CN100440819C - Network voice conversation detecting flow generation method based on conversation model - Google Patents

Network voice conversation detecting flow generation method based on conversation model Download PDF

Info

Publication number
CN100440819C
CN100440819C CNB2007100994074A CN200710099407A CN100440819C CN 100440819 C CN100440819 C CN 100440819C CN B2007100994074 A CNB2007100994074 A CN B2007100994074A CN 200710099407 A CN200710099407 A CN 200710099407A CN 100440819 C CN100440819 C CN 100440819C
Authority
CN
China
Prior art keywords
state
telephone user
test
time
current
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.)
Active
Application number
CNB2007100994074A
Other languages
Chinese (zh)
Other versions
CN101051960A (en
Inventor
尹霞
王之梁
施新刚
吉利
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.)
Tsinghua University
Original Assignee
Tsinghua University
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 Tsinghua University filed Critical Tsinghua University
Priority to CNB2007100994074A priority Critical patent/CN100440819C/en
Publication of CN101051960A publication Critical patent/CN101051960A/en
Application granted granted Critical
Publication of CN100440819C publication Critical patent/CN100440819C/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Abstract

The invention features the following: using a calling model to simulate the call interaction between both calling sides; calculating the state of call process at each time point; according to the sound activation detection technology, deciding if said state generates the traffic of network voice call; at the time point of generating the traffic, recording the offset between the time point and the initial time, the message sequence number and the size of the message corresponding to the code to form a traffic sequence; using the interaction between two test ends to determine the start time and the message actually-sending time, and finally generating the network voice call test traffic based on calling model.

