JP4849467B2 - Connectivity evaluation system - Google Patents

Description

  The present invention relates to a connectivity evaluation system that evaluates connectivity at a receiving terminal of media transmitted via a network, and in particular, provides a receiving terminal with quality-degraded media that simulates quality degradation in a network in a test environment. The present invention relates to a connectivity evaluation system for evaluating sex.

  In communication via a wide area network such as the Internet or a wireless network, unlike communication via a stable network such as a wired LAN, quality degradation such as packet loss and jitter tends to occur on the network. Therefore, when providing a service for transmitting media such as moving images and voices via such a network, it is necessary to evaluate in advance the practical effects of packet loss and jitter generated via the network on the media.

In Patent Document 1, in a state where two communication terminals for verifying mutual communication are actually in communication, transmission data and reception data of each communication terminal are collected, and a transmission / reception sequence of the communication terminals is analyzed. Techniques for performing communication analysis and testing are disclosed.
Japanese Patent Laid-Open No. 2003-177799

  Even if the frequency of occurrence of such quality degradation is the same, the connectivity at the receiving terminal is different depending on whether it occurs in a temporally distributed manner or temporarily in a concentrated manner. The effect on is completely different. Therefore, if the connectivity of the test terminal is to be evaluated, it is not a test that assumes accidentally observed quality degradation, or a test that assumes quality degradation that cannot occur in the test environment network. It is desirable to perform tests by simulating various modes and worst conditions of quality degradation that can occur.

  However, in the above-described conventional technology, it was not possible to perform tests by simulating various aspects and worst conditions of quality degradation that could occur in a test environment network.

  An object of the present invention is to solve the above-described problems of the prior art and provide a connectivity evaluation system capable of performing tests that simulate various aspects and worst conditions of quality degradation that may occur in a test environment network.

  In order to achieve the above-described object, the present invention provides a connectivity evaluation system for evaluating connectivity at a receiving terminal of media transmitted via a network, in which the quality of received media transmitted via an actual network to be evaluated is determined. Quality measuring means for measuring, statistic calculating means for calculating a statistic relating to quality degradation that may occur in the real network based on the measurement result, processing test media based on the statistic, It is characterized by including deterioration processing means for generating quality deteriorated media simulating quality deterioration, and transmission means for transmitting the quality deteriorated media to the receiving terminal.

  According to the present invention, it becomes possible to conduct tests that simulate various aspects and worst conditions of quality degradation that can occur in a network in a test environment, so that more accurate connectivity evaluation can be performed under conditions suitable for actual use. Become.

  DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, the best embodiment of the present invention will be described in detail with reference to the drawings. FIG. 1 is a block diagram showing a configuration of a main part of a connectivity evaluation system according to an embodiment of the present invention, and a quality measurement unit 1 that measures the quality of a reception medium via a real network, and a quality measurement Statistical processing unit 2 that statistically processes the results to obtain statistics of quality degradation, and based on these statistics, the quality of the test media is degraded, and quality degradation media that simulates quality degradation is generated and tested The connectivity test unit 3 applied to the terminal has a main configuration.

  In the quality measuring unit 1, the encoding unit 11 encodes media such as a moving image file, a still image file, or an audio file. The network sending unit 12 sends the encoded media to a real network 13 such as the Internet or a wireless network. The receiving unit 14 receives media addressed to itself from the real network 13 in units of packets. The quality measuring unit 15 measures packet loss and jitter based on the received packet.

  The statistical processing unit 2 statistically processes the measurement results of packet loss and jitter, and calculates the average value, maximum value, minimum value, probability value based on normal distribution, etc. of packet loss and jitter as quality degradation statistics. To the connectivity test unit 3.

  In the connectivity test unit 3, the media information database (DB) 31 stores test media such as moving image files, still image files, and audio files. The simulated parameter storage unit 32 stores quality degradation statistics provided from the statistical processing unit 2. The quality processing unit 33 reads a predetermined statistic from the simulation parameter storage unit 32 based on a separately designated test mode, and performs quality degradation processing on the test media based on the statistic to simulate the quality. To deteriorate.

