CA2084105C - Subscriber initiated non-intrusive network-based analysis of facsimile transmissions - Google Patents

Subscriber initiated non-intrusive network-based analysis of facsimile transmissions

Info

Publication number
CA2084105C
CA2084105C CA002084105A CA2084105A CA2084105C CA 2084105 C CA2084105 C CA 2084105C CA 002084105 A CA002084105 A CA 002084105A CA 2084105 A CA2084105 A CA 2084105A CA 2084105 C CA2084105 C CA 2084105C
Authority
CA
Canada
Prior art keywords
facsimile
call
telephone
network
signals
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Expired - Fee Related
Application number
CA002084105A
Other languages
French (fr)
Other versions
CA2084105A1 (en
Inventor
Richard C. Fuller
Thomas W. Goeddel
Robert B. Heick
Martin Herzlinger
Subramanian Krishnamurthy
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
AT&T Corp
Original Assignee
American Telephone and Telegraph Co Inc
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by American Telephone and Telegraph Co Inc filed Critical American Telephone and Telegraph Co Inc
Publication of CA2084105A1 publication Critical patent/CA2084105A1/en
Application granted granted Critical
Publication of CA2084105C publication Critical patent/CA2084105C/en
Anticipated expiration legal-status Critical
Expired - Fee Related legal-status Critical Current

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N1/00Scanning, transmission or reproduction of documents or the like, e.g. facsimile transmission; Details thereof
    • H04N1/00002Diagnosis, testing or measuring; Detecting, analysing or monitoring not otherwise provided for
    • H04N1/00007Diagnosis, testing or measuring; Detecting, analysing or monitoring not otherwise provided for relating to particular apparatus or devices
    • H04N1/0001Transmission systems or arrangements
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04MTELEPHONIC COMMUNICATION
    • H04M3/00Automatic or semi-automatic exchanges
    • H04M3/22Arrangements for supervision, monitoring or testing
    • H04M3/26Arrangements for supervision, monitoring or testing with means for applying test signals or for measuring
    • H04M3/28Automatic routine testing ; Fault testing; Installation testing; Test methods, test equipment or test arrangements therefor
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N1/00Scanning, transmission or reproduction of documents or the like, e.g. facsimile transmission; Details thereof
    • H04N1/00002Diagnosis, testing or measuring; Detecting, analysing or monitoring not otherwise provided for
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N1/00Scanning, transmission or reproduction of documents or the like, e.g. facsimile transmission; Details thereof
    • H04N1/00002Diagnosis, testing or measuring; Detecting, analysing or monitoring not otherwise provided for
    • H04N1/00026Methods therefor
    • H04N1/00029Diagnosis, i.e. identifying a problem by comparison with a normal state
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N1/00Scanning, transmission or reproduction of documents or the like, e.g. facsimile transmission; Details thereof
    • H04N1/00002Diagnosis, testing or measuring; Detecting, analysing or monitoring not otherwise provided for
    • H04N1/00026Methods therefor
    • H04N1/00034Measuring, i.e. determining a quantity by comparison with a standard
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N1/00Scanning, transmission or reproduction of documents or the like, e.g. facsimile transmission; Details thereof
    • H04N1/00002Diagnosis, testing or measuring; Detecting, analysing or monitoring not otherwise provided for
    • H04N1/00026Methods therefor
    • H04N1/00039Analysis, i.e. separating and studying components of a greater whole
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N1/00Scanning, transmission or reproduction of documents or the like, e.g. facsimile transmission; Details thereof
    • H04N1/00002Diagnosis, testing or measuring; Detecting, analysing or monitoring not otherwise provided for
    • H04N1/00026Methods therefor
    • H04N1/0005Methods therefor in service, i.e. during normal operation
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N1/00Scanning, transmission or reproduction of documents or the like, e.g. facsimile transmission; Details thereof
    • H04N1/00002Diagnosis, testing or measuring; Detecting, analysing or monitoring not otherwise provided for
    • H04N1/00026Methods therefor
    • H04N1/00058Methods therefor using a separate apparatus
    • H04N1/00061Methods therefor using a separate apparatus using a remote apparatus
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N1/00Scanning, transmission or reproduction of documents or the like, e.g. facsimile transmission; Details thereof
    • H04N1/00002Diagnosis, testing or measuring; Detecting, analysing or monitoring not otherwise provided for
    • H04N1/00071Diagnosis, testing or measuring; Detecting, analysing or monitoring not otherwise provided for characterised by the action taken
    • H04N1/00074Indicating or reporting
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N1/00Scanning, transmission or reproduction of documents or the like, e.g. facsimile transmission; Details thereof
    • H04N1/32Circuits or arrangements for control or supervision between transmitter and receiver or between image input and image output device, e.g. between a still-image camera and its memory or between a still-image camera and a printer device
    • H04N1/333Mode signalling or mode changing; Handshaking therefor
    • H04N1/3333Mode signalling or mode changing; Handshaking therefor during transmission, input or output of the picture signal; within a single document or page
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N2201/00Indexing scheme relating to scanning, transmission or reproduction of documents or the like, and to details thereof
    • H04N2201/32Circuits or arrangements for control or supervision between transmitter and receiver or between image input and image output device, e.g. between a still-image camera and its memory or between a still-image camera and a printer device
    • H04N2201/333Mode signalling or mode changing; Handshaking therefor
    • H04N2201/33307Mode signalling or mode changing; Handshaking therefor of a particular mode
    • H04N2201/33342Mode signalling or mode changing; Handshaking therefor of a particular mode of transmission mode
    • H04N2201/3335Speed or rate

Abstract

Non-intrusive monitoring and analysis of real-time facsimile transmissions is accomplished. Analog impairment measurements are made on the high speed page signal in those transmissions and protocol analysis is made on the low speed control messages in those transmissions. These measurements and analysis are a powerful tool for trouble shooting service problems afflicting facsimile transmissions. Actual customer traffic can be monitored to detect circuit impairments and to evaluate service being provided. Customer complaints can be handled by providing them with a special phone number which may be called to provide the ability to have the facsimile transmission experiencing a problem specially directed through the facsimile analyzer and forwarded to the intended destination while it is being non-intrusively monitored by service personnel whothen can quickly provide the customer with a diagnosis of the problem.

Description

- '' 2~8~

SUBSCRIBER INITIATED NON-INTRUSIVE NETWORK-BASED
~NALYSIS OF FACSIMILE TRANSMISSIONS
Technical Field This invention relates to facsimile commllni~a~ions~ More specifically, 5 this invention relates to subscriber initiated analysis of facsimile transmissions in a public switched telephone network.
Back~round of the Invention Commllnic Itit)n by facsimile is becoming increasingly important in many areas, patticularly, in business commllni~tions~ because doc~ )ell~s may be10 sent by farcimilP. from point to point virtually in~t:3nt~neously. The delaysexperienced in using other modes of sending doculllc~ ., such as the postal service, are avoided in tr~nemitting documents by f~rsimile The speed with which documents can be sent from one place to another has resulted in greatly increased use of f~rsimile, which makes it increasingly i~ that an uninterrupted and 15 reliable level of service be provided on public switched telephone networks so that successful f~c~imil~ transmission of an increasing number of documents may be accomplished without signifir~nt degrees of in~ t It, tlle.brol~, has becomeincreasingly illlpo~ t that the f:~r,simile traffic through a public switched telephone network be accura~ely characterized. It also has become increasingly illlpUI lant that 20 any illlpailllRnts of farsimile tr~n~mi~inns be rapidly identified and the source of those impairments be accurately ~etrrmin~(l so that correcdve acdon may be taken.
Until now, there has been no way to monitor and di~gnose actual fiqr~imik~ tr~nsmi~ n~. There is no available monitoring equiplllellt which obtains convenient switched access to des*ed portions of the traffic flowing through a public 25 switched telephone network to measure and ch~r~rteri7e f~c~imile tr~nsmi~sion~.
There is nothing available which analyzes the protocols present in f~c~imile tr:ln~mi~sions and ties that analysis to impairment measurements of the page data in those tr~n~micsions. There is no currently available capa'oility of accurately identifying network and ~,uslolllei premises i",p;~-""ents during f~rsimile 30 tr~nsmi~siûnc Prior techniques of measuIing illlpaillllellls affecting facsimile tr~nsmi~sions involve intrusive techniques which use special test signals which are monitored to ascertain any problems with the tr~n~mission~ path. This results in test conflitions which do not duplica~e or simulate the actual conditions experi~nced35 during real facsimile tr~ncmissions. Thus, the results obtained from these intrusive techniques may not accurately reflect the situation experienced in the course of .,., . . ., . - ~ - , . -, .... .

, 2as~l0~
making a facsimile call. Actual customer premises equipment is not involved in making intrusive tests. Therefore, problems in completing f~csimil~ tr~n~missions caused by the customer premises equipment will not be identified as such with these techniques. The sources of illlpailn~ent, therefore, may not be accurately identified 5 by intrusive techniques. In addition to the fact that different equipment is connected to the network, the source of those impairrnents may also not be found due to the artificial nature of intrusive testing. Moreover, the portion of a commnnirations network being tested must be taken out of service to accomplish the testing in accordance with prior techniques, and thus this part of the network will be 10 unavailable for norrnal use.
Prior devices which measure commllnications signals are not able to adequately ch~c~ ize a f~rgimile tr~ngmi~inn non-intrusively as the tr~ngmigsiortakes place. Moreover, if it were attempted to use this e~luiylllellt for characterizing facsimile tr~ngmi~cions the e~luiL Illent only would be able to obtain dedicated access 15 to a single facsimi1P. apparatus at a time. Thus, itl~.ntifir~ti-~n of all the possible problems in a large commllnic~tioni network, such as a public switched telephonenetwork, are impossible or impractical. There has been no e~luiplllel~t available which is capable of conveniently measuring in any meaningful fashion f~simil~
traffic in a public switched telephone network so that sources of i,~p~;l.n~nt to those 20 tr:~ngmigg;ons could be effectively identified and rapid and effective corrective action could be taken. There is no way for a subscriber of a public switched telephone network experiencing a problem in successfully completing a given f~rsimilP
tr~n~migcion to conveniently receive a ~ nQgic of why the tr~ngmisgion is not being successfully compl~tç~
Accordingly, there is a signifi- ~nt and long-standing uns~tigfiPcl need for equipment which can properly analyze facsi-m--ile tr~nsmigsinng in a public switched network in general and can conveniently diagnose problems experienced by individual network subscribers.
Summary of the Invention The need identified above is met by an apparatus which selectively and non-intrusively monitors real time facsimile tr~n~micsions as they are occurnng in a network. The apparatus can obtain convenient access ~o a relevant part of the commllnic~tion traffic flowing through a network node in response to a request from a network subscriber. The apparatus may measure certain characteristics of a 35 particular f~c~imile call and determines the reasons for any difficulties experienced by the netws)rk subscriber.

- ; ~.

2 ~
In one example of the invention, subscribers to a public switched telephone network may be given a customer service number which may be dialed when that subscriber is experiencing problems in cornpleting fslcsim;le, calls. Dialing that phone number permits the subscriber to send the attempted trangmi~sion to aS facsimile analyzer in accordance with this invention which will then forward the tr~ncmicsion to the inten~ed destination of the tr~ncmiccion i-1P,ntified by thesubscriber. The specific f~ccimile tr~ncmission may now be non-intrusively monitored by the f~csimil~ analyzer as it takes place to determine the identity and source of hllpai~ ellts experienced by that tr~ncmiccir~n. Once the nature of the 0 i...piqi. ~ ntC have been i~lentifiprl~ customer service personn~,l may notify the subscriber of the findings produced by the f~csimile, analyzer.

Brief Descr;ption of the Drawin~s FIG.l illustrates an example of a public switched telephone network in accordance with this invention co.llaining an apparatus for characterizing and 15 measuring i..)p~i" ,.~ntc of f~simil~ tr~ncmicsions carried by the network.
FIG.2is a detailed schematic diagram showing an example of a connection between a facsimile measurement apparatus and a node in a public switched telephone network like the one shown in FIG.l.
FIG.3 illustrates a full duplex ...c.,-ilu. ;ng function carried out by a 20 f~rcimile measurement apparatus in accordance with this invention.
FIGs.4 to 6 illustrate the steps carried out by a f.~csimil~. measurement apparatus in accordance with this invention to accomplish a full duplex directed test access function.
FIG.7 illustrates the main circuit elem~.ntg in a f~ccimile analysis apparatus in accordance with this invention.
FIG. 7a illustrates the main signal components flowing from a tr:lncmitt~r to a receiver during an example of a typical G3 f~simile call.
FIG.8is a more detailed sch~m~tic diagram of some of the circuit elements shown in FIG.7.
FIG.9is a more detailed schematic diagram of the circuitry used to implement the in-service quality measurement apparatus shown in FIG. 8.
FIG.lOis a more detailed schematic diagram of the circuitry in the demodu}ator shown in FIG.9.

" 2~8~10~
FIG. 11 is a more detailed schematic diagram of the circuitry in the constellation analyzer shown in FIG. g.
FIGs. 12a to 12c illustrate exAmrles of thresholds used by the impulse noise measurement circuit shown in FIG. 11.
S FIG. 13 is a flow chart specifying circuitry in the co-llputel shown in FIG.2 for accomplishing rli~gnosi~ Of f:lr~imile tr~n~mi~inns received by f~c~imil~-analysis equipment in accordance with this invention.
FIG.14is a state diagram illustrating an example of the operation of the state mac hine of FIG. 13 for a typical G3 type f~c~imile trAnimi~ion without an10 error correction mode.
FIG. 15 is a block diagram illustrating the database architecture in the computer of FIG.2.
FIG. 16 illustrates an example of a menu of options available to a user of a f~csimile measurement apparatus in accordance with this invention.
FIG.17is an example of a country list available to a user of a fAcsimile measurement apparatus in accordance this invention.
FIG. 1~ is an example of a trunk list available to a user of a f~similt~, measurement apparatus in accordance this invention~
FIG.19is an example of a session list available to a user of a f~ccimilP, 20 measurement apparatus in accordance this invention.
FIG.20is an example of a session ~.lull~ produced in accordance with one ex~mrle of this invention.
FIG. 21 is an exarnple of a call list produced in accordance with one example of this invention.
FIG.22is a call S~ ll~y produced in accordance with one example of this invention.
FIG. 23 is an events list produced in accordance with one example of this invention.
FIG.24is an example of analog i..lpAil . .~nt measurements displayed for one of the events in FIG.23.
FIG.25is an example of raw data taken for one of the events shown in FIG.23.