Description

Voice-over-net speaking test flow generation method based on the conversation model
Technical field
The invention belongs to Internet technical field, relate in particular to VOIP test traffic generation technique field.
Background technology
VOIP (Voice over IP) promptly carries out voice call by IP network, has become one of important application of the Internet, is Virtual network operator and the extensively employing of institute of large enterprise.For the performance test of VOIP mainly by access point deployment test point in on-premise network, generate the VOIP test traffic and transmit in test point, pass judgment on the performance of VOIP in network by analyzing indexs such as packet loss, delay, shake, audio distortion at receiving terminal.The performance test of VOIP has the meaning of two aspects: before actual deployment VOIP, determine the tenability of on-premise network for VOIP by testing, thereby determine whether the scale of disposing and disposing; After actual deployment VOIP, by the performance of regular test monitoring VOIP, thereby determine whether to reach re-set target, whether network needs increase-volume, whether has fault or the like.Therefore, the test of VOIP has great importance for the smooth deployment and the use of VOIP technology.
It is important part in the whole VOIP measuring technology that the VOIP test traffic generates, and its target is to generate by certain algorithm to be similar to the discharge characteristic that actual VOIP uses.At present, generation comprises two kinds of methods for the VOIP test traffic: use fixedly message flow (CBR) and the actual audio file coding of use.Use fixedly that G.729 message flow waits the coded system of cbr (constant bit rate) with reference to G.711, the fixed length UDP message by fixed intervals is as the VOIP flow, and flow Core Generators such as D-ITG have adopted this method.Fixedly the problem of message flow has been to ignore active (Voice Activity Detection) technology of the influence survey to(for) the VOIP flow of sound.Sound enlivens Detection Techniques and is meant by the detection of audio signal fluctuation being judged whether the telephone user is in state in a minute, does not produce flow under the situation that the telephone user does not speak, thereby avoids not having the flow of talking state, saves the network bandwidth.Studies show that the people is in time of about 50% in communication process answers and dumb state, therefore sound enlivens Detection Techniques for the influence of VOIP flow clearly, and be widely used in multiple coded system and actual use such as products such as Cisco IP phone.The flow that the support that shortage is enlivened Detection Techniques to sound causes fixing message flow to produce is untrue, has exaggerated the occupancy of VOIP stream for bandwidth, tenability that can not the Validity Test network and the actual performance of VOIP.Except fixing message flow, the another kind of flow generation method that is used at present is to adopt actual audio file coding, comprises Spirent, and the tester of companies such as Agilent is all supported this kind flow generation method.It generates flow by using actual .wav file to encode, and can use sound to enliven the time period shielding flow that Detection Techniques do not have conversation hereof.The problem of this generation method is that the .wav file that uses can only be used for unidirectional VOIP test, promptly unidirectional voice transfer, and can not test two interactive voices between the correspondent, therefore the flow with real VOIP application is not inconsistent.Use the mode of audio file coding can not test the VOIP performance of two correspondent simultaneously, the audio file of its use is not inconsistent with the two-way call audio frequency of reality yet, and there is error in the discharge characteristic that produces after using sound to enliven Detection Techniques.
At above-mentioned deficiency, the present invention proposes a kind of VOIP test traffic generation method based on the conversation model.Telephone relation model by reference ITU-T proposition, designed the conversation model that is applicable to VOIP, and according to this model generation VOIP flow, the behavior of both having considered the people makes the flow of generation have authenticity because sound enlivens Detection Techniques to the influence that the VOIP flow produces; Generate the flow sequence of two correspondent again by state transition simultaneously, make the flow of generation have interactivity, can test the quality of two-way call simultaneously.Therefore, the deficiency of existing VOIP test traffic generation method that the present invention is perfect uses test traffic of the present invention that more genuine and believable test result can be provided.
Summary of the invention
The object of the present invention is to provide a kind of VOIP test traffic generation method based on the conversation model.