  The network transmission unit 34 encodes a test medium whose quality is pseudo-degraded (hereinafter referred to as a quality-degraded medium) and transmits the encoded test medium to a network 35 that does not substantially deteriorate, such as a wired LAN. The terminal under test 36 receives the quality deteriorated media addressed to itself from the network 35 and reproduces it.

  FIG. 2 is a diagram showing a list of test modes in the present embodiment, where “as is” and “simulation” are prepared as “test types”. “As is” is a test mode in which a quality-degraded medium in which the packet loss and jitter in the actual network 13 observed by the quality measuring unit 1 are reproduced as it is is generated and transmitted.

  On the other hand, “simulation” of “test type” assumes various quality degradation patterns that can occur in the actual network 13 based on the packet loss and jitter observed in the quality measuring unit 1. This is a test mode unique to the present invention in which pattern quality deterioration is applied to a test medium and transmitted.

  Regarding “simulation” of the “test type”, “loss” and “jitter” are prepared as “simulation type”. “Loss” is a test mode that simulates quality degradation by dropping media packets, and “jitter” is a test mode that simulates quality degradation by delaying the packet transmission timing.

  In the “loss” of the “simulation type”, “period” and “random” are prepared as “simulation intervals” for specifying an interval for simulating quality degradation, that is, an interval for dropping packets. In the “cycle” of “simulation interval”, three modes of “none”, “average”, and “random” are prepared for “continuity” that specifies the number of consecutive lost packets. Are provided with four modes of “average value”, “random”, “worst value”, and “normal distribution” as “simulation amounts” for specifying the number of packets to be dropped. Similarly, four modes of “average value”, “random”, “worst value”, and “normal distribution” are prepared as “simulation amount” in “random” of the “simulation interval”.

  The “average value” mode of the “simulated amount” is a mode in which the average value Yave of the number of lost packets Yi in the real network 13 observed for each unit time Ti by the quality measuring unit 1 is set as the number of dropped packets. . In the “random” mode of the “simulated amount”, the number of dropped packets within the range of the maximum value Ymax and the minimum value Ymin of the number of lost packets Yi in the real network 13 observed by the quality measuring unit 1 for each unit time Ti. Is a mode in which is randomly set.

  In the “worst case” mode of the “simulated amount”, the worst value (maximum value) Ymax of the number of lost packets in the real network 13 observed for each unit time Ti by the quality measuring unit 1 is set as the number of dropped packets. Mode. In the “normal distribution” mode of the “simulated amount”, a plurality of distributions of packet loss rates (number / second) in the real network 13 observed for each unit time Ti by the quality measuring unit 1 are acquired, and for each unit time Ti. In this mode, the distribution with the maximum packet loss rate as a sample is applied to the normal distribution to set the number of dropped packets (probability value) Ynd.

  On the other hand, “simulation interval” and “continuity” are not defined for “jitter” of “simulation type”, and “average value”, “random”, “ Four modes of “worst value” and “normal distribution” are prepared.

  Next, a method for generating quality-degraded media in each test mode will be described. In the present embodiment, the state of occurrence of lost packets in the real network 13 observed by the quality measuring unit 1 is analyzed by the statistical processing unit 2 for each predetermined unit period Ti. Then, the average value Xave, the minimum value Xmin, the maximum value Xmax, and the random value Xrnd between Xmin and Xmax are determined for each unit period Ti with respect to the packet drop number X for each unit period Ti.

  Regarding the simulated amount Y, the average value Yave, the minimum value Ymin, the maximum value Ymax of the number of lost packets Yi obtained for each unit period Ti, the run-ram value Yrnd between the Ymin and Ymax, and the probability value Ynd of the normal distribution are Desired.