-' 2~8~

Detailed Description FIG. 1 is a schematic diagram showing an example of a public switched telephone network containing a farsimile analyzer for purposes of monitoring facsimile calls in the network. FIG. 1 also shows illustrative tr~ncmi~ting facsimile 5 e~uip~lle~lt, receiving facsimile equipment, and the lines, trunks, and nodes in the network used to complete a fflrsimilP call between the transmitting f~csimile equipment and the receiving farsimile equipment. In FIG. 1, a tran~imit~ing f~rcimile e4uiplllent 10, which may be a so-called Group 3 f~rsimile machine, isconnected to a public switch telephone network via a subscriber line 12, which may 10 be a pair of wires.
The subscriber line 12 is connected so as to form a subscriber's loop between the trflncmitting facsimile equipment 10 and an end o~fice 14 of the public switched network. In the U.S., the end of fice 14 may be a local exchange of fice of one of the Regional Bell Operating Comr~ni~s (RBOC's) and the like. The end 15 of fice 14 may be connected by a trlmk 15 to a toll switch 16 located in ane of the central of fices in a domestic or intern Ition~l inter-eYchflnge long distance network 17. The inter-exchange network 17 also contains other toll switches and trunks which are used to connect other calling parties not shown in FIG. 1 to other called parties also not shown in FIG. 1. The long distance network may be one of those 20 provided by a long distance calTier such as AT&T. Only the toll switches and trunks connecting the transmitting f~rsimile e4uip~ ,nt 10 with the receiving ~rsimile e(luiy~lent 11 are shown in FIG. 1. In addition to those items already described, the connection between the tr~nsmitting facsimile equipment 10 and the receiving facsimile equipment 11 further compri~es a trunk 18 connecting a toll switch 16 to a 25 toll switch 20, a trunk 22 connecting toll switch 20 to a toll switch 24, a trunk 26 connecting the toll switch 24 to another end office 28 which is connected to thereceiving fflcsimile equipment 11 via another subscriber line 30.
The public switched network of FI&. 1 has a system for identifying facsimile trflnimiision~i in selected portions of the total co~ r~tionci traffic on 30 the network and for pelÇo~ g certain measurements and analysis on the i~entifi~d facsimile tr~n~mi ~sions which are useful in identifying Ihe amount and kind of facsimile trfln~imi~i~iion~i flowing in the network and the extent to which successful f~c~imile service is being accomplished on the network. These measule.llen~s aremade non-intrusively in real time as the fflcsimile tran~micciinn~ are occurring.
35 These measurements are useful in rlj~gno~iing the causeis of difficulties in successfully completing facsimile tr~ncmi~ it)n~ through the network and in .. : . .,. - - . . - - ~ :

2 0 ~
char~rt~ri~ing in some meaningful way the amount and kind of fa~simile tr~n~mi~sions through the network.
In this regardj the network of FIG. 1 contains one or more facsimile measurement systems connected to components of the network through which the 5 measurement system can obtain non-intrusive access to some or all of the communications traffic in the network.
One of those f~c~imil~ measu~ ent systems 32 is shown in liIG. 1.
That measurement system 32 is connected to one of the toll switches 20 so that selected portions or all of the co,,l,,-,ll-ir itir,ns traffic through the toll switch 20 can 10 be observed in a non-intrusive manner. Facsimile calls can thereby be iclentifi~
certain char~t~ri~tirs of those ~ccimilt-. calls can be measured, and the nature and source of any i~ ail.llellts of the calls can be found and diagnosed. Although PIG. 1 explicitly shows only one measulG--Ien~ system 32 conn~ct~c1 to one switch in the public switched network, any number of mea~ul~l,lcllt systems may be used 15 anywhere else in the network where access to desired traffic through a selected part of the network may be easily gained. As is app~ from the description below, individual customer problems can be detected and addressed and aggregate fax service usage and quality in the network may be de~ermine~l FIG. 2 shows a more detailed example of an architect-lre which contains 20 a ~csimile measurement system 32 connected to a portion of a public switched telephone network so as to gain access to some or all of the co~ )ir,~tion~ traffic in the network. The alchile~;lu e of FM. 2 comrri~es a measurement system 32 foranalyzing f~r5imilP traffic on a selected portion of a public switched telephonenetwork. The measul~mellt system 32 also comrrises a ser~ice signal processor 3425 which analyzes one or more calls. Such an~lysis includes identifi~ ltion of the presence of f~csimil~ calls and the boundaries of those calls, p~ lfollllance of signal clSlssific~tion~ mcasulu~ nt of certain signal characteristics, and presçnt~tion of the results to a ~ nostic computer 35 and a user in~erf~e 36. The COIllpul-,f 35 receives data from the service signal ~loCeSSOl 34 on a suitable link 38, which may 30 be, for example, a X.25 9.6 kb/sec data link. The data received from processor 34 is stored in one or more data bases contained within the computer. The computer 35 contains software which pe.ro.ll.s diagnostic analysis of the data from processor 34 which may help to identify the potential sources of any abnormalities in the characteristics of f~csimile calls measured by processor 34. Any desired part of the 35 data collected by the processor 34 and stored in the data base in the co.llpulel 35 may be called up and displayed by a user through the user int~ e 36. The user interface 2 0 ~
36 operates in response to appropriate co"~"~A~ generated by a user via a peripheral device such as a mouse or keyboard connected to the compulel 35.
An access control unit 40 is linked to a network node 42 controlling the flow of communications traffic through the public switched telephone network of 5 FIG. 1. The network node may be any of a variety of equipment through which commllni~a~ions traffic flows. For example, the network node 42 may be a centralof fice switching system which controls the connection between a variety of inbound and outbound trunks of the public switched telephone network. By way of example,such a switching system may be an AT&T 4ESS or SESS~) switching system.
10 The network node 42 may also be similar switching systems from other m~nnf~rtllrers One other suitable example of a network node 42 in FlG. 2 is a digital cross-connect system such as a digital access cross-connect system (DACS) made by AT&T.
The access control unit 40 may be any circuit which can access a 15 selected portion of the commllnic~tioni traffic flowing through the network node 42.
For example, the access control unit 40 may be the digital test unit (DTU) in anAT~T remote m~iming system (RMS). The access control unit 40 is responsive to a comm~nfl from the colllpute~ 35 gen~ ed in l~;spol se to a request by a user of the f~csimil~ mea~u~ system 32 for infor~n~tinn about a specific portion of the 20 conl...,..~ ion~ traffic flowing through the network node 42. The access control unit 40 makes a suitable request, via an input/output l/O link 44, for the network node 42 to map one or more DS0 signals i'lowing through it, inc~ ling any sign~lling information corresponding to the phone calls represenîed by the selected DS0 signals, to a DSl link 46 connecting the access control unit d~0 with the network node 25 42. As those skilled in the art are aware, one phone call comprises a pair of DS0's, a transmit DS0 and a receive DS0, which may be mapped to adjacent DS0 time slots on the DS1 signal in the line 46. The techniques of mapping signals from a network node to a predetermined place in a blt stream like the one between the network node 42 and access control unit 40, by copying data relating to a specific DS0 stored from 30 time-to-time in specific loc~tion~ in the network node onto the bit strearn on line 46, are well-known to those skilled in the art and, therefore, are not described further.
Any such mapping technique may be used to carry out this invention. The selectedDS0 signals mapped onto the bit stream on line 46 are directed to one input of the service signal processor 34 via the operation a digital bridging repeater 48 3~ connecting the line 46 to the input of the service signal processor 34. The service signal processor 34 then i~1entifiPs call boundaries from signalling inforrnation . . . . . . . . . ' . .

:,;, , , ;, : ....... , .~,. . .. .. . .
-- , : ,.: :: : : : ;;: , - . :

208~1~~7 7,csOci~tçd with the DS0 signal, identifies the nature of any phone calls on theselected DS0, and makes certain measulG~ rlls and observa~ions regarding any facsimile tr~ncmi~sions associated with that DS0 signal. Signalling infolmation,which is used by the service signal processor 34 to identify call boundaIies, may be 5 commllnirated to the service signal processor 34 by setting the state of one or more predetermined bits in the stream of bits flowing between the network node 42 andthe access control unit 40. For example, two state robbed A-bit signals may be used.
Alternatively, the call signalling may be ~ ir~t~d to the service signal processor 34 via a cn~ iratic)n channel which is separate from the DS1 link 46.
The apparatus of FIG. 2 may be operated in two separate modes. The first mode is a full duplex monitoring mode which enables the service signal processor 34 to monitor both directions of one or more switched connections through a network node 42 such as a 4ESS~) type switch. The second mode is a directed test access mode in which there is the establi~hmP~t of a full duplex monitor connection 15 directed to a specific switched connection originated by a customer experiencing problems in completing a specific f~r~imile trpn~mi~si~n The full duplex monitor capability of the apl)~alus of FIG. 2 is illustrated in more detail in FIG. 3.. As shown in FIG. 3, the access control unit 40 causes the network node 42 to map a desired portion of the c~ ir~ti~n~ traffic 20 through the node onto the link 46 between the access control unit 40 and the network node 42. In one example of the invention, the access control unit 40 sends a command over the link 44 eo the network node 42 which carlses the network node 42 to map a specific DS0 signal flowing through the network node to a preselected slot on a DS l receive frame carried by the link 46. The selected DS0 signal is mappPd 25 from a trunk 50, as in~i~at~cl by arrow 52, to a slot design~ted primary TAT X
(st~nding for test access trunk X) in the DS1 receive frame on the link 46. A
col.es~ollding DS0 on a connPctecl trunk 54 is mapped, as inrlic~ed by line 56, to an adjacent slot in the DS 1 receive frame d~iign~ted associated TAT X+1 (standing for test access trunk X+1~ in FIG. 3. The DS0 on trunk 54 is the DS0 which is the return 30 a~soci:lt~cl with the selected DS0 on trunk 50 by the switch fabric of the network node 42. Monitoring both the DS0 associated with the send leg of trunk 50 and the col-.,slJonding DS0 associated with the receive leg of trunk 54 thus provides a full duplex monitor for a period of time specified by the user. A bridging repeater 48 provides the signals in the primary and associated TAT slots to the se~vice signal 35 processor 34 for measurement and analysis.

, ~ . , , , ; ~ . , . .: ~ .

2 ~

In one embodiment of the invention, the monitoring function is established by a suitable user cornrnand directed to the co~ u~t;r 35 which thencauses the access control unit 40 to provide the service signal processor 34 access to a desired portion of the traffic through the network node 42 for an in~lefinite. per~od S of time until the monitoring operation is termin~tecl by an abort comïnand entered by the user into the computer 35. In another embodiment, the computer 35 may be programmed to provide a monitoring operation for a predet~rmin~l arnount of timewith automatic discontinuance of the monitnring operation after the expiration of the predetermined amount of time.
Individual problems experienced by a network cu~7iol-ler in completing facsimile transmissions can be :lcsessed by facsimile analysis equipment i accordance with this invention by operating that e~lui~ll~nt in the directed test access mode m~ntion~d above. In this mode, a customer experiencing difficulty incompleting a facsimile trangmigsion first calls a special number, such as a special 15 800 number, which establishes a dialog with network personnel ~esponsible foroperating f~r~imile measulillg e4ui~n~ell~ in the telephone network. The customer notifies these personnel of the nature of the problem, the identity of the destination telephone number to which far~imilç tr~ncmi~sion is being attempted, and the identity of the telephone number from which the f~r~imilP. t~ncmi~sion is being 20 sent. The network personnel specify to the customer a special m~intpnanre telephone number (e.g., a "101" number). The cu~tom~r will use this telephone number in another attempt to complete his contP-mrlS-te~ fax tr~n~tion to its inten(led clestin~tion. This number will be used instead of the normal ~lest;n:~tion phone number used in the past which has just been given to m~inten~nce personnel.
25 The m~inten~nt~e number cc,ll~,sponds to a specific trunk ~l)e~ance in the switch which when called is then terminated by a predetermined test port module in the access control unit 40. After giving the customer the special m:-intens~nce number, the network personnel then program the access control unit 40 to answer and respond to an incoming facsimile call from the customer made to the m~int~n:~n~e number.30 The service signal processor 34 is activated to start a measurement process involving monitoring of appropriate DSO's in the DS 1 ch~nn.~ between the network node 42 and the access control unit 40. The monitored DSO's are those which have been clesign~ted to carry tr~n~mi~cion and signaling inform~tion flowing between the network node 42 and the access control unit 40 and which are associated with the35 f~csimile call from the customer. As in the case of the monitoring operation, two state robbed A-bit signalling may be used to convey ~ign~lling information from the ,,, ". - . . ~ ., , , ., ~ -~, ,, . , , ~ - ~,. .

2 ~

network node 42 to the access control unit 40 and the service signal processor 34.
As shown in FIG. 4, when the user then places his f~-~simile call to the special maintenance telephone number, a detection circuit 58 in the access control circuit 40 detects the presence of an incoming m~int~n~nre call which is directed by 5 the network node 42 to the access control unit 40. The presence of the m~intPn:lnce call may be detected by an electronic device which recognizes a special ringing signal sent by the network node 42 to the access control unit 40 when the customer places the maintenance call. For example, the special ringing signal may be a repeati~g cycle of off-hook and on-hook signals, each such signal of predetprminpd 10 duration, conveyed to the access control unit 40 in the signalling channel between the node 42 and the access control unit 40.
As shown in FIG. 5, the access control unit 40 contains a call termin~tinn device 60 which will answer the call and send an off-hook clesign:~tion back to the customer at the origin~tinn point. As also shown in E;IG. 5, a call 15 origination device 62 in the access control unit 40 places a call to the int~nded recipient of the fa~ simil~. call using conventional multifrequency outpulsing techniques in accordance with the recipient's telephone number received from thesending customer. This call fc.~ g function can be accomplished by an out-tandem call establichm~nt procedure in which the access control unit 40 emulates the 20 sign~lling funcdonality of a trunk circuit coming into the network node during the call set up procedure. The call forwarding function can also be accomplished by the network node itself est~bli~hing the call by outpulsing upon receiving an appropriate comm~nfl to do so on the input/output line 44 in FIG. S.
Once the digits of the telephone number given by the customer have 25 been outpulsed and the call between the access control unit 40 and the facsimile recipient has been established, the access control unit 40 reconfigures the tr~nsmi~iiQn and ~ign~lling chS~nn~l~ so that continuity is established between the origination and destination f~simile machines. Specifically, the receive signals(tr~n~mi~sion and sign~llin~) from the origin~tion point are connected to the transrnit 30 signals directed toward the ~lest;n:ltion point and vice versa. This cross-connect capability or unsplit connection allows the service signal processor 34 to measure signals from both directions via the bridging repeater 48 as shown in FIG. 6. After the call has been established, the service signal processor 34 makes measurements of the f~csimile call between the origination point and the destin~tion point and collects 35 data about the call.

, ~ .