The thinking of method proposed by the invention is: use the reciprocal process of VOIP conversation model emulation both call sides, calculate at the residing state of each time point communication process; Enliven Detection Techniques according to sound, judge whether generate the VOIP flow at this state; Generate the time point of the state of flow for needs, write down the side-play amount of this time point apart from zero-time, the message size of test serial number and corresponding coding forms the flow sequence; Determine to test the time started by the mutual of two test leads, determine the actual transmitting time of message, the final VOIP test traffic that generates based on the conversation model.
The invention is characterized in:
Described method realizes successively according to the following steps:
(1) determines VOIP conversation Model parameter, as shown in Figure 1, VOIP conversation model comprises one of four states, represent that respectively telephone user A is in talking state, telephone user B is in talking state, and no talking state and both sides are in talking state simultaneously, move by probabilistic relation between the state, owing to meet geometric distributions in the time of staying of each state, and there is certain symmetry in model, can calculate probable value by following formula:
P 11＝P 22＝1-T interval/T talk-avg
P 14＝P 24＝(1-P 11)×K talk-double
P 13＝P 23＝1-P 11-P 14
P 23＝1-T iterval/T stop-avg
P 44＝1-T interval/T double-avg
P 31＝P 32＝(1-P 33)/2
P 41＝P 42＝(1-P 44)/2
Wherein, T IntervalExpression testing audio coding sends the interval of message, T Talk-avgBe illustrated in the average time of state 1 or state 2, T Stop-avgBe illustrated in the average time of state 3, T Double-avgBe illustrated in the average time of state 4, can set according to the measured result of user behavior, also can use the empirical value 1.004 seconds of ITU-T suggestion, 0.508 second and 0.228 second is as default value, K Talk-doubleExpression is in talking state from a side and is in the transition factor of talking state together to both sides, can set according to the measured result of user behavior, and the empirical value 0.6 that also can use the ITU-T suggestion is as default value, and parameter and formula by as above calculate each probable value;
(2) set the time span that VOIP tests, and current state is set at state 3, and will the current time be set at 0, telephone user A test serial number is set at 0, current talking people B test serial number is set at 0, creates telephone user A and two sequence of message files of telephone user B;
(3) handle accordingly according to current state:
● current state is a state 1: add record＜telephone user A test serial number in the sequence of message file of telephone user A, current time, coding message size 〉, the flow of telephone user B is because sound enlivens Detection Techniques and conductively-closed;
● current state is a state 2: add record＜telephone user B test serial number in the sequence of message file of telephone user B, current time, coding message size 〉, the flow of telephone user A is because sound enlivens Detection Techniques and conductively-closed;
● current state is a state 3: the flow of telephone user A and telephone user B is because sound enlivens Detection Techniques and conductively-closed;
● current state is a state 4: add record＜telephone user A test serial number in the sequence of message file of telephone user A, current time, the coding message size 〉, in the sequence of message file of telephone user B, add record＜telephone user B test serial number, current time, the coding message size 〉;
(4) current time increases progressively the interval that coding sends message, i.e. T Interval, judge that whether the new current time surpass the time span of VOIP test, then change step (6) if surpass, if in the time span of VOIP test, then change step (5);
(5) according to current state and the new state of VOIP conversation model state transition probability calculation, new state value is set at current state, and changes step (3);
(6) close two sequence of message files of telephone user A and telephone user B, and store file into two test terminal main frames respectively, the main frame of the sequence of message file of storage telephone user A is called host A, and the main frame of the sequence of message file of storage telephone user B is called host B;
(7) host A calculates the test time started, and this time can be set at current time and a certain increment sum, and should the time send to host B by the tcp data connection;
(8) host A is with each the bar record＜telephone user A test serial number in the sequence of message file of telephone user A, current time, coded file size〉change to＜telephone user A test serial number, the current time+the test time started, the coding message size 〉, and preserve file;
(9) host B is with each the bar record＜telephone user B test serial number in the sequence of message file of telephone user B, current time, coded file size〉change to＜telephone user B test serial number, the current time+the test time started, the coding message size 〉, and preserve file;