  Regarding the jitter interval J, an average interval JYave, a minimum interval Jmin, a maximum interval Jmax, a random interval Jrnd between Jmin and Jmax, and a probability value Jnd of a normal distribution are obtained for each unit period Ti.

  Next, the operation of the quality processing unit 33 of the connectivity test unit 3 will be described. The quality processing unit 33 is instructed to test modes of “simulation” / “loss” / “cycle” / “none” for “test type” / “simulation type” / “simulation interval” / “continuity”, respectively. Then, the test media packet is dropped at a rate of 1 per [ai / Y].

  That is, if the “simulation amount” is “average”, one packet is dropped every time [ai / Yave] packets are transmitted. Similarly, each time [ai / Yrnd] packets are sent if the “simulation amount” is “random”, every time [ai / Ymax] packets are sent if “ai / Yrnd” packets are sent. If one packet is “normal distribution”, each time [ai / Ynd] packets are sent, one packet is dropped.

  FIG. 3 is a diagram showing an example of a packet sequence when “average” is instructed as “simulation amount”. If the number of packets ai in the unit period Ti is 24 and the simulation amount Yave is 4, The head packet interval d is six, and packets are dropped one by one from the head.

  FIG. 4 is a diagram showing an example of a packet sequence when “random” is instructed as “simulation amount”, where the number of packets ai in the unit period Ti is 40 and the simulation amount (Yrnd) is 5. For example, as shown in FIG. 9A, the head packet interval is eight, and one packet is dropped from the head one by one. If the number of packets ai in the unit period Ti is 28 and the simulated amount (Yrnd) is 4, the head packet interval is 7 as shown in FIG. 5B, and one packet is dropped from the head. Is done.

  In addition, when “test type” / “simulation type” / “simulation interval” / “continuity” are respectively designated as “simulation” / “loss” / “cycle” / “average” test modes, the test media Are dropped at a rate of 1 per [ai / (Y / Xave)] packets.

  That is, if the “simulated amount” is “average”, Xave packets are dropped every time [ai / (Yave / Xave)] packets are transmitted. Similarly, if the “simulation amount” is “random”, it will be Xave each time [ai / (Ynd / Xave)] packets are sent. If it is “worst value”, [ai / (Ymax / Xave)] Xave packets are dropped each time X packets are sent, and every time [ai / (Ynd / Xave)] packets are sent if they are “normally distributed”.

  FIG. 5 is a diagram showing an example of a packet sequence when “average” is designated as “simulation amount”, where the number of packets ai in the unit period Ti is 24, the number of consecutive drops X is 3, and the simulation amount If there are nine Yaves, the leading packet interval d is eight, and three packets are dropped from the beginning.

  In addition, when “test type” / “simulation type” / “simulation interval” / “continuity” are instructed to each of the “simulation” / “loss” / “random” test modes, ai / (Y / Xrnd)] is dropped at a rate of one.

  That is, if the “simulated amount” is “average”, Xrnd packets are dropped every time [ai / (Yave / Xrnd)] packets are transmitted. Similarly, if the “simulation amount” is “random”, it will be Xrnd each time it sends [ai / (Yrnd / Xrnd)] packets, and if it is “worst value”, it will be [ai / (Ymax / Xrnd)]. Xrnd packets are dropped every time a packet is sent, and Xrnd packets are dropped each time [ai / (Ynd / Xrnd)] packets are sent if “normal distribution” is sent.

  In addition, when “test type” / “simulation type” / “simulation interval” is instructed to “simulation” / “loss” / “random” test mode, the simulation interval d is set to [ai / (Ymin / Xmax )] And [ai / (Ymax / Xmin)] random value drnd, and if "simulation amount" is "average", every time drnd packets are sent, (drnd x Yave) / ai Drop the packet. Similarly, if the “simulated amount” is “random”, every time drnd packets are sent (drnd × Yrnd) / ai, and if the “worst case”, each time drnd packets are sent (drnd If (Yx) / ai, or "normal distribution", each time drnd packets are sent, (drnd × Ynd) / ai packets are dropped.

  On the other hand, if the “test type” / “simulation type” is instructed to be the “simulation” / “jitter” test mode, the transmission of each packet is delayed by Jave if the “simulation amount” is “average”. . Similarly, if “simulated amount” is “random”, it is delayed by Jrnd, if it is “worst value”, it is delayed by Jmax, and if it is “normal distribution”, it is delayed by Jnd.

It is the block diagram which showed the structure of the principal part of the connectivity evaluation system which concerns on this invention. It is the figure which showed the list of test modes. FIG. 6 is a diagram illustrating an example of a packet sequence when “simulated amount” is “average”. FIG. 6 is a diagram showing an example of a packet sequence when “simulated amount” is “random”. FIG. 6 is a diagram illustrating an example of a packet sequence when “simulated amount” is “average”.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1 ... Quality measurement part, 2 ... Statistical processing part, 3 ... Connectivity test part, 11 ... Encoding part, 12 ... Network transmission part, 13 ... Real network, 14 ... Reception part, 15 ... Quality measurement part, 31 ... Media Information database 32 ... Simulation parameter storage unit 33 ... Quality processing unit 34 ... Network transmission unit 35 ... Network 36 ... Terminal under test

Claims (8)

In a connectivity evaluation system for evaluating connectivity at a receiving terminal for media transmitted via a network,
Quality measuring means for measuring the quality of the received media transmitted via the real network to be evaluated;
Statistic calculation means for calculating a statistic regarding quality degradation that may occur in the real network based on the quality measured by the quality measurement means ;
A connection evaluation unit for inputting the statistic calculated by the statistic calculating means;
Deterioration processing means for processing test media based on the input statistics and generating quality degradation media simulating quality degradation by the real network;
A connectivity evaluation system comprising: a connection evaluation unit including transmission means for transmitting the quality-degraded media to a receiving terminal through a network with substantially no deterioration .   The connectivity evaluation system according to claim 1, wherein the statistic calculation unit calculates a statistic of packet loss.   The connectivity evaluation system according to claim 1, wherein the statistic calculation unit calculates a statistic of jitter. The statistic calculation means includes:
Obtain the number of received media packets ai and the number of lost packets Yi for each predetermined unit time Ti,
Statistically processing the number of lost packets Yi to obtain a simulated amount Y of packet loss,
The connectivity evaluation system according to claim 2, wherein the deterioration processing unit drops one packet of the test media at every packet interval d obtained by the following equation.
d = ai / Y   The simulated amount Y of the packet loss is any one of an average value Yave, a random value Yrnd, a worst value Ymax, and a probability value Ynd corresponding to a normal distribution of the number of lost packets Yi. Connectivity evaluation system. The statistic calculation means includes:
For each predetermined unit time Ti, obtain the number of received media packets ai, the number of lost packets Yi, and the number of consecutive lost packets Xi,
Statistically processing the number of lost packets Yi to obtain a simulated amount Y of packet loss,
Statistically processing the number of consecutive lost packets Xi to obtain the number of consecutive drops X,
3. The connectivity evaluation system according to claim 2, wherein the deterioration processing unit drops X packets of the test media every packet interval d determined by the following equation.
d = ai / (Y / X)   The simulated amount Y is one of the average value Yave, the random value Yrnd, the worst value Ymax, and the probability value Ynd according to the normal distribution of the number of lost packets Yi, and the number of consecutive drops X is the number of consecutive lost packets. 7. The connectivity evaluation system according to claim 6, wherein the average value Yave or random value Yrnd for each unit period Ti of Xi is used. The statistic calculation means obtains either the average value Yave of the jitter of the received media for each predetermined unit time Ti, the random value Yrnd, the worst value Ymax and the probability value Ynd according to the normal distribution,
4. The deterioration processing unit randomly delays the packet of the test media by any one of the average value Yave, the random value Yrnd, the worst value Ymax, and the probability value Ynd according to a normal distribution. The connectivity evaluation system described in 1.

Images