2v8~ 3~

A disconnect monitoring device 64 in the access control unit 40 monitors the connect;on between the origination facsimile machine and the destination fa~simil-o machine for a signal inf~ir~ting a disconnect. When a disconnect signal is detected, the access control unit 40 removes the unsplit S configuration and transmits an on-hook disconnect signal in both directions. When no further test calls from this customer are to be analyzed, the access control unit 40 should be disabled from responding to incoming call attempts by un~nthori7ed sources. This constitutes a security mechanism which prevents the execution of lln~u~h~rized call forwarding procedures in the access control unit 40.
FIG. 7 is a representS~-;c n of the main functional circuit el~mP.nt~ in the service signal processor 34, the diagnostic co~ ule~ 35, and the user int~ ce 36.
As shown in FIG. 7, the service signal processor 34 co.,-l)l ;ses a number of circuits which perform certain measurements on f~similp tr~nim;~ir~n~ passing through thenetwork node 42. Those circuits include a signal cl~csifir~tion device 66 which is 15 responsive to selected signals flowing through the network node 42 to de~ermine what kind of signals they are. Specific~lly, the signal el~esifir:lti~n device may determine the bit rates of those signals and what type of modem was used in producing them. A circuit 68 is responsive to selected parts of certain facsimile tr~nsmic~ion~ and makes a set of analog i l~p~il " ,~ ~t measurements relating to those 20 signals nonin~rusively and in real time. Further details of the signal cl~cifica~ion circuit 66 and tr~n~mi~sion mez~ulc~ nt circuit 68 are described below.
Another circuit element 70 in ~;IG. 7 detects the m~gnitude and delay relating to the appcd~ ce of certain echo signals associated with the farsimi1P
tr:~n~mi~sit ns being monitored by the service signal processor 34. The echo and25 delay measurements are made in both directions of a f:lrsimilP. call. In the example of a G3 f~r,simile call, the prede~e~minecl characteristics of V.21 signals sent during the tr~nimis~ion of leading HDLC flag characters and the V.29 "random CD"
training sequences are employed as known signals during the f~rsimil~ transaction.
The circuit element 70 searches for those signals continuously on both directions of a 30 farsimile call. The time delay between ~ltern~ing appearances of the sarne signal in two different directions is used as a measure of the delay experienced in a particular direction~ In ~ ion~ the difference in arnplitudes of the primary and echo V.21 signals may be used to compute an echo return loss. In order to obtain more exact measures of the echo path characteristics, echo canceling techniques may be used for 35 detecting echoes of the V.29 signal during page tr~ncmi~sion. The relatively large bandwidth characteristics of the V.29 signal are used advantageously to extract more : :

. 2 ~

inforrnation about the time and frequency response of the echo channel looking in the same direction as the page tr~ncmission Echo return loss and delay may be measured by colllp~ing the signal level and arrival times of demodulated facsirnile protocol messages on the primary 5 and echoed sides of the tr~nemiesion. The comparison of signal levels and arrival times may be made by directly observing the arrival times and signal levels of demodulated primary and echo signals in the fa~simil~ transmicsion or by using signal correlation techniclues. For example, if a demrxl~ ted V.21 HDLC ~csimileprotocol message on one side of a call is followed a short time later by the same 10 demo(lnl~ted V.21 HDLC facsimile protocol message on the opposite side, the echo return loss is given by the difference in signal level between the primary and echoed replications, while the difference in arrival time of the two m~.ss~gec gives echo delay. The half-duplex nature of the facsimile protocol and the robustness of the modem used for sending the protocol mP.ss~ges in the presence of channel 5 i~ te allow this technique to work.
The service signal processor 34 also includes a circuit element 72 which emo ~ tes and interprets certain protocol m~ss~es which are part of the rnonitored f~similP tr~nemiseioni~ For exarnple, the circuit element 72 may d~mocl~ te and interpret the T.30 protocol messslges associated with a G3 fal~simile 20 tr~nemiceion As illustrated in FIG. 7a, a G3 faccimile call comprises several components. FIG. 7a illustrates an example of a single page G3 f~eimilP.
transaction displayed in various levels of detail for the tr~nemieeionc made by the sending f~simil~ equipment to the receiving facsimile e4-.ip...~ ~t. Transmieeions 25 made from the receiving f~csimil~ equi~ ent to the tr~ncmining facsimile e~luip~le.lt are not shown in FIG. 7a, but they are apparent to those skilled in the art who are aware of the nature of G3 faceimile tr~nemiesion~e and the T.30 protocol.
Tr~nemissirne from the receiver to the sender may include signals notifying the tranemitter of the characteristics and capabilities of the receiver, confirmation of 30 receipt of training and page signals, notifil~tic-n of the tr:lnemitt~r that there has been an llnellccessful tr~nimiesion by the tr:~nemitter, and requests of the tr~nemitter for changes in characteristics of the tr~n~mi~sion such as a request for reduced bit rates and the like.
At its highest level, the entire call shown in FIG. 7a is considered as 35 simply a one page transaction at 9600 bits per second. At an interrnediate level, the various T.30 constituents are shown comprising tonal, V.21, and V.29 messages. At the lowest level, detailed line signals are shown in FIG. 7a~ When making non-intrusive impairment measurements, the protocol used during the G3 transaction must be tracked so that other functions such as echo measurements on the '~/.21 signals and impairrnent measurements on the V.29 signals are switched in at the 5 proper time. Besides controlling measurement functions, the T.30 tracking alsointerprees messages when f~csimil~ transactions have problems. The mPcc~ge~ are saved for later diagnosis by the computer 35. Real time tracking and control will be accomplished for all aspects of the T.30 protocol described below. Other inforrnation may be collected in accordance with this invention for other kinds of 10 f~ccimile calls and other kinds of protocols.
In the situ~tion of a G3 farsimil~. call using a T.30 protocol, the protocol signals in the tr~n~mi~cion may be measured to see if they are a standard version of the T.30 protocol or some nonc~n(l~rd protocol. Any inform~sion in the protocol signals identifying the m~nl~f~ctllrers of the sending and receiving f~csimile 15 equipment will be noted. Tr~ncmiscic-n rate infonnadon will also be sensed.
Specifically, the protocol tracking circuit will ascertain the initial bit rate at which information is being sent from the sending e4ui~ en~ to the receiving e~lui~ment and will note any situations where the e~luiylllent falls back to some lower bit rate. The protocol tracking circuit keeps a certain amount of st~ti~iti~l information about each 20 f~csimile call. The statistical ;nformadon may include the number of pages in the Iri~ncmiscion, the rates at which each page is sent, and the tdme it takes to complete the tr~ncmicsic-n of each page. The protocol tracking circuit may also note any turn around citll~tiQnc whereby the answering f~cimil~i machine sends pages to the calling f~simile ~n~chine. If a T.30 error correction mode is being ernployed, 25 whereby the receiving f~rsimil~ e4jui~,mcnt detects that some or all of a tran~mi~ted page has not been properly sent and received and then requests a retri~n~imiscion of the inadequately delivered portion of the page tr~n~imission, the protocol tracking circuit may detect the use of this error correction mode and may keep certain retr~ncmiccions statistics relevant to its use. The protocol tracking circuit may also 30 be configured to detect the presence of tones such as echo protection tones, the auto-originate 1100 Hz calling tone, and the auto-answer 210Q Hz answer tone.
In the circuit of FIG. 7, the outputs of the measurement circuit elements 66, 68, 70, and 72 are directed to a circuit 74 which integrates and arbitrates all of the measurement data produced by those measurement circuit elemen~s. The 35 integration and arbitration circuit element 74 sends the measu~ t data it receives in a predetermined order to the input of a high level diagnostic circuit 76 which flags , . .. ,, ~ , : ~ - .

20~3a~

certain Rnom~lies in the measured facsimile tr~n~mi~sions, such as protocol anomalies. The circuit 76 also analyzes the measurement data it receives from the integration and arbitration circuit 74 and produces a diagnosis of problems which occur in the monitored fi~csimilç tri~n~mi~siçns. The results of the signal 5 measurements are commnnic~tP,d to users through a variety of reports prvduced by the user interface 36 described in more detail below.
FIG. 8 is a more detailed description of the service signal processor shown in FIG. 7. The circuit of FIG. 8 is responsive to signalling information from the network node 42 appearing on an input line 78. A call boundary i~l~pntifir~tic)n 10 circuit 80 receives the signalling inforrnation and produces an output signal on line 82 which identifies when a telephone call carried by a monitored DS0 begins and when it ends.
An ~ ition~l function pelro~ ed by the circuit of FIG. 8 includes signal speed cl~ssifi~tion involving the real time il1~Pr~tifir~tinn of the characteristics 15 of unknown voice band signals. In this exarnple of the invention, the voice band signals comprise 64 kilobit pulse code mçrlnli~tion (PCM) le~ sent~tioni of the intelligence being tri~n~imittPd between two users of the public switched telephone network on a DS0 slot selected for monitoring by the f~c~imilP. measurement apparatus of this invention. A coarse voice band signal çl~iifir~ti~n circuit 8420 receives the PCM signals and produces an OUtpUt on line 86 iden~ifying the ~l~sence of elPct~c~l energy in the selected DS0. In this manner the idle portions of a telephone call can be separated from active signal spurts which are further processed.
Once an active signal spurt is identified by the coarse cl~cisi Rr~tion circuit 84, the signal into the circuit 84 is next cll~csifiPd as either a voice signal or a nonvoice 25 signal. If the signal is identified as a nonvoice signal, it is ~ snm~d to be voice band data and the circuit 84 performs a coarse speed cl~sification which iclen~ifi~si the signal as being in one of a plurality of speed categories, such as very low speed, low speed, medium speed, and high speed. If the nonvoice signal is not ln one of thespeed categories, it may be labeled "unknown" and the ~ umption of voice band 30 data may be considered to be wrong. Ex~mpl~s of very low speed voice band data signals include V.21 and V.23 signals. Examples of low speed voice band data signals include V.22 bis and V.26 signals. Examples of medium speed voice band data signals include V.27ter signals. Examples of high speed voice band data signals include V.32 and V.29 signals.

.
, The specific techniques ~or classifying the signal input to the classification circuit 84 are known to those skilled in the art and are described no further here. An example of a suitable voice band signal speed cl lcs;fication circuit 84 may be found in Benvenuto U.S. Patent 4,815~136, the content of which is hereby S incc,~y~lldted by reference in its entirety. A signal relating to the outcome of the cl~cifir~tion process perforrned by the circuit 84 is delivered to an output line 88 to one input of a modem identifit ~tion circuit 90 which identifies on line 98 the nature of the modem in the tr~nsmitting equipment in light of the speed cl~csific~tion made by the c1~ccification circuit 84.
If it is detP.rmin~.d that the PCM signal directed to the input of the coarse cl~ccifir,~tion ci-rcuit 84 is one of the high speed types of voice band data signals, an in~ tinn of such is directed by the coarse cl~c~ific~tirln circuit 84 on a line 92 to one input of a high speed cl~ccifir~ti~n circuit 94. The high speed cl~ccific7~irn circuit 94 also receives on another input the PCM signal which is being cls~csifi~-l lS The high speed cl~ccifir~tion ci-rcuit is activated by the signal on line 92 tO make a more specific determin~tion of what kind of high speed signial the input PCM signal is. The high speed clS~ccifir~tion circuit 94 directs a signal represer~ g which of the high speed signals describes the speed characteristics of the PCM signal on an output line 96. This signal is sent to one input of the modem identific~tion circuit 90. The 20 modem iclentific~tiQn ci-rcuit 90 is l~y~llsiv~; to the speed signal on line 96 to produce an output on line 98 which is related to the character of the modem in the tr~ncmitting f~rsimil~ e4uil",~ent The nature of the high speed s~lassifir~tinn process is known to those skilled in the art and is not described further here. An example of a high speed cl~ssifir~tion circuit may be found in Benvel~ o et al. U.S. Patent25 4,979,211, the entire contents of which is hereby incorporated by reference into this application.
A tone detection circuit 100 is also responsive to the input PCM signal to identify the presence of certain tones norrnally found in f~csim:lP tr~ncmicsion The tone detection circuit 100 itlentifips~ for example, the 1100 Hz. and 2100 Hz.
30 tones which may be present at the beginning of a f~simil~ call. The tone detection circuit in-lica~es the presence of such tones by providing a suitable signal on output line 102. The tone detection circuit 100 may be any circuit which can indicate the presence of specific sinusoidal signals present in f~ccimilf tr~ncmiscil~nc For exarnple, the tone detection circuit 100 may include adaptive notch filtering circui~y 35 which changes its output in response to the presence of the tone which it is desLred tO
detect. The tones detected by the detection circuit 100 also include those tones 2 ~

produced as a result of modem training sequences tr;~n~mitte~l from the tr~ncmit~ing facsimile equipment to the receiving facsimile equipment. These tones are the result of predeterrnined sequences of symbols sent to the receiving e-luipnlellt to train the modem in the receiving equipment so that the intelligence co,~ -ir~ted between S the tr:lncmitting equipment and the receiving equipment is properly demodulated by the receiving e~lui~ ent. The training sequences are such that the energy transmitted from one facsimile machine to another falls primanly at a nurnber of distinct frequencies. Identifir~tic n of these frequencies in the PCM signal is thus an in~lirsltirln of the presence of the tr~ncmieiiorl of training sequences between the two 10 farsimile m~rhines, The specific frequencies which are found by the tone detection circuit 100 in connection with detection of training tones and sequences is useful in identifying the characteristics of the modem tr~ncmitting the training sequences and thus is an intlir~tion of the nature of that modem. A signal ~ ,senting the nature of the training tones is directed on an output line 104 of the tone detection circuit 100 to 15 one input of the modem i~l~,ntifi~tiQn circuit 90. Tn~licatinni of certain training tones from the tone detection circuit 100, as well as the results of speed el~ccific;ltion~ are used by the modem i(l~ntifir~tion circuit 90 to identify the type of modem used in the tr~ncmitt~r. As inrlic~ted in FIG. 8, the signal on the output line 98 from the modem iden~ifi~tion circuit 90 represents the kind of modem which is tr~n~mitting 20 and producing the PCM signal input to the circuit of FIG. 8. For ex~mple, the output of the modem identifir~tion circuit 90 may indicate that the identifi~(1 modem is a V.29 or V.27 modem producing page signals in a G3 fP~iimilP. call.
The circuit of FIG. 8 also includes a tlPm~~ tion circuit 106 which ~lemo~ trs the protocol portions of a f~rsimile call. Those protocol portions 25 contain information and messages used by the sending and receiving f~ cimil~
m?~rhin~s to make sure that they are properly set up to send and receive f~r$imile signals and to make sure that the facsimile tr~ncmicsion is being successfully accompliched The demt~ r 106 produces an output signal on line 108 relating to the tlem~~ t~d protocol messages in the PCM signal. In addition to the 30 demodulation of primary protocol messages, the demn-lnlatnr 106 also rlemr~ tes echoes of the primary protocol signals. In one ex~mple, the demc-llll~tor 106 demod~ es the V.21 protocol messages in a G3 f~rcimile call as well as echoes ofsuch V.21 protocol mess~es. Logic circuitry s~soci~ted with the dem~cllllotion may be used to interpret the message content of the demodulated protocol signals from 35 the protocol demodulator 106.

, ~ , .