(10) host A and host B are set up the UDP session, and according to the order of the record in the sequence of message file, in time that each record indicates with the coding message size as the network application layer size of data, encapsulation realtime transmission protocol RTP heading, send by the UDP session, thereby realization is based on the VOIP test traffic generation method of conversation model.
VOIP test traffic generating algorithm proposed by the invention is because based on the conversation model, can be created on the VOIP test traffic of using sound to enliven both call sides under the Detection Techniques simultaneously, the authenticity and the interactivity of VOIP test traffic have been taken into full account, can be used to realize simultaneously the VOIP test of twocouese, for VOIP test job more accurately and reliably provides help.
Description of drawings
Fig. 1 .VOIP model of conversing.
Fig. 2 .VOIP test traffic deployed environment.
Fig. 3 .VOIP test traffic product process figure.
Fig. 4 algorithm generates the flow experimental result: the square points mark line is a state 1, and the Diamond spot mark line is a state 2, and triangle form point mark line is a state 3, and x point mark line is a state 4.
Embodiment
Concrete deployed environment used in the present invention as shown in Figure 2.Wherein, the VOIP test traffic generates main frame and realizes VOIP conversation model by software program, finishes the sequence of message file of telephone user A and B to the order of step (6) according to step (1).Test terminal host A and test terminal host B are stored the telephone user A of VOIP test traffic generation main frame generation and the sequence of message file of B respectively, finally finish the generation of VOIP test traffic to the order of step (10) according to step (7) by TCP session and UDP session.In addition,, can on test terminal host A and test terminal host B, install global positioning system additional in order to improve the precision of test, the clock of synchronous two main frames, thus make that the test zero-time is more consistent.
Concrete scheme flow process of the present invention as shown in Figure 3.The present invention is applicable to that the VOIP test traffic under the multiple audio coding condition generates, and only need get final product the relevant parameter adjustment that relates in the flow process according to different audio codings.Below, G.729 specify for the generation method of this kind coding in conjunction with audio frequency based on the VOIP test traffic of conversation model.
According to step (1) computing mode transition probability value.Wherein, T IntervalBe set at 0.02 second at interval, parameter T according to G.729 message Talk-avg, T Sotp-avg, T Double-avgAnd K Talk-doubleCan calculate by obtaining actual user's communication data, or (1) default value that provides that directly uses step, under situation about setting according to default value, it is as follows to calculate each probable value according to the formula of step (1):
P 11＝P 22＝0.980
P 14＝P 24＝0.012
P 13＝P 23＝0.008
P 33＝0.960
P 44＝0.912
P 31＝P 32＝0.020
P 41＝P 42＝0.044
State according to step (2) initialization conversation model is a state 3, and the initialization current time is 0, and the test serial number of initialization telephone user A and telephone user B is 0, and creates the sequence of message file of telephone user A and telephone user B.Because use the conversation model can repeatedly produce the VOIP test traffic of different time length, different qualities, for avoiding the file conflict, suggestion adopts " time on date telephone user sign " as filename.
To write down in the file of creating according to form write step (2) according to the present located state according to step (3), write the new row of fashionable establishment at every turn, write record according to character format, guarantee that the last sequence of message file that generates both can use for Test Host, also can read for the tester, wherein test serial number and current time are according to the value corresponding record, the coding message size is then set according to the requirement of variety classes coding, corresponding to G.729, it is under the situation that is set at 0.02 second at interval, the size of each message is 20 bytes, therefore record 20 when record.
Increase progressively the current time according to step (4), for G.729 increase progressively 0.02 second at every turn, compare in the time span that increases progressively back and testing setup, if the current time surpasses the time span of testing setup according to its coding regulation, the flow sequence of message that generates the enough time is described, then changes step (6); If the current time does not surpass the time span of testing setup, illustrate still and need write the inbound traffics message according to the state of new current time, then change step (5) and calculate new current state.
Calculate new current state according to step (5), new current state is by preceding state and probability decision, for example current state is a state 1, random number between program produces one 0 to 1, probable value according to step (1) calculating, if this random number is less than 0.980, then new state still is a state 1; If random number is greater than 0.980, less than 0.988 (0.980+0.008), then new state becomes state 3; If random number is greater than 0.988, then new state becomes state 4, after this kind method generation new state, then changes step (3).