A protocol tracking circuit 110 receives the demo~ul:~ted messages on line 108 and produces an output signal relating to the nature of page signals and trial tr~n~mi~sions (TCF's) in the façsimile tr~n~mis~inn. The page signals are the part of the facsimile tr~n~micsion which carries the inf ~rm~tion which is being sent from the 5 tran~mitting facsimile equipment to the receiving fS~simil~ equip-nent. In oneembodiment, the protocol tracking circuit 110 detects the speed at which the tr~n~mitting f~imil~ e(lui~meilt will be sending page data, hlro~ a~ion which isconr~in~d in the protocol signals sent between fal~imil~. machines. This page speed information is directed to an input of an in-service quality measurement control10 circuit 112 which activates an in-service quality measurement (ISQM) circuit 114.
The measurement circuit 114 makes certain analog imp~irment measurements on the page portion of the PCM fac~imilP tr:~n~mi~ion These in-service quality measurements are made in light of the nature of the page signals, namely, the nature of the modem which transmits the page signals from the sender to the receiver. The 15 nature of the modem along with the in-service quality mcasulbllle.~ are directed to an output line 116 of the in-service quality measurement circuit 114.
One variation of the circuit of FIG. 8 includes clet~çtin~ the nature of the tr~n~mittin~ modem from the output of the modem iflPntifirati~n circuit 90 instead of detecting the nature of the tr:~n~mitting modem from the information in the 20 protocol signals. In this situation, the output of the modem identifir~tion circuit 90 is directed on an input line 118 into the in-service quality measurement controlcircuit 112. This alternative may be employed in situ~tion~ where infolm~tion from the protocol tracking circuie 110 about the nature of the tran~mitting mç~em is unavailable or unreliable for some reason. The output of the protocol tracking from 25 circuit 110 is normally available before the start of a page or TCF tr~nerni~sion, while cl~ssific~tion completed by the modem irl~ntifir~ion circuit 90 normally is complete only after the page or TCF signal has begun.
The outputs of the coarse cl~ssific~tirrn circuit 84, the modem identifir~tion circuit 90, the tone defection circuit 100, the protocol demr~ tçJr 106, 30 the in-service quality measurement circuit 114, and the call boundary i~Pntifiçation circuit 80 are directed to inputs of a circuit 74 which integrates the mea~ .ellls and arbitrates them. The integrate and arbitrate circuit 74 produces an output signal which represents and identifies the nature of certain predetermined events occurring in the PCM signal representing the f~rsimilp~ tr:ln~mi~ion from the sender to the 35 receiver. The integrate and arbitrate circuit 74 produces a signal which accurately reflects the timing and sequence of the events which are of interest in the facsimilP

,., , , ~

2 ~ 0 ~

tr~nsmiision. The circuit 74 takes into account the differing amounts of time it takes to identify certain aspects of the facsimile tr~n~mis~ion. For example, it may take several milligeconds for activity to be detected in the PCM channel being monitored, 32 milli~econds for the voice/nonvoice c~ ifir~tion~ 128 milli.ceconcl.~ tor coarse 5 speed idensifir~tion~ and several seconds for the modem to be fully identified by the modem identification circuit 90. The integrate and arbitrate circuit 74 looks for the presence of energy in the monitored DS0 which is closely spaced in time with respect to certain other events to create an inrli~tinn that the event really occurred at the instant energy was detected in the DS0. For example, if signal activity or energy 10 is detected within a couple of seconds prior to modem identifir~tion, then the circuit 74 assumes that the modem which has been ic~entifi~d was actually turned on at the time of activity detection by the coarse cl~sifirsltir~n circuit 84 and not at the time the modem signal was formally i~lentifi~d by the modem i(lentifir~tion circuit 90.
The signal on the output 121 of the circuit 74 cr,mrri~es an event stream which 15 accurately reflects the events which have occurred in the tr~n~mic~ir,n and an accurate reflection of the sequence and timing of those events.
FIGs. 9-11 illustrate further details of the circuitry associated with making the in-service quality measur,mell~s referred to above. Once signal cl~csific~tion has been made by the circuit of FIG. 8 and the measured signal has 20 been specifir~lly ill~ntifi~l as a signal which is of interest as a facsimile call, non-intrusive illlpa~~ ent measurements are undertaken by the in-service quality measurement circuit of FIG. 8. For example, a V.29 or V.27ter modem signal used to transmit page data or TCF signals in a G3 f~rsimilp call is measured to determine the quality of the channel. These are ideal signals to measu~ in determining the25 quality of the channel because of their wide bandwidths, particularly, the V.29 and higher speed signals, and because actual cllctom~r signals are measured.
The in-service quality measurement circuit 114 compri~es a modem receiver which is switched on at a proper time to appropriately demodulate V.29 or V.27ter page signals to derive their data con~tell~tioni Impairments may then be30 measured by analyzing various aspects of the clemor~ tion process and by analyzing various aspects of the received data constellations. In this example of the invention, specific impairments to be measured for all V.29 and V.27ter rates are as follows:
1. Signal Level;

. - . .
": ~. ~

: . . .. . ,. . , ~

. . . ~ .
,. , . :: - :

2 ~

2. Frequency Offset;
3. Timing Offset;
4. Attenuation Distortion;
5. Envelope Delay Distortion;
6. Signal to Noise Ratio;
7. Signal to Noise Ratio after Removal of correlated mo~ll]l~tions;
8. Phase Jitter, 9. Amplitude Jitter, 10. Gain Hits;
11. Phase Hits; and 12. Impulse Noise.
As shown in FIG. ~, the PCM signal, which may be an A- or ~,1 - law PCM signal, is directed to the input of a Cartesian clem~~ tor 122 which removesthe carrier frequency from the PCM signal. The ~lemodlll~t~d PCM signal is directed 15 to an input of a con~tell~tion analyzer 124 which ascertains differences between the actu~l constellation received from the dPm~ tc~r 122 and the ideal congt~ tion exrect~d ~rom the known characteristics of the modem irl~ntifi~.d by the modem ~lentifir~tion circuit 90 or from the speed hlfo~ inn in the facsimile protocol ~emn~ul~tefl by the protocol demndl-l~tnr 106. The cr~ngtell:~tinn analyzer 124 may 20 contain a slicer circuit for determining the differences between the actual cnn~t~ tinn and the ideal cnngtell~t on Certain sample measu.~ ent~ are derived from vatious aspects of the processcs undertaken by the demodul~tor 122. Those sample measurements are directed on a output line 126 to an input of a microprocessor 128. Celtain other25 sample mea~ul~ ents are derived from the error between the actual modem constellation and the ideal modem co~gte~ ion detived by the constellation analyzer 124. The measurements derived from the constellation analyzer 124 are sent to another input of the microprocessor 128 on an output line 130.
In more specific terms, certain measurements may be obtained by 30 examining the state of certain adaptive processes in the demodulator 122. Theremaining measurclllents may be made by analyzing the constellation or "eye pattern" which is the output of the demodulator 122. At regular intervals, sarnple measurements are passed to the microprocessor 128 where they are processed to produce final measurements.

~ - ,. ~- .. .. -......... , The microprocessor 128 receives the sample measurements derived from the demodulator 122 and the constellation analyzer 124 over the duration of a modem tr~ncmiecion~ The microprocessor 128 reports a set of final measurements at the end of the modem tr~n~migcion~ For meas~.,.llenl, of transient events in the5 modem tr ,n~micgions~ counters either in the ~hmo in1~tor 122 and constellation analyzer 124 or in the microprocessor 128 record the number of such transient events which were observed in the modem tr:~ncmiscir)n~ The final measultlllellL, reported by the microprocessor are the values contained in the transient event counters at the conclusion of the-modem tr~ngmicsion The values of the transient event counters 10 maybe noted by the microprocessor 128 at periodic intervals during the tr~nsmicsinn~
for example, at one second intervals.
In this exarnple of the invention, the final mea~,ult,lle~ , of nontransient i~"l,~i""~ntc are reported dirre~c;ntly depending upon the length of the modem tr~ncmission For ex:lmpl~-, the nontransient ilnpaillllents are reported differently 15 depending on whether the tr~ncmicsion duration is less than four seconds or greater than four seconds.
Facsimile tr~ncmicsionc include modem signals which contain page inform~tion Such page tr~n~micsionc in typical G3 fax calls are usually well over four seconds in duration. Facsirnile tr~7ncmicsionc also usually include a short20 duration trial tr~ncmission known as a training check. These training check tr~ncmiCcil-ns contain predetermined sequences of bits and usually last about 1.5 sec~n~ic If the training check sequences are not received properly, the f~cimilem~chinec in certain inctslnres may fall back to tr~ncmiccion and reception at lower data rates than originally anticipated. In certain cilc~ s~nt~es~ the facsimile 25 tr:lncmiccion may be t.ormin~-ed if the training check sequences are not tr~ncmitted and received properly. In addition to the training checks, there also may be short duration partial page retr~nsmiCsions when the f~csimi1~ equipment is operable in certain error correction modes. If a f~simile tr~ncmigsion analysis apparatus inaccordance with this invention is to analyze why short duration tr~ncm;cgions such 30 as training check sequences and partial page retr~ncmisciong were not ~lop~illy received, as many analog imp~irnlen~c as possible must be measured during a short period of time. Trial tr~ncmission analysis is ess~.ntiS~1 when the connection is sufficiently bad that no actual pages are sent.
If the modem tr~ncmiss;on lasts less than about 4 secon~lc, a set of 35 sample measurements is reported to the microprocessor at the end of the short tr~ngmigsion. These sample measurements become the final measurements.

2 ~

Because of thç short duration of the modem tr~n~micsinn, some measurements may not yet be available or accurate. The available measurements in this example of the invention include the overall signal-to-noise ratio, signal level, frequency offset, amplitude jitter, phase jitter, attenuation distortion, envelope delay distortion, and 5 the quantity of transient events observed during the tr~ncmi~sion.
One obstacle to measuring the nontransient hllpa~lments during short tr:lncmiccic nc is that certain adaptive processes in the (lemoc~ tor 122 have not yet had a sufficient amount of time to stabilize. Since they have not fully adapted, the results of mea~u~ e.l~s based upon these adaptive processes may still be corrupted 10 at the end of the tr~ncmic~ion The preferred place to measure h~ e~ls is somewhere in the middle of the tr~ncm;icion when the ~lPmc~ tc~r is pe.rollll~ng at its best. To ~-- cc mrlich mea~ulemenl at the preferred place in the middle of the tr~ncmiccinn, I-leasul~lllents are kept for each block of a ~lcde~ i n~cl number of tr~nsmitted symbols. For e~mpl~, mea~.~rc.llen~ may be kept for each block of 15 sequentially occurring groups of 200 symbols each. If the duration of the tr~ncmicsiQn is less than 4 seconds, the Illeas~ ent block having the smallest overall noise power is reported to the microprocessor when the short duration tr~nsmiccion ends. The overall noise power provides a measure of good ~lemodl~ t~ r pe.r~ re For a given channel, the modem is pelrulllli"g its best 20 when the overall noise power is the lowest.
If the ~l~.n~.,licsion lasts more than 4 seconds in this example of the invention, the final mcas~llclll~nts are computed as the average of the sample measurements reported to the microprocessor at regular intexvals. The first set of sample measurements are reported to the microprocessor at the end of the 4th 25 second. Thereafter, sample measurements continue to be reported pe~io~ y to the microprocessor, for example, at S second intervals~
Most of the measurements are available by the fourth second, but certain measurements are not reported until later. At the fourth second, the signal level, overall signal-to-noise ratio, frequency offset, ~ttPnu~tir~n distortion, and envelope 30 delay distortion are reported. Timing offset may not be reported until later, for example, until the 9th second, because of a relatively slowly adapting feedback loop in a timing recovery algorithm used to supply the mea~ mellt. Values reported earlier may not be reliable. ~mpl;tllde mod~ tion, phase m~l~ tion, and uncorrelated signal-to-noise ratio may be reported for the first time at the 14th 35 second. These modulation measurements may require this additional amount of time because of the need for the processing elPm~nt~ to achieve steady state before those " :' ~ .: .' ' .. ::.. .: .. :~:.'': .. ' ''.. :':::: .',::'': .:: :,-, :: .: '' . , , 2~

measurements can be considered reliable.
Adaptive processes in the demodulator 122 provide measurements of signal level, timing offset, ampli~ude distortion, envelope delay distortion, and frequency offset. The demnc~ or used to obtain these measurements, for example, S may be similar to the demodulators found in standard modems in many respects.
For example, the demodulator 122 may be similar to those fl~mn~ tr~rs found in standard V.29 and V.27 modems. However, steps which convert modem symbols to data bits are not carried out by the clemtx~ tor 122 because the actual bits in the page tr~n~mi.c~ions are not needed to obtain the desired analog i."pai. ",~nt 10 meaa.l~nlellls. In an alternative embodiment, degradation can be measured by actually demo~ ting the page tr~n~mi~sir~n and observing page quality by lookingat certain aspects of the demodulated tr~ncmicsion such as scan line errors and the like.
FIG. 10 shows the main functional modules of the dem~ tor 122. An 15 ~ om~til~ gain control adjusts signal power to a desired level. The input PCMstream is converted from an 8 kHz to a 9.6 kHz sampling rate in this example of a demncluls~tnr. Timing recovery comre~ tes for differences between the transmit and receive modem. A Hilbert transform is used to recover the in-phase and quadrature parts of the signal. An adaptive equalizer removes linear impairrnents 20 from the signal. The output of the adaptive equalizer is then multiplied by acomplex sinusoid at the carrier frequency. The carrier frequency is recovered by the phase locked loop.
More specifically, FIG. 10 shows that the PCM signal is directed to the input of an ~utom~tir, gain control circuit 132 which ~mrlifif~s or attenuates the PCM
25 signal to provide a desired output level on line 134. The PCM signal from theautomatic gain control circuit 132 is directed to the input of a sample rate converter 136 which changes the sample rate of the PCM signal so that it exactly is an integral multiple of the symbol rate or baud rate of the modem which is tr~nimitting the PCM signal received by the circuit of FIG. 10. A timing recovery circuit 138 is 30 responsive to the output of the sample rate COnVCllC;i 136 via an irlput line 140. The timing recovery circuit 138 is basically a control circuit which produces an output signal on line 142 which is used by the sample rate converter 136 to accurately synchronize the output of the sample rate converter 136 precisely to the symbol rate of the tr~n~mi~ting modem. The actual sarnpling rate at the output of the sample rate 35 converter 136 in this example of the invention is, therefore, 9.6 kHz+e, where e is such that the rate at the output of the circuit 136 is a precise integral multiple of the , ! . . ~ ' .: i .
' ' , ' ,. . .. ~' , ; ' . ~ , .

"~ ' . ' ' ' . ~ '', " . :' ' ' ~ ' ' .. . , : , :. . ! ;
,, : , . . : . , ' ' . , , :

rate of the transmitting modem. A ~ilbert transform circuit 144 is responsive to the synchronized signal produced by the sample rate converter 136 to produce the in-phase and quadrature components of the quadrature amplitude modulated (QAM) page signals on lines 146 and 148. The in-phase and quadrature components are - 5 directed to two inputs of an adaptive equalizer 150 The adaptive equalizer 150 comren~slt~.s for the amplitude and phase distortion added by the network to thePCM signals The outputs of the adaptive equalizer 150, flerim~tecl to the symbol rate of the known modem, are directed to two inputs of a multiplier element 156 which10 removes the in-phase and quadrature carner waves from the PCM signal. A phased-lock loop circuit 158 receives the outputs of the multiplier element 156 on lines 160 and 162 and produces two outputs on lines 164 and 166 which are signals having frequencies which are the same as the actual carrier frequencies of the signals directed to the inputs of the multiplier element 156 on lines 152 and 154. The 15 frequency signals on lines 164 and 166 are used by the multiplier element 158 to demodulate the PCM signals on lines 152 and 154. The outputs of the mllltiplier element 156 on lines 168 and 170, which comprise (1~moclnl~ 1 in-phase and quadrature components of the input PCM signal, are directed to the inputs of theconstellation analyzer 124. The outputs on line 168 and line 170 comprise the actual 20 c~nstellation of the trs~n~mitting modem. The constellation analyzer determines differences between the con~tell~tion as represented by the signals on lines 16~ and 170 and an ideal con~t~ tion expected in light of the knowrl modem characteristics.
Certain adaptive processes in the clemo(l-~ tor or FIG. 10 allow some of the in-service quality measurements to be derived. These measurements include 25 signal level, timing offset, attenuation distortion, envelope delay distortion, and frequency offset.
Signal level is the power level of the input PCM signal. The signal level may be measured by observing two palametel~ in the ~utom~tic gain control process.
The signal level is computed by dividing the average power of the AGC output by 30 the square of the AGC gain. Both of these parameters are available as byproducts of any AGC operation. No ~ itic)n~l computation is required. The signal level couldalso be measured directly at the input.
Timing offset is the slight difference between the transmit baud rate and the receive baud rate. ~his offset is most co.~ lonly caused by a difference between 35 a reference clock in the receiver and a clock in the tr~n~mitting modem. The timing offset is measured from a velocity parameter of the timing recovery process. Timing ~, .

may be recovered ~sing a band edge timing recovery algorithm.
Attenuation distortion and envelope delay distortion are linear impairments related to deviations from an ideal channel response. Attenuation distortion is the deviation of the magnitude of the channel frequency response from S being ideal, namely, a channel amplitude response which is flat. Envelope delay distortion is the deviation of the envelope delay of the channel response from being ideal, namely, a channel frequency response which is flat. Flat envelope delay is equivalent to linear phase response since envelope delay is de~ined as the negative derivative with respect to frequency of phase. The ~tt~nll~tion distortion and 10 envelope delay distortion may be measured by examining the amplitude response of the adaptive equalizer filter, which has a characteristic which is the inverse of the channel response. The adaptive equalizer filter adjusts its amplitude and phase response to cancel the ~mplitude and phase distortions introduced by the networkinto the PCM signal. Thus the inverse of the frequency ~ yonse of the adaptive 15 equalizer provides an estim~inn of the frequency response of the channel. In order to provide measurements at the tra(lition:3l frequencies used for measuring linear paillllents, such as 1804 Hz and 1004 ~Iz, the channel response may be computed by Fourier transform techniques.
The frequency offset measurement is the mea~u~ ,nt of a shift in the 20 carrier frequency from its desired value. The frequency offset may be measured from a parameter in the phase-lock loop circuit 158 which recovers the carrier frequency.
The con~t.oll~tion analyzer 124 may measure signal-to-noise ratio, gain hits, phase hits, impulse noise, amplitude jitter, phase jitter, ~mplit~ . modulation, 25 and phase modlllation All of these mcasul~lllellts are obtained by exS~minin~ each output received from the ~lemoduls1tor 122 and comp~ring the location of each output to the location of the nearest ideal con~tell~tion point.
FI&. 11 shows in more detail the functional modules of the constellation analyzer 124. The outputs of the Cartesian demn~ trr 122 are directed to a circuit 30 172 on lines 174 and 176. The circuit 172 measures the overall signal-to-noise ratio of the signals produced by the demodulator 122. The overall signal-to-noise ratio is calculated by the circuit 172 as the ratio of the ~emoclul~tor's output power to the power of an error signal. The e~ror signal is the deviation of each clem- dnl~tor output point from the closest ideal cc n~t~ tion point. The overall signal-to-noise 35 ratio computed in this way gives a good indication of the likelihood that thetransmitting modem is making bit errors. However, this measurement does not - - -., .~ ~: . ............................. ,;., . .. :
.