Close the sequence of message file according to step (6), the time of calculating this moment has arrived the time that test needs, show that the sequence of message that writes down in the file finishes, therefore close the sequence of message file of telephone user A and B, above work all goes out the software program of realizing on VOIP test traffic generation main frame finishes according to flow process, and need store the sequence of message file of telephone user A and B respectively into test terminal host A and test terminal host B this moment.This stores optional majority kind mode, both can pass through HTTP, agreements such as FTP are finished, and also can connect by set up TCP between test terminal main frame and VOIP test traffic generation main frame, are sent to the test terminal main frame automatically after the VOIP test traffic generates the Host Shutdown file.
Calculate the testing time according to step (7), the work that begins of step is from then on finished by the software program of test terminal host A and test terminal host B.At first between test terminal host A and test terminal host B, set up TCP and connect, in order to control information transmission by software program; Determine the time started of test then at the test terminal host A, this time is at first obtained current system time by calling the kernel order, and should increase a certain increment the time, and for example 1 minute, as the time started of reality test; At last, for test terminal host A and test terminal host B have the common test time started, the software program of test terminal host A connects by TCP this testing time is sent to end host B.
Revise the sequence of message file of the telephone user A of test terminal host A storage according to step (8), open the sequence of message file of telephone user A by program software, read each line item, and the numeric field that will wherein represent the current time change to test time started that current time and step (7) set and, again write record, close file after all records are all revised, the sequence of message file upgraded according to the test time started and finished this moment, can write down corresponding time transmitted traffic according to each in test beginning back.
Revise the sequence of message file of the telephone user B of test terminal host B storage according to step (9), open the sequence of message file of telephone user B by program software, read each line item, and the numeric field that will wherein represent the current time change to test time started that current time and step (7) set and, again write record, close file after all records are all revised, the sequence of message file upgraded according to the test time started and finished this moment, can write down corresponding time transmitted traffic according to each in test beginning back.
Carry out the actual transmissions of test traffic according to step (10), by step (8) and step (9), the sequence of message file of telephone user A and B upgrades according to the testing time of reality, at this moment, the UDP session that test terminal host A and Test Host B set up test data according to separately address and port, read each record of sequence of message file separately afterwards in order, with the record the coding message size as load length, because the actual content of load can not influence the transmission course of message in network, therefore can generate at random, time value according to record after encapsulation RTP heading sends to the other side by the UDP session, thereby finishes the generation work of test traffic.
More than according to feature of the present invention, at specific G.729 coded system the embodiment of VOIP test traffic generation method based on the conversation model has been described, Fig. 4 has shown the final mean residence time that generates flow at each state under the different testing time situations of setting, when the testing time of setting greater than 200 seconds after, discharge characteristic tends towards stability, with each state error average time of setting according to user behavior less than 10%, explanation is under the situation that guarantees certain data volume, VOIP test traffic generating algorithm based on the conversation model can be similar to the emulation user behavior, has verified the validity of this algorithm.For the coding of other kinds, only need make amendment to the numerical value in the corresponding steps in the execution mode and promptly can use according to the parameter of specific coding, guaranteed universality of the present invention.The flow that generates by the present invention meets the standard of the model of converse, possesses authenticity and interactivity simultaneously, can provide data more accurately for VOIP tests.
This shows that the present invention has reached intended purposes.

Claims (1)

  1. Based on the conversation model voice-over-net speaking test flow generation method, it is characterized in that this method generates between main frame, test terminal host A and the test terminal host B at voice-over-net speaking test flow and alternatively realizes as follows successively: step (1): generate at described voice-over-net speaking test flow and determine VOIP voice-over-net conversation Model parameter on the main frame:
    Described VOIP voice-over-net conversation model comprises one of four states, expression respectively:
    The telephone user A of test terminal host A is in talking state and is called state 1,
    The telephone user B of test terminal host B is in talking state and is called state 2,
    No talking state is called state 3,
    A, B both sides telephone user are in talking state simultaneously and are called state 4,
    Move by probabilistic relation between each state, wherein:
    P 11＝P 22＝1-T interval/T talk-avg
    P 14＝P 24＝(1-P 11)×K talk-double
    P 13＝P 23＝1-P 11-P 14
    P 33＝1-T interval/T stop-avg
    P 44＝1-T interval/T double-avg
    P 31＝P 32＝(1-P 33)/2
    P 41＝P＝ 42(1-P 44)/2
    Wherein, P represents probability, and following target different digital is successively represented to transfer to back one state by previous state, and following target same numbers represents to keep the present located state,
    Tinterval: the expression audio coding sends the interval of message,
    Ttalk-avg: be illustrated in the average time of state 1 or state 2,
    Tstop-avg: be illustrated in the average time of state 3,
    Tdouble-avg: be illustrated in the average time of state 4,
    Ttalk-avg is 1.004 seconds, or by the user behavior actual measurement,
    Tstop-avg is 0.508 second, or by the user behavior actual measurement,
    Tdouble-avg is 0.228 second, or by the user behavior actual measurement,
    Ktalk-double: expression is in talking state from a side and is in the transition factor of talking state simultaneously to both sides, is 0.6 or is surveyed by user behavior;
    Step (2): on the described voice-over-net speaking test of step (1) flow generation main frame, set:
    The time span of voice-over-net speaking test, and current state is set at 3, and the current time is set at 0, and telephone user A and telephone user B test serial number separately are set at 0, and create telephone user A and telephone user B totally two sequence of message files, this document " date _ time _ telephone user's sign " by name;
    Step (3): do different following processing at different current states:
    ● current state is a state 1: in the sequence of message file of telephone user A, add record＜telephone user A test serial number, when
    Before the time, the coding message size 〉, telephone user B this moment do not produce voice, so sound enliven Detection Techniques can not the perceptual speech signal, so telephone user B does not produce flow;
    ● current state is a state 2: add record＜telephone user B test serial number in the sequence of message file of telephone user B, current time, the coding message size 〉, telephone user A does not produce voice at this moment, therefore sound enliven Detection Techniques can not the perceptual speech signal, so telephone user A does not produce flow;
    ● current state is a state 3: telephone user A and telephone user B all do not produce voice this moment, thus sound enliven Detection Techniques can not the perceptual speech signal, so telephone user A and telephone user B all do not produce flow;
    ● current state is a state 4: add record＜telephone user A test serial number in the sequence of message file of telephone user A, current time, the coding message size 〉, in the sequence of message file of telephone user B, add record＜telephone user B test serial number, current time, the coding message size 〉;
    Described coding message size is set according to the requirement of variety classes coding;
    Step (4): increase progressively the interval T interval that sends message based on the current time, judge whether the current time after increasing progressively surpasses the testing time length of voice-over-net conversation, if surpass, then change step (6), if do not surpass execution in step (5);
    Step (5): according to current state and the new state of voice-over-net conversation model state transition probability calculation, new state value is set at current state, changes step (3), its concrete steps are as follows:
    Step (5.1): the random number between program produces 0 to 1;
    Step (5.2): (1) migration probability of calculating set by step under described this random number of comparison step (5.1) and the current state:
    If this random number is less than the migration probability under the current state, then current state is constant, changes step (3);
    If this random number is greater than the migration probability under the current state, then the probability of all the next states that can move to from current state, find an interval that can make the migration probability of migration probability that this random number drops on current state and selected next state, again selected next state as new state, and new state value is set at current state, change step (3);
    Step (6): voice-over-net speaking test flow generation Host Shutdown telephone user A and telephone user B be totally two sequence of message files, and two files are deposited in test terminal host A and test terminal host B respectively;
    Step (7): the test terminal host A calculates the test time started, and this time set is an increment sum of current time and setting, and the current time of new settings is sent to the terminal test host B by the TCP connection;
    Step (8): the current time of current time for newly establishing of the sequence of message file of test terminal host A change oneself;
    Step (9): the test terminal host B changes the current time of current time for newly establishing of the sequence of message file of oneself after receiving the current time of newly establishing that step (7) is sent;
    Step (10): set up the UDP session between test terminal host A and the test terminal host B, and according to the order of the record in the sequence of message file, in time that each record indicates the coding message size as the network application layer size of data, encapsulation realtime transmission protocol RTP head, send by the UDP session, thereby realization is based on the voice-over-net speaking test flow generation method of conversation model.
CNB2007100994074A 2007-05-18 2007-05-18 Network voice conversation detecting flow generation method based on conversation model Active CN100440819C (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CNB2007100994074A CN100440819C (en) 2007-05-18 2007-05-18 Network voice conversation detecting flow generation method based on conversation model