2 ~ e~
-2~-reveal whether or not the impairment faced by the modem is an uncorrelated impairrnent such as additive white noise or a correlated impairment such as frequency or phase modulation. The circuit of FIG. 11, therefore, also makes measurements which separate certain correlated illlpailllle~ from uncorrelated 5 impairments. Specifically, the circuit of FIG. 11 measures an uncorrelated signal-to-noise ratio similar to the measurement of overall signal-to-noise ratio. Somepreliminary processing is pelrolllled on the constellation error signal to remove the effects of correlated imp~innPnts. In this regard, the output of the demn(l~ torl22 is directed on input lines 178 and 180 to two inputs of a circuit 182 which computes 10 the phase error and the m~gnit~lde error of the received constellation points with respect to their nearest ideal constellation points. The error computation circuit 182 produces two outputs on lines 184 and 186 which relate to the radial component of the difference between the received con~t~ tion point and the ideal con~tell:ltion point and the angular component of the difference between the received conitçll~ti~-n 1~ point and the ideal conctell~tion point. The two signals on lines 184 and 186 are delivered on lines 188 and 190 to the inputs of a high pass filter 192, which rnay be configured to eliminate low frequency components below about 300 Hz from the phase and msl~nit~ errors received from the eIror co~llpulaLion circuit 182. Thecutoff frequency of the high pass filter 192 is set at a frequency such that the phase 20 and amrlitll-le m~~ tion components in the error signals are removed. Typically, those components occur at frequencies below the 300 Hz cutoff frequency noted above. The high pass filter 1~2 produces filtered phase and amplitude error signals on lines 194 and 196 which are connected to the inputs of a second signal-to-noise ratio measuring circuit 198. The circuit 198 then cornputes an error power from the 25 filtered phase error signal and the filtered ~mplitu~l~9 signaL This power computation may then be scaled to replace the power of the white noise removed by the high pass filter. Because the uncorrelated signal-to-noise ratio measurement circuit 198 measures the signal-to-noise ratio only after removing phase and ~mI~litude m~d--l~tion, it provides a good indi~tion of the presence or absence of addit;ve30 white noise.
When the predominant impairrnent is uncorrelated noise, the overall signal-to-noise ratio measurement produced by the circuit 172 and the ~mcorrelated signal-to-noise ratio measurement produced by circuit 198 will be close in value.
However, when the predomin~nt impairment is arnplitude mod~ tion or phase 35 mod~ tif~n~ the overall signal-to-noise ratio will be significantly worse (smaller) than the uncorrelated signal-to-noise ratio.

- :,. ............. , , . .......... :
.. . . ... .

,. .. . .. ... ....

-'' 2 ~

Since the demodulator's adaptive equalizer 150 removes noise outside the transmitted energy band, modems with trs~n~mitte(l energy in different frequency bands would report different signal-to-noise ratios for a given amount of additive noise. If the signal-to-noise ratio measurements are unscaled for this phenomenon, 5 the signal-to-noise ratio mea~ul~l.lellts would result in dif~erent reported numbers for different modems e~en though the amount of noise is the same. To facilltate comparison between the relative quality of ch:3nn~l~ on which modems with transmitted energy in different frequency bands were observed, both the overall signal-to-noise ratio measurement and the uncorrelated signal-to-noise ratio 10 measu~ ent are scaled to reflect the signal-to-noise ratio that would have been measured had the tr:~n~mitte~l energy been in a frequency band which is 2400 Hz wide, which is the width of the frequency band tr:ln~mits~cl by a V.29 modem. The scaling assumes that the noise is evenly distributed across such 2400 Hz wide frequency spectrum.
In addition to the circuits describe~l above, the circuit of FIG. 11 contains a number of devices which measure transient i~ ent~ which are not st~ion~ry in time. These impairments include impulse noise, gain hits, and phasehits. Transient i..l~aill.lents are measured by counting the number of transientevents or "hits" which occur during an observation period. The number of hits 20 counted has a useful m~ning only in the context of the observation time. For example, a hit occ~lrring during a one second time period has a meaning which isdifferent from that of one hit occurring in a 100 second time period. Longer observation times allow one to make reasonable ~ses~mP.nt~ of the extent to which hits are appearing at regular intervals in f~csimil~ tr~ncmi~sir)n~
Two parameters are used in this example of the invention to implement the transient measurements. They are qualifying time and blocking time. Qualifying time is how long a transient impairment must exist before it qualifies as a hit. There is no qualifying time used for impulse noise measurements. Examples of qualifying times for gain hits and phase hits may be about 5.0 msec. for V.27 transmissions and 30 about 2.5 msec. for V.2g tr~n~mi~sions. It should be noted that modems which produce more complicated constell~ionc than those produced by V.27 and V.29 rnodems require shorter qualifying times because spatial aliasing will cause theimpairment to appear to persist only ~or a few symbols at a time. Blocking time is the amount of time after a hit has been counted during which a new hit cannot be35 counted. For all of the transient mea~ulGme~ , the blocking time may be about 125 msec. Thus, in a given second no more than 8 hits of a transient irnpairment will be , ~ . . . . . . . ~

counted.
A measurement circuit 200 in FIG. 11 receives the democlul~or output on a pair of input lines 202 and 204. The measurement circuit 200 measures impulse noise in the demodulator outputs. The measurement circuit 200 colllpa c;s the 5 inct~nt~neous power of the error of each ~lemo~ tor output to the average power of the constellation currently being received. The ratio of the error power and theaverage power is related to the impulse noise. Impulse noise ~lleasul~lllents may be referred to a number of different severity levels. In this exarnple of the invention~
impulse noise is referred to three levels of severity, for example, -7 dB, -12 dB, and 10 -18 dB. FIGs. 12a to 12c show circles corresponding to these thresholds as applied to a V.29 9600 bit per sec. constellation. If a demodulator output does not lie within any of the circles associated with a particular threshold, then an impulse noise hit is counted. To qualify as an impulse noise hit, a ~emod~ tor output must fall outside of all 16 circular regions associated with that threshold. It should be pointed out that 15 -7 dB impulse noise on the inner constellation points, as shown in F~G. 12c~ will not be accurately measured because the threshold circles overlap. If the error in the demodulator output is bad enough to exceed the -7 dB threshold, then the error is bad enough to cause the demodulator output to be miet:~kPnly ~soci~tPd with the wrong con~tell~tion point. This phenomenon is known as spatial aliasing. For large 20 errors in the ~em~ tnr output, spatial aliasing degrades measurement accuracy.
In these sitll~tinng, measurements of only the outer constellation points may beadvisable.
In FIG. 11, a measurement circuit 206 receives the output of the phase and m~nituflP error measurement circuit 182 on lines 208 and 210. The 25 measurement circuit counts the number of gain hits and phase hits in the modem tr:~n~mi~ion Gain and phase hits are measured by examining the m~nitlldP errors and the phase errors of the demod~-ln~or outputs. Gain hits are counted when an increase or decrease in power level of a predeterrnined amount persists for more than the qualifying time. For example, a gain hit may be counted when the power level30 increases or decreases by about 1 dB or more with respect to the power of the closest ideal constellation point and persists for more than the specific qualifying time referred to above. Phase hits are counted when phase deviations of more than a predetermined amount occur for longer than the qualifying time. For example, phase hits may be counted when there is a phase deviation of greater than about 7 ~ with 35 respect to the phase of the closest ideal conctPll~tion point for a period of time equal to the specific qualifying time referred to above. In the cases of both gain hits and 2 ~

phase hits, the hit counter of interest must not be in a blocking period for it to be incremented by the occurrence of these gain and phase deviations.
The phase and m~gni~ e error computed by the circuit 182 are also directed on lines 212 and 214 to the inputs of an amplitude and phase jitter 5 measurement circuit 216. Amplitude jitter is the mslgnituf3t- component of theconstellation error and phase jitter is the phase component of the constellation error Although these measurements may be made for all tr~ngmi~sions, it is preferred that they be made only during a short period at the beginning of the tr:-ngmi~gion, for example, during the first 4 seconds of the tr~ngmig~ion. These amplitude and phase 10 jitter measurements are most useful for modem tr~ncmigc;on.c which are too short for pe~ru~ ing accurate amplitude m~lnl~tinn and phase mo~lul~tion measurements described below. The measurement circuit 216 first C5lk''ll~teS amplitude jitter by colnlJu~ing the mean square gain error of the (lemol1nl~t~r output. This computed number then is scaled to give the percent of a-m--plitude m~nl~tion which wûuld 15 produce the computed mean square gain error. Phase jitter is coll~puled by the measurement circuit 216 as the mean square phase error of the ~1em~ tor output.
This number is also scaled to give the peak-to-peak degrees of phase mcxl-ll~ltil-n which would produce the observed phase error. The mean square phase error may becomputed from the phase error observed in the phase locked loop of the tlemn I~ tor 20 described above. The phase error could also be computed directly from the received constellation points.
The circuit of FIG. 11 also computes certain amplitude and phase mod~ tic)n char~rt~rigtics of the flemnd~ tûr output. AmplitlldA and phase modllls~tion mea~ enls indicate the severity of certain co,rrelated m~nitu~l~
25 errors and correlated phase errors in the ~lAm(x~ ted output. To remove uncorrelated components from measured impai~rnentg~ the m;~gnitudp error and phase e~or signals produced by the error computation circuit 182 are directed to two inputs ûf an adaptive predictive filter circuit 218 which removes the nonpredictive components of the error signals from the computation circuit 182. The outputs 220 30 and 222 of the adaptive predictive filter 218 are then processed by a first band pass filter 224 and a second band pass fil~er 226. The outputs of the band pass filters 224 and 226 provide sotne inforrnation about the spectrum of the phase and m~gnitucle error mod~ tions which can give some inrli~ ~tion about the source of the errors. For example, modulation of the msl~nitllde error at a frequency of 60 Hz and its 35 harmonic frequencies may indicate power line interference with the f~rsimile tr~ngmi~sion being monitored. As shown in FIG. 11 the band pass filter 224 has a - . . . .. . .

0 ~

pass band from about 4 Hz to about 20 Hz. The band pass filter 226 has a pass band from about 20 Hz to about 300 Hz. A peak detector 228 and a peak detector 230 measure the severity of the modulations indicated by the outputs of the band pass filters 224 and 226, respectively. Amplitude mocl~ ;nn~ are determined by S measurement circuits 232 and 234 in percent gain. Phase mr,(1~ tir,nc are measured in peak-to-peak degrees by the measurement circuits 232 and 234.
This completes the description of the circuitry which derives the infnrrn~t;on needed to create the event strearn produced by the integrate and arbitrate circuit 74. The event stream may include signals le~lc,sentillg the occurrence of the 10 start of a f~rsimillo call, the end of a f~csimilf, call, 1100 Hz or 2100 Hz tones at the beginning of a facsimile call, demodulated V.21 messages, V.29 or V.27 trial tr~ncmi~cinnc and page signals, with and without ISQM ll.ea~ul~.llents, and non-f~rSimile events occurring during the call. The event strearn may include time stamps for each event, the duration of some events, and the energy in some events.
15 The event stream produced by the integrate and arbitrate circuit 74 in FIG. 8 is directed on a line 121 to the input of a fli~gnostic circuit shown in FIG. 13, which is located in the colllp-l~er 35. A block 236 causes the circuit of FIG. 13 to wait for the next event in the event st~eam. A block 238 delel,l~illes if the next event is the start of a call. If there is no start of call as inrlicated by block 238, the operation of the 20 circuit of FIG. 13 returns to the input of the block 236. At the start of a call, as in(lir~ted by block 238, the circoit of FIG. 13 initi~li7~s a call ~ulllmaly in block 240.
The call SUllllll&ly iS a group of data and mea~ lllenls which are produced during the course of a f~rsimil~ call. The call ~ lllll~y may comprise infonn~tion about whether the call was a f~rsimile call, a voice call, or other kind of call. The call 25 summary may also comprise infolTn:~ion about the kinds of activities that took place in a f~fsimile call as well as certain p~lrollllance measurements ~iso~ ted with the call. The information in the call summar,v is updated periodically as the call progresses.
After the call summary has been initi:lli7~.d in block 240, the circuit of 30 FIG. 13 waits for the arrival of the next event from the integrate and arbitrate circuit 74 as in~ic~d by block 242. When the next event arrives, block 244 makes a determinS~tinn of whether or not the next event con~titlltes the end of the call. If the next event is the end of the call as determined in block 244, then the call summary is fin~li7~d in block 246. If the next event, as determin~cl in block 244, is not an end-35 of-call event, then a determination is made in block 246 as to whether the event is one which is part of a ~csimile call or an echo of such an event. If it is found that .
. ~, ~ ~ ., ; ,. . . .

the event is a nonf~r~imil~ event, it is labeled as such in block 248 and a signal classification designation is updated. The operation of the circuit of FIG. 13 then returns the input of block 242 and repeats the operation of the circuit from that point.
If block 246 determines that the event is a f~csimil~ event or an echo of 5 a facsimile event, then block 250 makes a determination as to whether or not the event was a talker echo event. If block 250 finds that the event was a talker echo event, then it is labeled as such in block 252 and echo measurements are updatedaccordingly. Specifically, the level of the echo event as compared with the level of the primaTy event which produced the echo and the time delay between the 10 appearance of the primary event and the appearance of the corresponding talker echo event is noted in block 252. The operation of the circuit of FIG. 13 then returns to the input of block 242. If the determination in block 250 is that the event is not a talker echo event, then block 254 makes a determin~tion as to whether or no~ theevent is a listener echo event. If the event is a listener echo event, block 256 labels it 15 as such in the operation of the circuiL of FIG. 13 returns to the input of block 242.
If the event is not a listener echo event as detemlined in block 254, then the circuit of FIG. 13 makes a determin~tion in block 258 as to whether or not the event is a tone. If the event is a tone, the event is labeled satisfactory and the call SU111111~ / iS updated with an inrli~tion that a specified tone was found in the20 f~3r~imile call in block 260. The operation of the circuit of FIG. 13 then returns to the inpu~ of block 242.
If the event was not a tone, as tlet~rminecl in blc,clc 258, then the circuit of FIG. 13 makes a deterrnination in block 262 as to whether or not the event is a V.29 or a V.27 event having in-service quality measurements appended to an 25 inrlicatic)n of the occurrence of the event. If the event is a V.29 or a V.27 event with appended in-service quality measurements, then the measurements are compared in block 264 with a set of thresholds. The thresholds are predetermined signal leve~s deterrnined empirically or by some other means in-lin~ing values of the ISQM
ll.e&s~ ents expected for a properly pelrcilll~g network h~nflling f~esimile traffic 30 produced by properly pe.rolllling facsimile equipment for a network and f~simil.-.
equipment connected to the network. Block 264 characterizes the in-service quality measurements with respect to the thresholds in a plurality of ways. As inrlir~ted in FIG. 13, the in-service quality measurements may be labeled as good, bad, or marginal depending on how they compare with the levels of the thresholds.

, 2 ~

If the event from the circuit 74 is not a V.29 or V.27 signal with ISQM, or if the operation of block 264 has been performed by the circuit of FIG. 13, adetermination then is made in block 266 to see if the event is a V.21 digital information signal (DIS) or a digital transmit comm~nrl (DTC) signal. If the yes5 path on the output of block 266 is followed, then a determin~ tir n is made in block 268 as to whether or not the event was a repeat of a previous DIS or DTC event. If the event is such a repea~, then it is labeled as such in block 270 and the operation of the circuit of FIG. 13 returns to the input of block 242. If the event was not a repeat, as determined in block 268, a page orientation pa al-lelt;r is set in block 272, the call 10 summary is accordingly updated in block 276, and the operation of the circuit of FIG. 13 returns to the input of block 242.
If block 266 determines that the event does not involve a DIS signal or a DTC signal, a low }evel state machine 276 then tracks the protocols in the facsirnile tr~nimi~ion, labels the occull~,nce of certain events in the facsimile tran~mi~ion~
15 notes any observed ~nom~liçs such as failures of mlul~m~ to train, and further updates the call ~ulllm~y with information such as page counts, data rates, manufacturers, identific~tions, fall backs, retraining, durations of tr~n~mi~sion~ and errors in page tr~n~mi~sinn FIG. 14 illustrates a diagram of some of the states which may be entered 20 ints) by the low level state machine 276 of FIG. 13 in r~s~ol~se to an example of a typical G3 f~rsimil~ call. The state m~hin~ 276 begins at an initial state 278.
Detection of the receiver sending to the tr~n~mitter a digital i~l~ntifi~tic)n signal DIS(R~ in accold~lce with the T.30 protocol, causes the state machine 276 to enter into a start state 280 which reflects the fact that a DIS signal was detected.
25 Continued DIS mes~ges from the receiver cause the state machine 276 to remain in the start state 280. Eventually, continued sending of unanswered or i~lplup~,~lyanswered DIS m~ss~es from the receiver will cause the receiver to terminate the call by sending a disconnect DCN(R) signal and the state machine 276 will then enter a done state 282.
The next proper signal normally present in a G3 f~e~imi1~ call after the receiver sends a DIS signal is a digital comm~ntl signal DCS sent by the tr~n~mitt~r to the receiver, identified in FIG. 14 as a DCS(X) signal. Sending of a DCS signal from the trancmi~ter to receiver causes the state m~hin~ 276 to enter a DCS state 284 indicative of the sending of a DCS signal from ~he tr~n~mitter to receiver. The 35 next proper signal in a G3 f~simile call is a trial ~ran~mi.~sinn which may be a V.29 tr~n~micsion from the tr~n~m;ster to the receiver iclentifi~d as V.29 (X), as shown in ', : ' :,, ' '' - ' , - ~, , FIG. 14. The state machine 276 then enters a training check ~TCF) state 286 because the first such V.29 signal will be a training sequence from the transmitter to the receiver.
Proper reception of a training sequence by the receiver causes the 5 production of a confirmation signal CFR(R) sent by the receiver to the tr~n~mitter and entry of the state machine 276 into a conrlllllation tCFR) state 288. A
disconnect DCN(X) signal from the tr~n~mittPr causes the state machine to enter a done state ~90.
If there is an i~ upes reception of the training sequence, the receiver 10 produces a failure to train signal F'I~(R) and the state machine 276 enters a failure to train (E~) state 292. A disconnect signal DCN(X) from the tran~mitter at this time causes the state machine to enter a done state 294. The occurrence of a digital comm~n~ signal DCS(X) from the tr~n~mitt~r when the state machine is in state 2~2 or state 286 causes the state machine to reenter state 284.
After conrllmation of proper receipt of the training sequence and entry of the state machine into state 288, a page signal is sent frorn the tr~lnimitter to the receiver. As shown in the example of FIG. 14, the page signal may be a V.29 signal.
In response to the presence of the V.29 signal, the state machine 276 enters a page state 296. If the f~c~;mile tr~n~mi~inn involves sending only a single page to the 20 receiver, the tran~mitter sends an end of procedure signal EOP(X) to the receiver at the completion of the page signal and the state m~hin~ 276 enters an end of procedure EOP state 298. Cr~ntinn~d sending of end of procedure signals to the receiver results in the state machine 276 rem~ining in state ';298. Production of a disconnect signal DCN(X) by the tr~n.imitter results in the state machine entering a 25 done state 300.
If the receiver properly receives the page signal and the EOP signal from the tr~ncmitt~r~ the receiver sends a message confirms~tion signal MCF(R) to thetr~n~mit~er and the state machine 276 enters a message confirmation MCF state 302.
A fiiicr,nn~ct signal from the tr~nsmitter to the receiver DCN(X) when the state30 m~shine is in stata 302 causes the state machine to enter a done state 304. When the state machine 276 is in the end of procedure state 298, a retrain positive signal or a retrain negative signal RTP/RTN(R) from the receiver to the tr~n~mitter causes the state m~shine 276 to enter an RTN/RTP state 306. A disconnect signal from the tr:~n.~mitter to the receiver DCN(X) when the state machine is in state 306 causes the 35 state m~shin~ to enter a done state 308.

' ~ :,' .: ' ~ , . ~
- , , .- . . :

If the facsimile tr~nemieeion involves more than one page, the tr~nemitt~r sends to the receiver a multipage signal MPS(X) when tr~nemiesion ofeach page has been completed. The state machine 276 in this situation enters a mllltip~l~e signal MPS state 310 from the page state 296. Continued sending of aS m-lltipage signal from the tranemitt,or to the receiver causes the state machine 276 to remain in the MPS state 310 as shown in FIG. 14. If the receiver has properly received the multipage signal from the tr~nemi~t~r, the receiver sends a messageconfirmation signal MCF(R) to the tr:~nemitter and the state machine 276 enters a message confin~l~tion state 312. The IlAllelllillto( then sends the next page in the 10 tr~nemiesion, for example, the tr:~nemitt~r sends the next page by a V.29 tr~nemiesion as inflic~t~d in FIG. 14. The state machine enters the page state 296 in response to this page tr~nemigsion. The state m~chine loops between states 296, 310, and 312 as each page is properly tr~nemitted and received in the course of the fareimilP tr~nemi~sion. The production of a ~1ieconnect signal ~om the tr~nemitter to 15 the receiver causes the state machine to enter a done state 314 from st~te 310. If the receiver sends a retrain negative signal or a retrain positive signal RTN/RTP(R) to the tranemitter in the course of a page tranemieeion~ the state m~rhine 276 enters an RTN/RTP state 316 shown in FIG. 14. A digital co~ ti signal DCS(X) from the tr:~nemitter to the receiver causes the state m~rhinP.276 to re-enter the DCS state 284 20 from state 316.
The states which were entered into by the state m~chine during the course of a facsimile tranemiesion are recorded for use in producing char~-rteri7:-tions and cli~gnoses of the f~rcirnile tranemieeirlrl observed by a facsimile measuremen~ apparatus in accordance with this invention.
FIG. 15 illustrates the data base architecnlre in the computer 35 shown in FIG. 2. The apparatus of FIG. 15 includes a trunk ~tnb~se 317 which is connected to the access control unit 40. The trunk r~ b~ee 317 contains relevantinform~tion about all the trunks being processed by the network node 42. This information may include some iclentifi~ nn number for each trunk and the 30 geographical areas served by each trunk. For call or session control, the tr mk database 317 will be ~ccessed and comm~n-le will be issued to the access controlunit 40 from the trunk rl:lt~b~ee 317 requesting the monitoring of specific portions of the co,~."~ tions traffic through a network node as described above. As described in the exarnple above, both directions of a selected call represented by two 35 DSO's will be placed on two adjacent DS0 time slots of a DSl bit stream flowing between the network node 42 and the access control unit 40. These two adjacent time slots are tapped by the service signal processor 34 by means of the bridging repeater 48.
The database architecture of FIG. 15 also contains an event database 318 which is responsive to signals produced by the service signal processor 34. The S database 318 collects and stores representations of the specific events observed by the service signal processor 34 in the course of monitoring one or more facsimile calls. A diagnostic module 320 is responsive to the contents of the event database 318 and determines if there are any abnormalities in the calls observed by the f~r~imile measurement apparatus. The diagnostic module 320 tracks the protocol 10 events occllrring on both directions of each fax call. The (ii~nostic module 320 generates appropriate information about the nature of the fax call which can be called up on the user intPrf~ce The rli~gnostic module 320 contains templates ofnorrnal f~ imil~ calls which are compared to the events occurring in an actual call.
Exceptions to normal opera~ions in the f~rsimile protocols are noted and may be 15 appropriately displayed on the user interfaçe 36 along with the results of the nonintrusive in~ nt, echo, and delay measurements described in more detail above. When a f~rsimilP, call has been completed, the rli~gnosti~ module 320 uses the information collected about mea~ulemell~s and events to render an overall decision about the reason for the occurrence of an abnormal f:~r~imile call. Forexample, the diagnostic module 320 may indicate that an abnormal call was causedby an undue amount of echo or noise or that there was some protocol incompatibility between the sending and receiving f~rsimill~ m~rhines, The t~ nr)stir module maysu~ ~ize the results for the user and may send appropriate data to a session database and a call database described below.
In addition to the diagnostic module 320, the circuit of FIG. 15 contains a call summary module 322 which organizes certain inform:ltion about every facsimile call observed by an apparatus in accordance with this invention. The call s.n~ is a collection of data in~liçating certain normal and abnormal behavior ofeach observed ~csimil~ tran~mi~ion The data in the call ~u~ ~y may be used for 30 subsequent analysis and traffic characterizadon. A call ~ulllm~y may contain the following illustradve information:
1. The number of pages in the incoming and outgoing directions;

2. Inform~tion about the manufacturers of the machines used in the f~r~imile call;

- - . ~

.. ~ . . . . . . ...
- - -, ,. . ~

'2~ 3~

3. Information about the geographical areas involved in the f~l~simile call;
., 4. Information about the numbers of pages tr~ncmitted at each of a plurality of bit rates, for example, 9600, 7200, 4800, and 2400 bits per second;

5. The speed of the first trial or training tr~n~mi~sion;

6. The speed of tr~ncmission of the first page;

. The average page duration at each of a plurality of bit rates, for example, 9600, 7200, 4800, and 2400 bits per second;

8. The number of retrains before the tran~miccion of the first page;

9. The number of retrains after the tr~n~mi.~cion of ehe first page;

10. An intlic~tion of the presence of a ~ l~uulld, specifically an in(lic~tion of whether the receiving f~ imilP. m~chine was requested to send pages to the tr7~n~mitting f~simih m~- hin,o.;

11. An indication of whether an error correction mode was used, for example, a T~30 error correction mode;

12. An indication of the existence of partial page tr~n.~miccinns;
13. ~n~lin~tion.~ of the presence of an echo protection tone, an auto-originate tone, or an auto-answer tone;
14. An inrlie~tion of normal and abnormal termin~tion~ of the f~csimile tr~nimission; and 15. The results of call diagnostics, ~or example, the results of protocol tracking and in-service quality measurements and the results 2 ~

of comparison of those results to a set of predetermined expected mea~u~ el ts.
Information such as the informadon described above is accllm~ te(l by the call Sull~ll~y module 322 for each call during a monitoring session. A call 5 l1~tS~b~e 324 stores the call sul-llllal;cs produced by the call ~ullllllal y module 322.
When this apparatus is used in a directed access mode, this information is collected for the specific call which is being monitored. A session Sul~ ~y will be accnm~ t~(l for all calls in a given session. Further SUI11111~iCS may also be accumulated for selected items at the trunk, trunk sub-group, and geographical 10 region level.
A session rl~t71tl~5e 326 contains a superset of the inforrn~tion in the call ~ t:l~ase, specifically, the session 13~t:~b~e 326 contains infnrm~tion regarding mnnitoring sessions occurring over a predeterminecl time which may encompass more than one f~r~imil~. call. The operation of the databases 317, 318, 324, and 326 15 are controlled by a dat~base management module 328. The contents of the databases and the output of the diagnostics module 320 may be displayed to a user via a user interface described in more detail below.
The details of the user int~rfare 36 are illustrated by a series of represent~tinn~ of various screens which may be produced on a mr,nitnring device20 connected to the computer 35 shown in FIG. 2. These representations are shown in FIGs. 16-25. There are two display modes, each respectively associated with the previously described monitor mode and directed access mocle. There is also a retrieval display mode in which certain data collected in the past and now stored in the ~ t:lb~es described above may be retrieved for display to a user of an apparatus 25 in accordance with the principles of this invention.
In the monitor display mode, the user may request monitoring of all calls made during a predetermined time period on a specific portion of the co~.",.~ ir~tion~ ch~nnels handled by the network node 42. The llloniloring for a predetermined time is referred to as a session. The user will be able to view 30 measurements and st~tietirs about each call as it passes through the network node 42.
These measurements and statistics may be displayed on a computer screen as they are generated in real time during the course of each call. In addition to displaying inform~tion about an ongoing f~csimil~ call as it takes place, the user may alsorequest and a view overall statistics for some or all the prior calls measured to date 35 during the session. In the directed access mode, where access to a particularfacsimile call is set up in response to a specific customer request, the user normally ~ -; - ~ : . . ., :

2~

will view the events associated with that call as it progresses in real time. The events may also be saved for later display. In the analysis mode, the user can retrieve data from any past measurement session and can either display information in a manner similar to that of the display of live calls or the user can view a series of 5 report summ~es about the measurement sessions.
The user interface 36 will make available to the user a menu from which the specific mode of operation can be selected. Specifically, a computer screen connected to the colllput~r 35 may display the options of looking at data from amonitoring mode or a directed test access mode. The computer 35 may also indicate 10 the availability of an access or retrieval mode. FIGs. 1~-25 illustrate the behavior of the user interface in the retrieval mode which illustrates the entire c~p~bility of the user interface. Operation in the other modes are essentially s~milar and are notdescribed here.
If the user selects the retrieval mode, he is given the opportunity to 15 select the display of certain categories of in~onn~ion, as shown in FIG. 16. The user may select to display information about f~esimile calls directed to one or a number of geographical areas, for example, the user may elect to display data about f~rcimile calls to a certain country or to display data handled by a certain trunk sub-group which may handle f~c~imil~ calls to a certain city within a country. As shown in20 FIG. 16, the user may also elect to observe the farsimile calls made on a pa~icular trunk, in this case, the user may elect to observe the calls made on a precletennined DS0 handled by the network node 42.
Selection of a particular country in the example shown in FIG. 16 will cause the display to produce a list of countries and cities such as the list shown in 25 FIG. 17. FIG. 17 comprises a menu of options from which the user can select acountry or city to observe. A f~csimile lneasulement apparatus in accordance with this invention will provide information about the nature of the f~r~irnilP traffic to that country and city.
FIG. 18 shows a list of trunks which comprise a selected trunk sub-30 group from the menu of FIG. 17. The screen of FIG. 18 shows in the far right handcolumn the number of sessions stored in the facsimil~ analysis equipment for which data has been taken for each listed trunk. The user may then select for observation the results of measurements on ~nks indicated in FIG. 18 as having had one or more monitoring sessions.

.. . .

2 ~ 0 ~

FIG. 19 shows an example of a sessions list which can be produced in response to selection from the list of FIG. 18 of a trunk for which measurementshave been taken and stored in the f~csimile analysis equipment. The sessions list of FIG. 19 identifies the start time of the session, a session number, the number of calls 5 monitored during the session, and the number of calls which were facsirnile calls.
Further info~n~tion about a session listed in a sessions list like the one in FIG. 19 may be produced in a sessions su~ llaly such as the one shown in FIG. 20. Among other things, the session ~u~ ~ y identifies certain characteristics of the session and the calls within the session. The session ~iUllllll~lly also id~ntifi~s certain diagnostic 10 and statistical aspects of the measurements made on the farsimilç calls made during the session. As intli~ ltçd in FIG. 20, certain percentages of the calls have been ident;fi~d as good, marginal, and bad as intlic~ted by the columns headed by these ~le~ignP.tiC~ni. The session sul~ may also indicate tabular data relating to numbers of pages and a time duration of tr~ncm~ n at specifiPd rates.
The user interface 36 may also produce a display of a call list which basically identifies the start time of each call in the session, identifies each call with a number, id~ntifi~s the number of measured events in each call, characterizes the type of each call, and gives an indi~ion of the duration of each call during a session.
See FIG. 2}. The call list may include info~n~ti-~n for every call which was 20 observed during the monitoring session or the non-facsimile calls may be filtered from the list as they have been in FIG. 21.
The user may call up a display of further inÇc,~ alion about selected ones of the calls listed in the call list of FIG. 21. In this reg~lrd, the previously described call ~UIllll~y for the selected call may be displayed, one example of which 25 is shown in FIG. 22. The info~n:l~ion in FIG. 22 is self-exrl~n~t~-ry in light of the discussion of call summaries above and, therefore, ls described no further here.The user interface 36 of a f~csimil~ measurement apparatus in accordance with the principles of this invention also permits a user to display a list of the observed events which had occurred in each measured f~rsimile tr:~ncmiision.
30 FIG. 23 shows such a list for the call summary illustrated in FIG. 22. Each event is numbered and the time of occurrence for each event is noted near the left hand margin of the display shown in FIG. 23. As inrlic ited by the nonconsecutive event numbering in the far left hand column of FIG. 23, some of the call events have been filtered from the display. Those events filtered from the display of FIG. 23 may35 include some idle events and activity events caused by noise and the like which can be ignored as irrelevant artifacts in the faesirnile tr:~nsmi~siQn. The display shown in ~ .. .. . . . . . .

2 ~

FIG. 23 also indicates whether or not each event was produced by the tr~n~mitting f~rsimile machine or the receiving facsimile machine. The events produced as a result of transmi~sions from the receiver to the transmitter are listed in a column headed by the word "incoming". The events produced as a result of tr~n~mi~sions S from the tr~n~mitter to the receiver are listed in a column headed by the word"outgoing." For each event, the display of FIG. 23 in~lic~tes the signal level of the tr~n~mi~sion and the kind of event which took place, for example, there is a display of the fact that the event was the result of a V.21 HDLC tr~n~mi~ion, a V.29 9600 or 7200 bps page or training tr~n~micsion, or a V.27ter 4800 bps page or training 10 tr~n~mic~ion.
FIG. 23 also indicates some of the content of the protocol m~ss~ges for the V.21 events and also in(1icates whether the V.29 and V.27ter are page or trial tr~ncmi~ ns The display of FIG. 23 also contains rli~gnos~ info. .,~lir~n deduced by the ~rsimil~. analysis equipment. This diagnostic information can include some 15 in~liratisn of the signal level and quality of each tr~n~mi~sil n which produces the observed event~s. The event numbered 25 at the top of list in FIG. 23 was a V.21HDLC tr~nsmi~ion from the receiver to the t~n~mitter in accordance with the T.30protocol having a signal level of -17.20 dBm. The event 25 was iclentifi~l by large cih~uill~ a~soci~ted with the V.21 clem~xh~l~tor 106 described above as a properly 20 tr~n~mitted protocol message involving a called subscriber identifi~;~tinn CSI and a digital iclentifil~ti- n signal DIS from the receiver to the tr:~n~mitt~r. Event 25 is accordingly labeled OK. The box surrounding the infnrm:~tit~n about event 25 in FIG.
23 may be colored with a suitable color inriir~ting that the tr~nimicsic!n was proper.
For example, the box may be colored green. Event 28 was the result of a proper 25 V.21 HDLC protocol tr:~nimig~ion from the tr~ncmit~er to the receiver involving a digital comm~n-l signal DCS and is labeled OK. Event 28 indicates that the tr~nemitte~ will send page data to the receiver at 9600 bps. As in the case of event 25, event 28 may be colored green. Event 29 in~ teS that the tr~ncmit~r next sent a V.29 9600 bps trial tr:~n~mi~sion TCF to the receiver. The SNR clesign~tion 30 inrlic~tes that the in-service quality monitor found that the signal-to-noise ratio was bad during the trial tr~n~mi~sion The box surrounding the inform~tiQn about event ~9 may be colored in a manner in(lic~ting that the event was bad, for example, event 29 may be colored in red. Event 32 was a V.21 HDLC tr~n~m;~sion from the tr~ncmitter to the receiver repeating the digital comm~n-1 signal of event 28. Event 35 32 is labeled as a repeat CMD REPEAT of a prior comm~nfl which may be considered a marginal event. Marginal events such as event 32 may be given a - 40 ~
suitable color such as yellow. Event 34 is an attempt to repeat the training tr~n~micsion of event 29 since there was no response from the receiver. Again, the tr~n~mi~ion was bad and an indication is placed next to event 34 that this was not the first tr~n~micsion which was bad. Again, a bad event like event 34 is colored red.
5 Event 38 is a V.21 HDLC tr~nsmicsion from the receiver to the transmitter in~ ting that there was a failure to train ~ l l . Failures to train such as event 38 are considered marginal at this point and event 38 accordingly is colored yellow. Event 38 is also in~1ir;~ted by the diagnostic module to have resulted from a bad training sequence BAD TCF. Event 41 is a V.21 HDLC tr insmi~sion from the tr~n~mitter to 10 the receiver involving a digital c-3mm~n~1 signal DCS notifying the receiver that the tr~n~mirt~r will fall back to a tr~n~mii~ion rate of 7200 bps for the trial tr~n~m;~sion Event 41 is labeled OK and maybe coiored green. Event 44 is a V.29 7200 bps trial tr In~mic~ion ~rom the tr~ncmi1ter to the receiver for which the in-service quality monitor has in~lir~ted a marginal rating. Event 44 is intliraterl as not being the first 15 rnarginal tr~ln~mi~;on and is colored yellow. Event 47 is a Y.21 HDLC tr:~n~mi~sion from the receiver to the tr~n~mitter involving a con~.llation CFR of the receipt of the trial tr~nsmic~ion of even~ 44. Event 47 is labeled OK and maybe colored green.
Event 50 is a V.29 7200 bps page tran~mi~ion from the tr~nimitt~r to the receiver.
ISQM measurements indicate that the page tran~mi~sion was bad due to poor 20 signal-to-noise ratio. Event 50 accordingly is labeled as not being the first bad tr~n~mi~sion and may be colored red. Event 55 is a V.21 HDLC signal sent from the tr~nsmitt~r to the receiver involving a mnltir~e MPS signal which is labeled OK
and maybe colored green. Event 60 is a repeat of the V.21 HDLC of event 50.
Event 60 is labeled a comm~nrl repeat CMD REPEAT which occurred because the 25 tr:~ncmitter did not receive a response to the multipage signal of event 55. CommS~nri repeats such as event 60 are considered marginal events and may be colored yellow.
Event 69 is a retrain negative signal which is sent by the receiver to the tr:~ncmit~er.
The display identifies the cause of the retrain negative signal as being a bad page and event 69 accordingly may be colored red. Event 74 is a V.21 HDLC signal 30 involving a digital c-mm~nd signal DCS in which the tr~n~mitter notifies the receiver that it will fall back to a V.27ter 4800 bps rate. The diagnostics identifies event 74 as being caused by a request to retrain. Event 75 is a V~27ter 4800 bpstraining sequence TCF sent from the ll,.n~ e~ to the receiver. ISQM
measurements have in(1it ated that this trial tr~n~mi~cion was marginal due to signal-35 to-noise considerations. Accordingly, event 75 may be colored yellow. Event 81 is a V.21 HDLC signal from the receiver to the tr~ncmit~er confirming the receipt of the - ,-. . ...

2 ~ 0 -3 training sequence. The diagnostics have labeled event 81 OK. Event 81 accordingly may be colored green. Event 83 is a V.27ter 4800 bps page tr~n~mi~sinn which hasbeen identified as bad in light of ISQM mcasu~ ents which indicate a signal-to-noise ratio which is too low. Event 84 is a V.21 HDLC sequence identifying the end 5 of procedure EOP which was OK and accordingly, may be colored green. Event 100 is a V.21 HDLC signal from the ~n~mitter to the receiver repeating the tr~n~mi~sic n of the EOP signal of event 84. As in the case of event 60, this is a C(lmm~n(l repeat CMD REPEAT which is considered marginal and may be colored yellow. The receiver next confirmed the receipt of the EOP signal in event 104.
10 Event 104 was OK and may be colored green. Event 108 is a proper V.21 HDLC
disconnect DCN signal which was OK and may be colored green.
A user of a f~n~imile measurement apparatus in accordance with the invention may call up a display of individual analog ISQM measu~ ent~ made by the apparatus for each event in the calls which have been ,lloni~c,led for which ISQM
15 has been made. FIG. 24 is a display corresponding to the event 50 of the call shown in FIG. 23. FIG. 24 shows the level of amplitude and phase jitter, the level of amplitude and phase modulation, the level of attenuation distortion at a number of different frequencies from about 604 Hz to about 2804 Hz, and the envelope delaydistortion at the same frequencies. FIG. 24 also reports various other measurements 20 in the right-hand column which are self-explanatory. FIG. 24 also shows an expanded ~ gno$tic message identifying a problem with the subject transmission.
In this example, all of the measu~ e~ were OK with the exception of bad measurements for the two signal-to-noise ratios S/N 0 and SIN 1 and the IMP 2 measurement.
A user of a f:lcsimile me&~ultilllellt apparatus in accordance with this invention may also display information about the actual bits which are in the protocol messages found in f~csimile tr~n~missions~ FIG. 25 shows such a displayfor event 32 in ~IG. 23. FIG. 25 also shows an expanded diagnostic message whichexplains the comm~ and repeat ~lesi~n:~tion shown in FIG. 23.

:

Claims (32)

1. A telephone network, comprising:
a facsimile analysis unit in the network for making non-intrusive impairment measurements of facsimile telephone calls handled by the network; anda means for connecting the facsimile analysis unit to a predetermined facsimile telephone call handled by the network in response to another telephonecall placed to the network by a user of the network for making at least one non-intrusive measurement of a selected parameter associated with the predetermined facsimile telephone call to identify an impairment of the predetermined facsimile telephone call.
2. The telephone network of claim 1, in which the means for connecting comprises:
a means in the network for answering a facsimile call from a transmitting facsimile machine;
a means for making another call from the network to a receiving facsimile machine at a predetermined destination for the facsimile call; and a means for establishing continuity between the facsimile call from the transmitting facsimile machine and the call from the network to the receiving facsimile machine.
3. The telephone network of claim 1, in which the facsimile analysis unit comprises a means for measuring protocol signals in the facsimile telephone calls.
4. The telephone network of claim 1, in which the facsimile analysis unit comprises a means for measuring page signals in the facsimile telephone calls.
5. The telephone network of claim 3, in which the means for measuring further comprises a means for measuring page signals in the facsimile telephone calls.
6. The telephone network of claim 1, further comprising an identifying means comprising a means for automatically classifying signals in the facsimile telephone calls.
7. The telephone network of claim 1, in which the facsimile analysis unit comprises a means for detecting echo signals in the facsimile telephone calls.
8. The telephone network of claim 1, in which the facsimile analysis unit comprises:
a means for classifying signals in the facsimile telephone calls;
a means for measuring transmission impairments related to the signals in the facsimile telephone calls;
a means for detecting echo signals in the facsimile telephone calls; and a means for interpreting protocol messages in the facsimile telephone calls.
9. The telephone network of claim 8, further comprising a means for integrating signals from the classifying, measuring, detecting, and interpretingmeans into a data signal representing an event stream in the facsimile telephonecalls.
10. The telephone network of claim 9, further comprising a means responsive to the integrating means for analyzing the facsimile telephone calls and diagnosing the facsimile telephone calls.
11. The telephone network of claim 1, in which the facsimile analysis unit comprises a means responsive to signaling in a central office switching system for identifying boundaries of the facsimile telephone calls.
12. The telephone network of claim 1, in which the facsimile analysis unit comprises a means for performing voice band signal classification.
13. The telephone network of claim 1, in which the facsimile analysis unit comprises:
a means for performing speed classification of signals appearing in the facsimile telephone calls; and a means responsive to the means for performing speed classification of signals for identifying at least one characteristic of one or more modems involved in the facsimile telephone calls.
14. The telephone network of claim 1, in which the facsimile analysis unit comprises a means for detecting predetermined tones in the facsimile telephone calls.
15. The telephone network of claim 14, in which the tone detecting means detects modem training tones in the facsimile telephone calls.
16. The telephone network of claim 13, further comprising a tone detecting means for detecting modem training tones in the facsimile telephone calls.
17. The telephone network of claim 16, further comprising a modem identifying means responsive to the means for performing speed classification ofsignals and also responsive to the tone detecting means to identify said at least one characteristic of one or more modems involved in the facsimile telephone calls.
18. The telephone network of claim 1, in which the facsimile analysis unit comprises a means for demodulating protocol signals in the facsimile telephone calls.
19. The telephone network of claim 18, in which the demodulating means distinguishes between primary and echo signals related to the protocol signals.
20. The telephone network of claim 13, in which the facsimile analysis unit comprises a means for demodulating protocol signals in the facsimile telephone calls.
21. The telephone network of claim 20, in which the facsimile analysis unit comprises a means for performing protocol tracking in response to the demodulated protocol signals; and a means responsive to signals from the modem identification means and the protocol tracking means for producing in-service quality measurement controlsignals.
22. The telephone network of claim 20, in which the facsimile analysis unit further comprises a means responsive to page signals in the facsimile telephone calls and to in-service quality measurement control signals for performing predetermined in-service quality measurements.
23. The telephone network of claim 1, in which the facsimile analysis unit comprises a means for measuring trail transmissions in the facsimile telephone calls.
24. The telephone network of claim 1, further comprising a tone detecting means for detecting modem training sequences in the facsimile telephone calls.
25. The telephone network of claim 23, in which the facsimile analysis unit further comprises a means responsive to trial transmissions and to in-service quality measurement control signals for performing predetermined in-service quality measurements.
26. The telephone network of claim 1, in which the facsimile analysis unit comprises:

a means for detecting protocol transmissions from a transmitting facsimile machine to a receiving facsimile machine and protocol transmissions from the receiving facsimile machine to the transmitting facsimile machine; and a means responsive to the detecting means for making non-intrusive impairment measurements of trial transmissions and page signals in facsimile telephone calls.
27. A telephone network, comprising:
a means for establishing a facsimile maintenance telephone number in a telephone network;
a means in the network for answering a facsimile call made to the facsimile maintenance telephone number by a transmitting facsimile machine;
a means in the network for placing a call to a receiving facsimile machine in response to a facsimile call made to the facsimile maintenance telephone number;
a means for establishing continuity between the call to the receiving facsimile machine and the call to the facsimile maintenance telephone number to create a facsimile call between the transmitting and receiving facsimile machines through the network; and a means for non-intrusively monitoring selected parameters in the facsimile call between the transmitting and receiving facsimile machines to identify an impairment in the facsimile call between the transmitting and receiving facsimile machines.
28. A method of making impairment measurements of facsimile calls in a telephone network, comprising the steps of:
providing a facsimile analysis unit in the network for making non-intrusive impairment measurements of facsimile telephone calls handled by the network; and connecting the facsimile analysis unit to a predetermined facsimile telephone call handled by the network in response to another telephone call placed to the network by a user of the network for making at least one non-intrusive measurement of a selected parameter associated with the facsimile telephone call to identify an impairment of the facsimile telephone call.
29. The method of claim 28, in which the connecting step comprises the steps of:
answering in the network a facsimile call from a transmitting facsimile machine;
making another call from the network to a receiving facsimile machine at a predetermined destination for the facsimile call; and establishing continuity between the facsimile call from the transmitting facsimile machine and the call to the receiving facsimile machine.
30. The method of claim 28, in which the connecting step comprises the steps of:
detecting protocol transmissions from a transmitting facsimile machine to a receiving facsimile machine and protocol transmissions from the receiving facsimile machine to the transmitting facsimile machine; and making non-intrusive analog impairment measurements of trial transmissions and page signals in the facsimile telephone call in response to the detecting step.
31. A method of making impairment measurements of facsimile telephone calls in a telephone network, comprising the steps of:
establishing a facsimile maintenance telephone number in a telephone network;
answering in the network a facsimile call made to the facsimile maintenance telephone number from a transmitting facsimile machine;
placing a call from the network to a predetermined receiving facsimile machine in response to a facsimile call made to the facsimile maintenance telephone number;

establishing continuity between the call from the network to a predetermined receiving facsimile machine and the call to the facsimile maintenance telephone number to create a facsimile call between the transmittingfacsimile machine and the receiving facsimile machine; and non-intrusively monitoring selected parameters in the facsimile call between the transmitting and receiving facsimile machines to identify an impairment in the facsimile call between the transmitting and receiving facsimile machines.
32. A method of making impairment measurements of facsimile telephone calls in a telephone network, comprising the steps of:
receiving a telephone call in a telephone network made to a customer service telephone number from a user of the network;
collecting information from the user relating to a telephone number of a transmitting facsimile machine and a telephone number of a receiving facsimile machine;
notifying the user of a maintenance telephone number;
receiving a facsimile call in the network from the transmitting facsimile machine at the maintenance telephone number;
making another telephone call to the receiving facsimile machine in response to the receiving step;
establishing continuity between the facsimile call from the transmitting facsimile machine and the telephone call to the receiving facsimile machine to create a facsimile call between the transmitting facsimile machine and the receiving facsimile machine; and making a non-intrusive impairment measurement of the facsimile call between the transmitting facsimile machine and the receiving facsimile machine.
CA002084105A 1992-02-21 1992-11-30 Subscriber initiated non-intrusive network-based analysis of facsimile transmissions Expired - Fee Related CA2084105C (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US07/839,975 US5299257A (en) 1992-02-21 1992-02-21 Subscriber initiated non-intrusive network-based analysis of facsimile transmissions
US839,975 1992-02-21

Publications (2)

Publication Number Publication Date
CA2084105A1 CA2084105A1 (en) 1993-08-22
CA2084105C true CA2084105C (en) 1998-09-08

Family

ID=25281134

Family Applications (1)

Application Number Title Priority Date Filing Date
CA002084105A Expired - Fee Related CA2084105C (en) 1992-02-21 1992-11-30 Subscriber initiated non-intrusive network-based analysis of facsimile transmissions

Country Status (6)

Country Link
US (1) US5299257A (en)
EP (1) EP0557055B1 (en)
JP (1) JP2622063B2 (en)
AU (1) AU646988B2 (en)
CA (1) CA2084105C (en)
DE (1) DE69312760T2 (en)

Families Citing this family (41)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5490199A (en) * 1992-02-21 1996-02-06 At&T Corp. Non-intrusive network-based analysis of facsimile transmissions
NL9301094A (en) * 1993-06-23 1995-01-16 Nederland Ptt Device for measuring analog telephone signals.
CA2129942C (en) * 1993-09-30 1998-08-25 Steven Todd Kaish Telecommunication network with integrated network-wide automatic call distribution
US5539804A (en) * 1994-02-25 1996-07-23 Ctel Compression Telecommunications Corporation Common channel signalling communication monitoring system
US5752216A (en) * 1994-07-06 1998-05-12 Dimensions International, Inc. Non-intrusive data interface system for air traffic control
US5937004A (en) * 1994-10-13 1999-08-10 Fasulo, Ii; Albert Joseph Apparatus and method for verifying performance of digital processing board of an RF receiver
US5526410A (en) * 1995-01-12 1996-06-11 Texas Instruments Incorporated Method and system for determination of signal/noise ratio of telephone transmission line used for facsimile transmission
US5809110A (en) * 1996-01-31 1998-09-15 Dialogic Corporation Technique of controlling telephone calls
US6023470A (en) * 1996-05-17 2000-02-08 Lee; Warren S. Point of presence (POP) for digital facsimile network with virtual POPs used to communicate with other networks
US5739919A (en) * 1996-05-17 1998-04-14 Nko, Inc. Point of presence (POP) for digital facsimile network
US5815669A (en) * 1996-05-17 1998-09-29 Nko, Inc. Method of routing a data transmission
US5999274A (en) * 1996-05-17 1999-12-07 Nko, Inc. Apparatus and method for transmitting facsimile data
US5828468A (en) * 1996-05-17 1998-10-27 Nko, Inc. Point of presence (POP) for digital facsimile network with spoofing capability to maintain fax session
US5905782A (en) * 1996-05-17 1999-05-18 Nko, Inc. Facsimile jack for selectively routing transmissions through a facsimile network
US6075849A (en) * 1996-05-17 2000-06-13 Nko, Inc. Method of monitoring a data transmission
WO1998000784A1 (en) * 1996-06-28 1998-01-08 Mci Communications Corporation System and method for reporting telecommunication service conditions
US6519723B1 (en) * 1996-09-27 2003-02-11 Applied Digital Access, Inc. Firewall performance monitoring and limited access system
US5917896A (en) * 1996-10-30 1999-06-29 Mci Communications Corporation Bridgeable fax monitor for help desk support
US5940376A (en) * 1997-01-29 1999-08-17 Cabletron Systems, Inc. Method and apparatus to establish a tap-point in a switched network using self-configuring switches having distributed configuration capabilities
US6424627B1 (en) * 1997-02-24 2002-07-23 Metrobility Optical Systems Full-duplex medium tap apparatus and system
US6041042A (en) * 1997-05-27 2000-03-21 Cabletron Systems, Inc. Remote port mirroring system and method thereof
US6445733B1 (en) * 1997-10-03 2002-09-03 Conexant Systems, Inc. Method of and apparatus for performing line characterization in a non-idle mode in a subscriber line communication system
US7126713B1 (en) * 1998-03-17 2006-10-24 Brother Kogyo Kabushiki Kaisha Communication terminal apparatus and method
US6519281B1 (en) * 2000-01-31 2003-02-11 Agilent Technologies, Inc. Jitter measurement
US7085019B2 (en) * 2001-03-27 2006-08-01 Canon Kabushiki Kaisha Facsimile apparatus
US7904454B2 (en) * 2001-07-16 2011-03-08 International Business Machines Corporation Database access security
US7506371B1 (en) 2004-01-22 2009-03-17 Guardium, Inc. System and methods for adaptive behavior based access control
US7690040B2 (en) 2004-03-10 2010-03-30 Enterasys Networks, Inc. Method for network traffic mirroring with data privacy
EP1782293A2 (en) * 2004-08-20 2007-05-09 Enterasys Networks, Inc. System, method and apparatus for traffic mirror setup, service and security in communication networks
US7639383B2 (en) * 2005-05-10 2009-12-29 Oliszewski Michael A Systems and methods for enhanced electronic facsimile communication network and preference selection
WO2006122182A2 (en) * 2005-05-10 2006-11-16 Michael Oliszewski Systems and methods for determining user preferences and/or facsimile device capabilities before call initiation
US7970788B2 (en) * 2005-08-02 2011-06-28 International Business Machines Corporation Selective local database access restriction
US7933923B2 (en) * 2005-11-04 2011-04-26 International Business Machines Corporation Tracking and reconciling database commands
US20070206700A1 (en) * 2006-03-06 2007-09-06 Mediatek Inc. Quadrature Amplitude Modulation receiver and diagnostic method thereof
US20080123788A1 (en) * 2006-11-29 2008-05-29 Supat Wongwirawat Method and apparatus for detecting and correcting modulated signal impairments
US20080123787A1 (en) * 2006-11-29 2008-05-29 Supat Wongwirawat Method and apparatus for detecting and correcting modulated signal impairments
US8141100B2 (en) 2006-12-20 2012-03-20 International Business Machines Corporation Identifying attribute propagation for multi-tier processing
US8495367B2 (en) 2007-02-22 2013-07-23 International Business Machines Corporation Nondestructive interception of secure data in transit
US8261326B2 (en) 2008-04-25 2012-09-04 International Business Machines Corporation Network intrusion blocking security overlay
US9477537B2 (en) 2010-12-13 2016-10-25 Microsoft Technology Licensing, Llc Reactive coincidence
WO2014063116A2 (en) * 2012-10-18 2014-04-24 Ikanos Communications, Inc. Method and apparatus for cancelling impulse noise in dsl systems

Family Cites Families (16)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4484030A (en) * 1982-04-19 1984-11-20 At&T Bell Laboratories Method and apparatus for identifying faulty communications circuits
JPH0722324B2 (en) * 1983-08-30 1995-03-08 富士通株式会社 Data transmission method
US4698841A (en) * 1985-08-02 1987-10-06 Gte Laboratories, Incorporated Methods of establishing and terminating connections in a distributed-control burst switching communications system
JPS6251326A (en) * 1985-08-30 1987-03-06 Oki Electric Ind Co Ltd Retransmission control system in data communication
DE3765947D1 (en) * 1986-01-18 1990-12-13 Hewlett Packard Ltd NON-INFLUENCY ANALYZER FOR DETECTING CHANNEL FAULTS.
US4815136A (en) * 1986-11-06 1989-03-21 American Telephone And Telegraph Company Voiceband signal classification
US4815137A (en) * 1986-11-06 1989-03-21 American Telephone And Telegraph Company Voiceband signal classification
JPH0681210B2 (en) * 1987-09-28 1994-10-12 シャープ株式会社 Facsimile line monitor system
EP0332286A3 (en) * 1988-03-09 1990-02-28 Hewlett-Packard Company Programmable protocol analyzer
US4994926C1 (en) * 1988-09-22 2001-07-03 Audiofax Ip L L C Facsimile telecommunications system and method
US4979211A (en) * 1988-11-16 1990-12-18 At&T Bell Laboratories Classifier for high speed voiceband digital data modem signals
JPH02153652A (en) * 1988-12-06 1990-06-13 Ricoh Co Ltd Control method for facsimile equipment
US4969184A (en) * 1989-02-02 1990-11-06 Alphanet Technology Corporation Data transmission arrangement
US4965676A (en) * 1989-02-28 1990-10-23 Ricoh Corporation And Ricoh Company, Ltd. Facsimile remote diagnostic system
US5163088A (en) * 1991-03-06 1992-11-10 Locascio Peter Facsimile security system
US5166977A (en) * 1991-05-31 1992-11-24 Encrypto, Inc. Protocol converter for a secure fax transmission system

Also Published As

Publication number Publication date
JP2622063B2 (en) 1997-06-18
JPH06105112A (en) 1994-04-15
AU3218693A (en) 1993-08-26
US5299257A (en) 1994-03-29
EP0557055A1 (en) 1993-08-25
AU646988B2 (en) 1994-03-10
DE69312760D1 (en) 1997-09-11
EP0557055B1 (en) 1997-08-06
CA2084105A1 (en) 1993-08-22
DE69312760T2 (en) 1998-02-05

Similar Documents

Publication Publication Date Title
CA2084105C (en) Subscriber initiated non-intrusive network-based analysis of facsimile transmissions
CA2084146C (en) Non-intrusive network-based analysis of facsimile transmissions
CA2523482C (en) Telecommunication diagnostic information management
CA2241157C (en) Means for and methods of real time detection of fraud, lack of credit, and the like, in the ss#7 system network
EP0774876B1 (en) Cell testing in cellular telecommunication network
US5166925A (en) T carrier network test apparatus and method
JPH01233856A (en) Facsimile equipment
CA2481203A1 (en) Method and apparatus for measuring communication market statistics
US5905785A (en) Detecting high usage telecommunications lines
US5396342A (en) Automatic remote facsimile testing system
US7050549B2 (en) Real time call trace capable of use with multiple elements
CA2417710C (en) Telecommunications systems
US6304655B1 (en) Signal capture for telecommunications switch
CN100461680C (en) Display system and method for incoming telephone number for digital program controlled user exchanger
US6480576B1 (en) Method and device for detecting path faults in a network
US5426647A (en) Data communication apparatus detecting interrupt requests through data error detection
EP0824294A2 (en) Detecting high usage telecommunications lines
EP1005210A2 (en) Method and system for transmitting messages to subscribers during the set-up stage of incoming telephone calls
Bertocco et al. An in-service measurement system for the quality analysis of facsimile transmissions
Hartanto QualiView: an in-service, non-intrusive monitoring device (INMD) for call quality performance monitoring
JPH04280547A (en) Misdialing detector
Murthy et al. Elucidation of an ISDN performance management model
JPH04280561A (en) Erroneous dialing deciding device
Seshadri et al. Performance assessment of AT&T international switched voice services
JPS6173457A (en) Testing device of modem

Legal Events

Date Code Title Description
EEER Examination request
MKLA Lapsed