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CNB2007100994074A CN100440819C (en) 2007-05-18 2007-05-18 Network voice conversation detecting flow generation method based on conversation model

Publications (2)

Publication Number Publication Date
CN101051960A CN101051960A (en) 2007-10-10
CN100440819C true CN100440819C (en) 2008-12-03

Family

ID=38783166

Family Applications (1)

Application Number Title Priority Date Filing Date
CNB2007100994074A Active CN100440819C (en) 2007-05-18 2007-05-18 Network voice conversation detecting flow generation method based on conversation model

Country Status (1)

Country Link
CN (1) CN100440819C (en)

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101547259B (en) * 2009-04-30 2012-08-08 华东师范大学 VoIP test method based on analog data flow
EP2254277B1 (en) * 2009-05-20 2015-03-18 Accenture Global Services Limited Control management of voice-over IP parameters
CN104168155B (en) * 2014-07-31 2019-02-15 南京赛拜诺信息技术有限公司 Application traffic generation method and device
CN109921947B (en) * 2019-03-26 2022-02-11 东软集团股份有限公司 Network flow simulation method, device, equipment and network equipment test system
CN110460732B (en) * 2019-09-24 2021-05-14 腾讯科技（深圳）有限公司 Network quality monitoring method and device and communication server

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20040120312A1 (en) * 2002-12-16 2004-06-24 Eung-Moon Yeom Method for call processing and link test in a VoIP gateway and system thereof
US20050135391A1 (en) * 2003-12-17 2005-06-23 Jung-Sic Sung Apparatus and method for providing VoIP service
CN1750488A (en) * 2005-09-22 2006-03-22 广东省电信有限公司研究院 Monitor system and its monitor method for internet network telephone
CN1794765A (en) * 2006-01-06 2006-06-28 中国移动通信集团公司 Test method VoIP language technology capacity of access network
CN1913456A (en) * 2006-08-30 2007-02-14 中国科学院计算技术研究所 Method of identifing VOIP flow based on SIP protocol process performance

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20040120312A1 (en) * 2002-12-16 2004-06-24 Eung-Moon Yeom Method for call processing and link test in a VoIP gateway and system thereof
US20050135391A1 (en) * 2003-12-17 2005-06-23 Jung-Sic Sung Apparatus and method for providing VoIP service
CN1750488A (en) * 2005-09-22 2006-03-22 广东省电信有限公司研究院 Monitor system and its monitor method for internet network telephone
CN1794765A (en) * 2006-01-06 2006-06-28 中国移动通信集团公司 Test method VoIP language technology capacity of access network
CN1913456A (en) * 2006-08-30 2007-02-14 中国科学院计算技术研究所 Method of identifing VOIP flow based on SIP protocol process performance

Also Published As

Publication number Publication date
CN101051960A (en) 2007-10-10

Similar Documents

Publication Publication Date Title
Jelassi et al. Quality of experience of VoIP service: A survey of assessment approaches and open issues
US6885987B2 (en) Method and apparatus for encoding and decoding pause information
JP4504429B2 (en) Method and apparatus for managing media latency of voice over internet protocol between terminals
JP4964859B2 (en) Capture and reproduce network status
CA2560425C (en) Determining quality of voice calls over a packet network
US10044875B2 (en) Codec selection and usage for improved VoIP call quality
US20020015387A1 (en) Voice traffic packet capture and analysis tool for a data network
US20040034492A1 (en) Passive system and method for measuring and monitoring the quality of service in a communications network
US20040165570A1 (en) Call routing method in VoIP based on prediction MOS value
US20020016937A1 (en) Method and apparatus for utilizing a network processor as part of a test system
CA2359991A1 (en) Methods, systems and computer program products for packetized voice network evaluation
WO2004086741A1 (en) Talking quality evaluation system and device for evaluating talking quality
US20210150393A1 (en) System and Method for Machine Learning Based QoE Prediction of Voice/Video Services in Wireless Networks
CN100440819C (en) Network voice conversation detecting flow generation method based on conversation model
US7274714B2 (en) Emulating of information flow
CN103841275A (en) Interactive audio experience quality evaluation platform and method based on QoS
US20130208614A1 (en) Method of providing voice over ip at predefined qos levels
KR100738162B1 (en) Method for measuring interactive speech quality in VoIP network
US20020016708A1 (en) Method and apparatus for utilizing a network processor as part of a test system
Ortega et al. Evaluation of the voice quality and QoS in real calls using different voice over IP codecs
US20090257455A1 (en) Method and apparatus for synchronizing timing of signal packets
Zhang et al. Perceived QoS assessment for Voip networks
Ren et al. Assessment of effects of different language in VOIP
CN100488216C (en) Testing method and tester for IP telephone sound quality
Sun et al. Voip quality of experience (QoE)

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant