JP2017050753A - Quality monitoring device, system, and method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To allow the cause of quality deterioration in a network to be isolated in relation to a technique to monitor the quality of a network including a radio network.SOLUTION: A throughput analysis processing unit 609 monitors communication of each radio terminal in a network, and thereby calculates a throughput value for each data communication at each radio terminal. A main cause determination processing unit 610 calculates a throughput deterioration rate per data communication on the basis of the throughput value, checks the throughput deterioration rate with information on main cause determination stored in a main cause determination analysis table 611, and thereby determines the main cause of quality deterioration in the network.SELECTED DRAWING: Figure 6

Description

  The present invention relates to an apparatus, a system, and a method for monitoring quality of a network including a wireless network.

  With the background of wireless communication standards such as LTE (Long Term Evolution) and WiFi (Wireless Fidelity) that realize high-capacity data communication and response speed of wireless terminals, there are an increasing number of users who handle important information in wireless terminals. There is a need for technology that can quickly detect and respond to troubles by monitoring wireless quality.

  As a technique for monitoring wireless quality, a technique for measuring response and throughput with respect to operation of a wireless terminal is known.

  For example, Patent Document 1 is known as a conventional technique for performing quality control and monitoring of a network using measurement of throughput.

  Further, for example, Patent Document 2 is known as a conventional technique for selecting an appropriate throughput improvement measure according to a factor of throughput reduction.

JP 2007-036839 A JP 2006-217495 A

  In the above-described conventional technique, the throughput is measured on an average for a predetermined time (for example, 1 minute), and only the presence or absence of a decrease in throughput is grasped. Also, only the total amount (capacity) of the line base on the EPC (Evolved Packet Core) network side where the wireless networks are aggregated was seen.

  For this reason, it has not been possible to determine whether the cause of quality degradation in the network is in the wireless network or in the EPC network, and more specific cause determination.

  Furthermore, as a result of relying on the know-how of engineers to determine the cause, labor costs have been high.

  It is also conceivable to install a dedicated application for identifying the cause of the throughput decrease in the wireless terminal. However, in this case, since it is necessary to make the application correspond to each of the terminal model, the operating system, and the hardware implementation, the cost is always increased. In addition, all or many of the terminal users have to install the application themselves, which makes it difficult to promote the introduction.

  Accordingly, an object of one aspect of the present invention is to enable identification of causes of quality deterioration in a network.

  In one example, a means for calculating a throughput value for each data communication in a session in a network including a wireless network and a packet network that aggregates the wireless networks, a means for calculating a throughput deterioration rate from each throughput value, There is provided a quality monitoring apparatus comprising: means for identifying a cause of quality deterioration in a network based on a deterioration rate.

  The cause of quality degradation in the network can be identified.

It is a figure which shows the network structural example of this embodiment. It is explanatory drawing of the calculation process of a throughput value. It is explanatory drawing of the division | segmentation process of packet data. It is explanatory drawing (the 1) of the isolation | separation process by main cause identification. It is explanatory drawing (the 2) of the isolation | separation process by main cause identification. It is a block diagram which shows the structural example of a quality monitoring measurement apparatus. It is a figure which shows the hardware structural example of a quality monitoring measurement apparatus. It is a figure which shows the data structural example of a main cause specific analysis table. It is a figure which shows the data structural example of a data communication management table. It is a flowchart (the 1) which shows the example of the whole control processing of a quality monitoring measurement apparatus. It is a flowchart (the 2) which shows the example of the whole control processing of a quality monitoring measurement apparatus. It is a flowchart which shows the detailed example of a main cause specific process. It is a flowchart which shows the example of the threshold value setting process of the allowable range of a throughput degradation rate. It is a figure which shows the operation example (throughput value calculation operation | movement) of embodiment. It is a figure which shows the operation example (contents of a data communication management table) of embodiment.

Hereinafter, embodiments for carrying out the present invention will be described in detail with reference to the drawings.
FIG. 1 is a diagram illustrating a network configuration example of the present embodiment.

  A plurality of wireless terminals 104 are connected to the neighboring base stations 105 according to, for example, the LTE wireless standard, whereby the wireless network 101 is formed.

  Each base station 105 is aggregated by the EPC network 102. The EPC network is mainly configured by a packet network that aggregates radio networks of base stations. The EPC network may include a radio transmission path of a microwave communication network, but the stability of communication is maintained because it is not a mobile transmission network. The EPC network includes an S-GW (Serving Gateway) that processes user packet data and a P-GW (Packet data network Gateway) that processes connection to the core network 103 in the subsequent stage such as the Internet.

  The wireless network 101 and the EPC network 102 constructed by the base station 105 together form an LTE network.

  The wireless terminal 104 communicates with the application / Web server 106 via the base station 105, the EPC network 102, and the core network 103.

  In this embodiment, a quality monitoring / measuring device (quality monitoring device) is connected to the communication line of the capture point 108 in the EPC network 102 that aggregates the base stations 105. The quality monitoring / measurement apparatus 100 captures (passive measurement) HTTP (hypertext transfer protocol) communication data flowing in the LTE network at the capture point 108. Here, HTTP is an example of TCP (Transmission Control Protocol), the present embodiment can be applied to TCP, and the following HTTP session can be read as a TCP session and applied. The quality monitoring and measuring apparatus 100 calculates a throughput value and further calculates a throughput deterioration rate for each data communication in an HTTP session to be described later with the application / Web server 106 by each wireless terminal 104. At this time, the allowable range of the throughput deterioration rate is determined, and the cause of delay of the network from the wireless network 101 to the EPC network 102 is determined. When a determination other than “no abnormality” is made as a result of the delay cause isolation process, the determination result (specified delay cause) is notified from the quality monitoring measurement device 100 to the system operation terminal 107. The allowable range of the throughput deterioration rate is set by the system operator accessing the quality monitoring / measurement apparatus 100 from the system operation terminal 107.

  FIG. 2 is an explanatory diagram of the throughput value calculation process in the quality monitoring and measuring apparatus 100. This figure exemplifies HTTP communication flowing in the LTE network. This communication is an example in which the wireless terminal 104 communicates with the application / Web server 106 via the base station 105, the EPC network 102, and the core network 103. The quality monitoring and measuring apparatus 100 connected to the capture point 108 in the EPC network 102 through which data between the wireless terminal 104 and the application / Web server 106 shown in FIG. 1 passes monitors the communication.

  Now, a data transmission request is transmitted from the wireless terminal 104 to the application / Web server 106. This data is called a request. In response to the request, the application / Web server 106 transmits packet data to the wireless terminal 104. This packet data is called a response. When receiving a response from the application / Web server 106, the wireless terminal 104 returns an affirmative response indicating that the reception has been completed. This acknowledgment is called ACK. This combination of response and ACK is called data communication. After receiving the request, the application / Web server 106 sends a first response to the request via the core network 103, the EPC network 102, and the base station 105 in FIG. 1, as indicated by the solid line R1 in FIG. Is transmitted to the wireless terminal 104 that transmitted. In response to this response R1, the wireless terminal 104 sends an ACK to the application / Web server via the base station 105, the EPC network 102, and the core network 103 in FIG. 1, as indicated by a broken line ACK1 in FIG. Respond to 106. The application / Web server 106 sequentially transmits R2, R3, and R4 responses following the first response R1. Upon receiving these responses, the wireless terminal 104 receives and responds to the corresponding ACK2, ACK3, and ACK4. When the application / Web server 106 receives ACK4 corresponding to the final response R4 from the wireless terminal 104, the application / Web server 106 returns data indicating completion of data transmission to the wireless terminal 104. This is called FIN. A set of a series of data communications from the above request to FIN is called one session (201 in the figure).

  Here, the response returned from the application / Web server 106 to the wireless terminal 104 is, for example, data for one page of a home page. In this case, when the packet data indicating the response is larger than the window size indicating the maximum data size of data transfer on the network, the following division process is generally executed. FIG. 3 is an explanatory diagram of the packet data dividing process. The original data 301 before the division, which is a response, is divided into data 1 to 3 each having an MTU (Maximum Transfer Unit) length as indicated by 302, and each of the data 1 to 3 has an IP. (Internet protocol) A header is added and packet data is generated and transmitted. At this time, the packet data including the final data 4 is less than the MTU length, and the data amounts of the data 1 to 3 and the data communication are not uniform.

  Each packet data of data 1 to 4 transmitted as described above corresponds to R1, R2, R3, and R4 (final data packet) in FIG.

  Each of I and II in FIG. 2 indicates the measurement timing of the throughput value. The throughput value is calculated by the following equation. In the following description, “response time” and “ACK time” refer to the times when the response monitoring and the ACK packet data are respectively captured (passive measurement) by the quality monitoring measurement device 100.

Throughput value [bps (bits / second)] =
Response traffic [bytes (bytes)] ÷
Time from response time to corresponding ACK time [s (seconds)]

  Here, the amount of ACK data is negligible compared to the amount of response data, and is ignored.

  Based on the above equation (1), for example, the throughput value I [bps] in FIG. 2 captures the communication amount [byte] of the response R2 measured by the quality monitoring measurement device 100, and the quality monitoring measurement device 100 captures the response R2. It is calculated as a value divided by the difference time [s] from the time to the time when ACK2 was captured.

  Similarly, the throughput value II [bps] in FIG. 2 is obtained by capturing the communication amount [byte] of the response R3 measured by the quality monitoring and measuring apparatus 100 and ACK3 from the time when the quality monitoring and measuring apparatus 100 captures the response R3. It is calculated as a value divided by the time difference [s] up to the time.

  Further, in the present embodiment, the quality monitoring and measuring apparatus 100 uses the throughput values of the data communication one before and the last data communication of the session (throughput values I and II in the case of FIG. 2), For example, the throughput deterioration rate is calculated by the following equation.

Throughput degradation rate =
(Throughput value II−throughput value I) ÷ throughput value I × 100 [%]
... (2)

  Here, the reason why the throughput value of the final data communication is not used is that the data amount of the final data is not uniform with the data amounts of the other data 1 to 3 as shown as data 4 in FIG. .

  According to the present embodiment, whether the cause of quality degradation in the network is an EPC network, that is, a packet network in which wireless networks are aggregated, a lack of resources in the wireless network, packet non-arrival, or the like, accompanied by interruption in the wireless network It is possible to clearly determine the cause isolation such as whether it is due to handover or reconnection. Then, it is possible to determine which area is the cause of the wireless network side or the EPC network side.

  Since regularity can be defined and shared for the throughput measurement results, it is not necessary to rely on the know-how of engineers to investigate the cause, thereby reducing labor costs.

  It is not necessary to install a dedicated application for identifying the cause of quality degradation, and measurement can be performed only with a quality monitoring device placed at an aggregated capture point of the network line on the base station side.

  The allowable range of the data communication throughput degradation rate between each wireless terminal session and the application / WEB server can be defined, and the cause of delay between the wireless network and the EPC network can be separated.

  In the present embodiment, the throughput deterioration rate is calculated based on the above formula (2) based on the throughput values of the data communications one session before and two times before the final data communication of the session. The calculation of the deterioration rate is not limited to this. For example, the throughput deterioration rate may be calculated based on the throughput values of adjacent data communications other than the last data communication in the session, such as two previous, three previous, and four previous. In the following, for the sake of simplicity of explanation, the throughput deterioration rate is calculated in accordance with the above equation (2) based on the throughput values of the data communication before and after the last data communication of the session. To do.

  4 and 5 are explanatory diagrams of the separation process by specifying the main cause using the throughput deterioration rate calculated according to the above equation (2).

  In the present embodiment, the quality monitoring and measuring apparatus 100 in FIG. 1 performs the wireless network 101 and the EPC network 102 in FIG. 1 based on the throughput deterioration rate calculated based on the equation (1) and in accordance with the equation (2). The main cause of quality degradation in the LTE network including In the present embodiment, the quality monitoring and measuring apparatus 100 can classify main causes of quality degradation in the LTE network into those caused by the EPC network 102 that aggregates the wireless networks 101 and those caused by the wireless networks 101. . Further, the quality monitoring / measurement apparatus 100 is mainly caused by the radio network 101 due to resource shortage in the radio network 101 and due to handover and reconnection with non-arrival or interruption of packets in the radio network 101. Can be divided into things.

  Here, the “resource in the wireless network 101” refers to a data communication band in the base station 105 in the wireless network 101, for example. “Resources” are generally prepared with a margin with respect to the average number of wireless terminals 104 connected to one base station 105. Therefore, if the number of wireless terminals 104 connected to one base station 105 exceeds the average number, communication exceeding the allowable amount of “resource” is performed, and a resource shortage state may occur.

  “Handover” refers to switching the base station 105 communicating with the wireless terminal 104 while the wireless terminal 104 is moving. For example, in a wireless terminal 104 such as a mobile phone which is a mobile communication adopting a cell system, when the wireless terminal 104 moves to a cell boundary or for other reasons, if the radio wave from the base station 105 becomes weak, it can communicate as it is. Disappear. Therefore, before the radio wave is weakened or weakened, control is performed to switch to the base station 105 having a strong radio wave in another cell. At this time, packet non-arrival or interruption may occur.

  FIG. 4A is a diagram showing a sequence example of a session when the main cause of quality degradation in the LTE network is caused by the EPC network 102. In FIG. 4A, the throughput deterioration rate is calculated by the following equation, for example, according to the equation (2).

Throughput degradation rate =
(Throughput value II−throughput value I) ÷ throughput value I × 100 [%]
= Less than 0.1 [%]

  In FIG. 4 (a), it can be seen from the throughput deterioration rate that delays occur at substantially equal intervals or minutely. In this case, when there is almost no delay in the wireless network 101 from the wireless terminal 104 to the base station 105, that is, there is a problem in quality degradation in the LTE network, the LTE network is not connected to the EPC network 102 side but the wireless network 101 side. It can be determined that there is a main cause of quality degradation in the network.

  FIG. 4B is a diagram illustrating a session sequence example when the main cause of quality degradation in the LTE network is due to the wireless network 101. In FIG. 4B, the throughput deterioration rate is calculated by the following equation, for example, according to the equation (2).

Throughput degradation rate =
(Throughput value IV-Throughput value III) ÷ Throughput value III x 100 [%]
= 0.1 [%] or more

  In FIG. 4B, it can be seen that the delay is increased from the throughput deterioration rate. In this case, it can be seen that there is a delay in the wireless network 101 from the wireless terminal 104 to the base station 105. From the session example of FIG. 4B, it can be seen that the response is delayed between the wireless terminal 104 and the base station 105 as compared to FIGS. 2 and 4A.

  In this embodiment, as an example, the throughput deterioration rate of FIG. 4A is less than 0.1 [%] and the throughput deterioration rate of FIG. 4B is 0.1 [%] or more. The user can change the allowable range of the throughput deterioration rate for each 105 or the wireless terminal 104 (see the description of FIG. 8 and FIG. 13 described later).

  FIG. 5A shows a sequence of a session when resource shortage or packet non-arrival in the base station 105 of the wireless network 101 is the main cause of quality degradation in the LTE network. FIG. 5B shows a sequence of sessions in the case where handover and reconnection with interruption of the radio network 101 are the main causes of quality degradation in the LTE network.

  In FIG. 5 (a), the throughput deterioration rate is calculated by the following equation, for example, according to the equation (2).

Throughput degradation rate =
(Throughput value VI−throughput value V) ÷ throughput value V × 100 [%]
= 0.1 [%] or more and less than 10 [%]

  As shown in FIG. 5A, when the throughput deterioration rate gradually increases, communication is performed in the wireless network 101 between the base station 105 and the wireless terminal 104 as compared with between the base station 105 and the quality monitoring and measuring apparatus 100. Insufficient bandwidth. That is, it can be determined that the shortage of resources of the base station 105, non-arrival of packets, and the like are the main causes of quality degradation in the LTE network.

  When the resources of the base station 105 are insufficient, places such as the base station 105 and the access point have a certain amount of resources. As the degree of packet data retention gradually increases with respect to the resource, the throughput value gradually delays and the throughput deterioration rate increases.

  In FIG. 5B, the throughput degradation rate is calculated according to the following equation, for example, according to the equation (2).

Throughput degradation rate =
(Throughput value VIII-throughput value VI) ÷ throughput value VI x 100 [%]
= 10 [%] or more

  As shown in FIG. 5 (b), when the throughput deterioration rate greatly increases, it is determined that the wireless network 101 has a reduced throughput due to a cause such as handover, reconnection, etc., which takes a long time in the wireless network 101. can do.

  In this embodiment, as an example, the throughput deterioration rate in FIG. 5A is 0.1 [%] or more and less than 10 [%], and the throughput deterioration rate in FIG. 5B is 10 [%] or more. The user can change the permissible range of the throughput deterioration rate for each base station 105 or wireless terminal 104 (see the description of FIG. 8 and FIG. 13 described later).

  The detailed configuration and operation of the present embodiment based on the principle of the operation for isolating the main cause of quality deterioration in the LTE network will be described below.

  FIG. 6 is a block diagram illustrating a configuration example of the quality monitoring measurement apparatus 100 of FIG. The quality monitoring and measuring apparatus 100 includes a net interface (“Net I / F” in the figure) 601, a collection unit 602, an analysis unit 603, a storage unit 604, a notification unit 605, and a setting unit 606. The collection unit 602 includes a packet capture processing unit 607. The analysis unit 603 includes a collected information analysis unit 608, a throughput analysis processing unit 609, and a main cause identification processing unit 610. The storage unit 604 includes a main cause identification analysis table 611, a data communication management table 612, and a main cause identification result storage unit 613. The notification unit 605 includes a failure notification unit 614. The setting unit 606 includes a main cause identification threshold setting unit 615.

  The packet capture processing unit 607 in the collection unit 602 captures the packet data flowing on the communication line 617 of the capture point 108 in the EPC network 102 in FIG. 1 from the TAP 616 and delivers it to the analysis unit 603. The TAP 616 is a device capable of branching and monitoring packet data on the communication line 617 by being inserted into the communication line 617.

  The collected information analysis unit 608 in the analysis unit 603 updates the contents of the data communication management table 612 in the storage unit 604 based on the packet data information received from the packet capture processing unit 607.

  A throughput analysis processing unit 609 in the analysis unit 603 sequentially calculates a throughput value corresponding to each data communication for each session and writes it in the data communication management table 612.

  The main cause identification processing unit 610 in the analysis unit 603 calculates a throughput deterioration rate in the session at the timing when the FIN packet data is received for each session. Then, the main cause identification processing unit 610 collates the calculated throughput degradation rate with the main cause identification information stored in the main cause identification analysis table 611 in the storage unit 604, thereby reducing the quality degradation in the LTE network. Identify the main cause. After the main cause identification processing unit 610 stores the main cause identification result in the main cause identification result storage unit 613 in the storage unit 604, the main cause identification processing unit 610 identifies the main cause identification in the failure notification unit 614 in the notification unit 605. Deliver the result.

  FIG. 7 is a diagram illustrating a hardware configuration example of the quality monitoring and measuring apparatus 100 corresponding to the configuration of FIG. The hardware of the quality monitoring and measuring apparatus 100 is, for example, a computer having a configuration in which a CPU 701, a ROM 702, a RAM 703, a disk storage device 704, a network interface controller (NIC) 705, and the like are connected to each other via a bus 706. The configuration shown in FIG. 7 is an example of a computer that can implement the quality monitoring and measuring apparatus 100, and such a computer is not limited to this configuration.

  The CPU 701 controls the entire quality monitoring and measuring apparatus 100. The ROM 702 stores various control programs for controlling the quality monitoring and measuring apparatus 100, initial data, and the like. The CPU 701 implements the functions of the blocks 608 to 615 in FIG. 6 as shown in FIG. 7 by reading the control program in the ROM 702 and developing it in the RAM 703.

The NIC 705 corresponds to the Net I / F 601 in FIG.
The disk storage device 704 stores data to be stored for a long time.

  The quality monitoring and measuring apparatus 100 according to the present embodiment is realized by the CPU 701 executing a program having functions realized by the flowcharts of FIGS. 11 to 13 and the like. The program may be recorded and distributed on, for example, a disk storage device 704 or a portable recording medium (not shown), or may be acquired from a network via the NIC 705.

  FIG. 8 is a diagram illustrating a data configuration example of the main cause identification analysis table 611 in FIGS. 6 and 7. The allowable range is different depending on the external environment, the model of the wireless terminal 104, the communication standard (LTE, 3G (third generation mobile communication network), WiFi), the base station 105, and the like. Therefore, the main cause identification analysis table 611 has a configuration in which the setting of the allowable range of the throughput deterioration rate can be changed according to each environment. Since the delay rate of the network changes with the amount of data, in this embodiment, the allowable range of the throughput deterioration rate is set in the main cause specifying analysis table 611 and each cause of the network quality deterioration is specified.

  In the main cause identification analysis table 611, the% ratio of the increase amount of the throughput, that is, the allowable range of the throughput deterioration rate is set, and the cause is classified according to the allowable range. If the throughput deterioration rate calculated according to the above-described equation (2) is α [%], the relationship between the allowable range of α and the cause of quality deterioration of the LTE network can be determined as follows, for example. .

0.01%> α: No abnormality 0.01% <α ≦ 0.1%: Caused by EPC network 102 0.1% <α ≦ 10%: Caused by wireless network 101 (radio resource shortage, etc.)
10% <α: Caused by the wireless network 101
(Handover / reconnection with interruption)

  What stores such a relationship for each wireless terminal 104, base station 105, and the like is a main cause identification analysis table 611 shown in FIG.

  FIG. 9 is a diagram illustrating a data configuration example of the data communication management table 612 in FIGS. 6 and 7.

  When request data, that is, new session information is received, a new entry is added to the data communication management table 612, and a management item group of “session information” in the entry is registered. As the management item group of “session information”, the following items can be managed.

● “Session number” item: A number identifying the session is recorded.
“Source IP address” item: The IP address of the source device (application / Web server 106 or wireless terminal 104) is recorded.
"Destination IP address" item: The IP address of the destination device (wireless terminal 104 or application / Web server 106) is recorded.
“Source MAC address” item: The MAC (Media Access Control) address of the source device (application / Web server 106 or wireless terminal 104) is recorded.
“Destination MAC address” item: The MAC address of the destination device (wireless terminal 104 or application / Web server 106) is recorded.
● “VLAN” item: Records identification information (ID) of a virtual local area network.
“Communication direction” item: (wireless terminal-> server) or (server-> wireless terminal) is recorded.
(Wireless terminal-> server): The direction in which packet data is transmitted from the wireless terminal 104 to the application / Web server 106.
(Server-> wireless terminal): Reverse direction as described above.

  In this embodiment, the throughput degradation rate is calculated using data communication in which “server → wireless terminal” (download) is recorded in the “communication direction”. (Upload) data communication may be recorded.

  In each entry of the data communication management table 612 in FIG. 9, the value of the “reception time” item is captured every time the request or response of the session is captured by the packet capture processing unit 607 in the quality monitoring measurement device 100. Updated with the time value.

In each entry of the data communication management table 612 in FIG. 9, the “throughput value for each data communication [bps]” item is calculated based on the above-described equation (1) every time an ACK for the session is received. Throughput values are recorded sequentially. In the above item of FIG. 9, N ₁ is a throughput value calculated from the first data communication (for example, the pair of R1 and ACK1 in FIG. 2) of the session corresponding to the entry. Nn Is a throughput value calculated from the last data communication (for example, a pair of R4 and ACK4 in FIG. 2) of the session corresponding to the entry. Nn-2 And Nn-1 Is a throughput value (for example, I in FIG. 2) that is calculated from each data communication (for example, the pair of R3 and ACK3 in FIG. 2 and the pair of R2 and ACK2 in FIG. 2) from the end of the session corresponding to the entry. And II).

  As described above, in the present embodiment, after the FIN (see FIG. 2) is received, the throughput value II of the previous data communication and the previous two data communication indicated by the broken line frame 901 in FIG. Is used to calculate the throughput deterioration rate according to the above-described equation (2). The reason why the throughput value of the final data communication is not used has been described above with reference to FIGS.

  FIGS. 10 and 11 are flowcharts showing an example of the overall control processing of the quality monitoring and measuring apparatus 100 of FIGS. Hereinafter, unless otherwise specified, each part in FIG. 6 or FIG. 7 is referred to.

  When the power of the quality monitoring and measuring apparatus 100 is turned on, the CPU 701 first transfers a program for overall control processing from the ROM 702 to the RAM 703 and starts the program process (step S1001 in FIG. 10).

  The CPU 701 executes processing of the packet capture processing unit 607 (step S1002 in FIG. 10). The packet capture processing unit 607 captures, from the TAP 616, packet data that flows on the communication line 617 of the capture point 108 in the EPC network 102 of FIG. Then, the packet capture processing unit 607 captures the packet data captured by the TAP 616 through the Net I / F 601 and determines whether the packet data can be analyzed. The packet capture processing unit 607 passes the packet data that can be analyzed to the processing of the collection information analysis unit 608 in the analysis unit 603, and discards abnormal packet data that cannot be analyzed.

  The CPU 701 executes processing of the collection information analysis unit 608, and first analyzes packet data delivered from the processing of the packet capture processing unit 607 (step S1003 in FIG. 10).

  Thereafter, the collected information analysis unit 608 determines whether or not the analyzed packet data is a request (step S1004 in FIG. 10).

  When the received packet data is a request (the determination in step S1004 is YES), the collected information analysis unit 608 additionally registers a new entry in the data communication management table 612 in FIG. The collected information analysis unit 608 captures the packet information in the new entry, in addition to the IP address and MAC address of both the transmission and reception extracted from the header of the packet data, the session information such as the communication direction and VLAN, and the like. The reception time of the hour is recorded (step S1005).

  Thereafter, the CPU 701 returns to the process of the packet capture processing unit 607 in step S1002.

  If the received packet data is not a request (NO in step S1004), the collected information analysis unit 608 further determines whether the packet data is a response (step S1006 in FIG. 11).

  When the packet data is a response (the determination in step S1006 is YES), the collected information analysis unit 608 displays the “reception time” item of the corresponding entry in the data communication management table 612 in FIG. 9 when the response is captured. The reception time is updated (step S1007 in FIG. 11).

  Thereafter, the CPU 701 returns to the processing of the packet capture processing unit 607 in step S1002 of FIG.

  When the received packet data is not a response (NO in step S1006), the collected information analysis unit 608 executes processing of the throughput analysis processing unit 609.

  The throughput analysis processing unit 609 determines whether the packet data received from the collection information analysis unit 608 is FIN (step S1008 in FIG. 11).

  If the received packet data is not FIN (determination in step S1008 is NO), that is, if it is ACK, the throughput analysis processing unit 609 executes the following processing. The throughput analysis processing unit 609 reads from the data communication management table 612 the time value of the “reception time” item of the session entry corresponding to the ACK (the reception time of the response of the same data communication). Then, the throughput analysis processing unit 609 calculates a throughput value based on the above equation (1) using the reception time when the ACK is captured and the read reception time (step S1009 in FIG. 11). ).

  Then, the throughput analysis processing unit 609 stores the calculated throughput value in the “throughput value for each data communication [bps]” column of the entry of the session corresponding to the ACK in the data communication management table 612 of FIG. (Step S1010 in FIG. 11).

  Thereafter, the CPU 701 returns to the processing of the packet capture processing unit 607 in step S1002 of FIG.

  When the packet data received from the collected information analysis unit 608 is FIN (determination in step S1008 is YES), the CPU 701 executes processing of the main cause identification processing unit 610.

  First, the main cause identification processing unit 610 executes main cause identification processing (step S1011 in FIG. 11). Details of this processing will be described later with reference to the flowchart of FIG. 12, but here, the throughput deterioration rate is calculated and the main cause is specified.

  The main cause identification processing unit 610 determines whether or not the throughput deterioration rate is “no abnormality” as a result of the main cause identification processing in step S1012 (step S1012 in FIG. 11).

  When the throughput deterioration rate is within the allowable range of “no abnormality” (see FIG. 8) (determination in step S1012 is YES), the following processing is executed. The main cause identification processing unit 610 deletes the entry of the session corresponding to the FIN packet data received from the throughput analysis processing unit 609 in the data communication management table 612 in FIG. 9 (step S1015 in FIG. 11).

  Thereafter, the CPU 701 returns to the processing of the packet capture processing unit 607 in step S1002 of FIG.

  If the throughput deterioration rate is not within the allowable range of “no abnormality” (see FIG. 8) (the determination in step S1012 is NO), the following processing is executed.

  The main cause specifying processing unit 610 stores the main cause specifying result obtained in the main cause specifying process in step S1011 together with the session information in the main cause specifying result storage unit 613 (step S1013 in FIG. 11).

  Next, the main cause specifying processing unit 610 notifies the processing of the failure notifying unit 614 of the specified result of the main cause (step S1014 in FIG. 11).

  Then, the main cause identification processing unit 610 deletes the entry of the session corresponding to the FIN packet data received from the throughput analysis processing unit 609 in the data communication management table 612 in FIG. 9 (Step S1015 in FIG. 11).

  Thereafter, the CPU 701 returns to the processing of the packet capture processing unit 607 in step S1002 of FIG.

  The details of the processing of the failure notification unit 614 are omitted, but the processing is started by the event notification of the specific result of the main cause from the processing of the main cause specifying processing unit 610, and the main cause is specified in the e-mail address of the system administrator The result of the transmission is notified, or the specific result of the main cause is popped up on the display screen of the system administrator's computer.

  FIG. 12 is a flowchart showing a detailed example of the main cause identifying process in step S1011 of FIG.

  The main cause identification processing unit 610 reads a set of session information and a throughput value from the session entry corresponding to the FIN packet data received from the throughput analysis processing unit 609 in the data communication management table 612 of FIG. As a set of throughput values, as described above, the previous and second previous throughput values are read (step S1201 in FIG. 12).

  The main cause identification processing unit 610 calculates a throughput deterioration rate according to the above-described equation (2) using the throughput values one and two before and after the last read (step S1202 in FIG. 12). .

  The main cause identification processing unit 610 refers to the main cause identification analysis table 611 illustrated in FIG. 8 using the read session information as a key, and reads information on the allowable range of the throughput deterioration rate (step S1203 in FIG. 12). .

  The main cause identification processing unit 610 compares the throughput deterioration rate calculated in step S1202 with the allowable range of the throughput deterioration rate read in step S1203, and specifies the main cause (step S1204 in FIG. 12).

  FIG. 13 is a flowchart illustrating an example of the threshold setting process for the allowable range of the throughput deterioration rate. This process is activated by a setting instruction from the system operation terminal 107 in FIG.

  The CPU 701 acquires the contents of the main cause identification analysis table 611 illustrated in FIG. 8 for each wireless terminal 104 model and the base station 105 (step S1301 in FIG. 13).

  Next, the CPU 701 determines whether the instruction from the system operation terminal 107 is setting or deleting an allowable range threshold (step S1302 in FIG. 13).

  When the instruction from the system operation terminal 107 is deletion of the threshold value of the allowable range, the CPU 701 deletes the corresponding part instructed to be deleted from the main cause identification analysis table 611 (step S1303 in FIG. 13).

  When the instruction from the system operation terminal 107 is a setting of the allowable range threshold, the CPU 701 further determines whether the instruction from the system operation terminal 107 is a new setting or an update setting of the allowable range threshold (step S1304 in FIG. 13). ).

  When the instruction from the system operation terminal 107 is an update setting of the threshold value of the allowable range, the CPU 701 updates the corresponding part of the main cause identification analysis table 611 instructed to update (step S1305 in FIG. 13).

  When the instruction from the system operation terminal 107 is a new setting of the threshold value of the allowable range, the CPU 701 adds a new allowable range condition to the main cause identification analysis table 611 (step S1306 in FIG. 13).

  After the processing in step S1303, S1305, or S1306, the CPU 701 saves the setting contents of the main cause identification analysis table 611 in the distortion device 100 or the like (step S1307 in FIG. 13).

  Thereafter, the CPU 701 ends the threshold value setting process of the allowable range of the throughput deterioration rate shown in the flowchart of FIG.

  14 and 15 are diagrams illustrating an example of the operation of the embodiment. FIG. 14 illustrates an example of the throughput value calculation operation, and FIG. 15 illustrates an example of the contents of the data communication management table 612. In the example of FIG. 14, the IP address aaa. aaa. aaa. wireless terminal 104 having aaa and IP address bbb. bbb. bbb. The application / Web server 106 having bbb performs HTTP communication similar to the case of FIG. Note that the value of each IP address is a schematic value for ease of explanation.

  The example of FIG. 15 is a recorded content example of the data communication management table 612 corresponding to the session of FIG. As described above, the throughput values used for obtaining the throughput deterioration rate are the data communication immediately before the last and the data communication two times before. The final data communication in FIG. 14 is a pair of response R4 and ACK4. Therefore, using the throughput value III of FIG. 15 corresponding to the pair of R3 and ACK3 of FIG. 14 and the throughput value II of FIG. 15 corresponding to the pair of R2 and ACK2 of FIG. Thus, the throughput deterioration rate is calculated.

  From FIG. 15, the throughput value II is 9.5 [bps], and the throughput value III is 9 [bps]. From this, the throughput deterioration rate can be calculated according to the above-described equation (2) as in the following equation.

Throughput deterioration rate = (9.5-9) ÷ 9 × 100 = 5.6 [%]
From the main cause identification analysis table 611 illustrated in FIG. 8, it can be determined that the main cause is a shortage of base station resources in the wireless network.

  As described above, according to the present embodiment, it is possible to determine and isolate a region where a cause of packet delay exists in the LTE network as a network or a more specific cause of packet delay. In addition, the operation cost for the cause identification investigation can be reduced, such as high cost reduction and efficiency improvement by know-how holders.

  Compared with introducing a dedicated application to each of the wireless terminals 104, the quality monitoring / measuring device 100 is disposed at the capture point 108 in the EPC network 102 on the side where the base stations 105 are aggregated. The introduction cost can be reduced.

  Furthermore, since the quality monitoring and measuring apparatus 100 does not depend on the number of wireless terminals 104, it is possible to construct a monitoring system at a low cost even if the wireless quality monitoring scale increases.

  In the embodiment described above, the example in which the quality monitoring / measurement apparatus 100 is connected to the LTE network including the wireless network 101 and the EPC network 102 in FIG. 1 has been described. The present embodiment may be applied to a wireless network such as a network.

Regarding the above embodiment, the following additional notes are disclosed.
(Appendix 1)
Means for calculating a throughput value for each data communication in a session in a network including a wireless network and a packet network in which the wireless networks are aggregated;
Means for calculating a throughput degradation rate from each throughput value;
Means for identifying the cause of quality degradation in the network based on the degradation rate;
Quality monitoring device consisting of
(Appendix 2)
The network includes a wireless network and a packet network in which the wireless networks are aggregated,
The quality monitoring device monitors communication from a packet network in which the wireless networks are aggregated;
The quality monitoring apparatus according to Supplementary Note 1, wherein:
(Appendix 3)
The means for identifying the main cause is to identify the cause of quality degradation by comparing the storage unit storing information for identifying the cause of quality degradation in the network and the degradation rate with the information. Consisting of the main cause identification processing part of
The quality monitoring apparatus according to appendix 1 or 2, characterized in that:
(Appendix 4)
The data communication is a set of response data in a session in TCP (Transmission Control Protocol) including a hypertext transfer protocol and acknowledgment data for the response data, and the means for calculating the throughput value includes the response in the data communication. The throughput value corresponding to the data communication is calculated from the difference between each reception time of the acknowledgment data with respect to the data and the response data,
4. The quality monitoring apparatus according to any one of appendices 1 to 3, wherein
(Appendix 5)
The main cause identification processing unit divides a main cause of quality degradation in the network into at least one of those caused by the radio network or a packet network that aggregates the radio network,
The quality monitoring apparatus according to appendix 3 or 4, characterized in that:
(Appendix 6)
The main cause identification processing unit determines that there is no abnormality when the throughput deterioration rate is small compared to the first threshold, the throughput deterioration rate is large compared to the first threshold, and the first When it is smaller than a second threshold value that is larger than a threshold value, it is determined that the main cause of quality degradation in the network is caused by a packet network that aggregates wireless networks included in the network, and the throughput degradation rate is Determining that the main cause of quality degradation in the network is attributed to the wireless network when larger than the second threshold;
The quality monitoring apparatus according to appendix 5, characterized by:
(Appendix 7)
The main cause identifying processing unit further causes a main cause of quality degradation in the network due to a resource shortage in the wireless network, or due to handover and reconnection with packet non-arrival or interruption in the wireless network. Carved into at least one of the things,
The quality monitoring apparatus according to any one of appendix 5 or 6, characterized by the above.
(Appendix 8)
The main cause identification processing unit further determines that the main cause of quality deterioration in the network is caused by a shortage of resources in the wireless network when the throughput deterioration rate is smaller than a third threshold, and the throughput Determining that the main cause of quality degradation in the network is due to handover and reconnection with packet non-arrival or interruption in the wireless network when the degradation rate is large compared to the third threshold;
The quality monitoring apparatus according to appendix 7, characterized by:
(Appendix 9)
The wireless network is either a WiFi network, a wireless network in an LTE network, or a third generation mobile communication network.
The quality monitoring apparatus according to Supplementary Note 1, wherein:
(Appendix 10)
Identification of the main cause of quality degradation in the network is performed for each virtual local area network, for each MAC address, or for each IP address.
The quality monitoring apparatus according to Supplementary Note 1, wherein:
(Appendix 11)
Calculating a throughput value for each data communication in a session in a packet network that aggregates at least wireless networks;
Calculating a throughput deterioration rate from each throughput value;
Identifying a cause of quality degradation in the network based on the degradation rate;
Quality monitoring method consisting of
(Appendix 12)
A wireless network including wireless terminals;
A base station that transmits and receives wireless data to and from the wireless network;
A packet network that communicates with and aggregates the base stations;
A server communicating with the wireless terminal via a packet network that aggregates the base stations;
The base station is connected to a network capture point, a throughput deterioration rate is calculated from the throughput value in the communication session, and the result is collated with cause setting information stored in the storage unit in advance. A quality monitoring device for setting the cause of the quality degradation of the communication,
System with.

DESCRIPTION OF SYMBOLS 101 Wireless network 102 EPC network 103 Core network 104 Wireless terminal 105 Base station 106 Application / Web server 107 System operation terminal 108 Capture point 201 Session 601 Net I / F (net interface)
602 Collection unit 603 Analysis unit 604 Storage unit 605 Notification unit 606 Setting unit 607 Packet capture processing unit 608 Collection information analysis unit 609 Throughput analysis processing unit 610 Main cause identification processing unit 611 Main cause identification analysis table 612 Data communication management table 613 Main cause Specific result storage unit 614 Failure notification unit 615 Main cause specific threshold setting unit 616 TAP
617 Communication line 701 CPU
702 Memory 703 RAM
704 Disk storage device 705 NIC
706 Bus 901 Original data before division 902 Data transmitted after division

Claims (10)

Means for calculating a throughput value for each data communication in a session in a network including a wireless network and a packet network in which the wireless networks are aggregated;
Means for calculating a throughput degradation rate from each throughput value;
Means for identifying the cause of quality degradation in the network based on the degradation rate;
Quality monitoring device consisting of The network includes a wireless network and a packet network in which the wireless networks are aggregated,
The quality monitoring device monitors communication from a packet network in which the wireless networks are aggregated;
The quality monitoring apparatus according to claim 1. The means for identifying the cause is for identifying a cause of quality degradation by comparing a storage unit storing information for identifying the cause of quality degradation in the network and the degradation rate with the information. Consists of the cause identification processing part,
The quality monitoring apparatus according to claim 1 or 2, characterized in that The data communication is a set of response data in a session in TCP (Transmission Control Protocol) including a hypertext transfer protocol and acknowledgment data for the response data, and the means for calculating the throughput value includes the response in the data communication. The throughput value corresponding to the data communication is calculated from the difference between each reception time of the acknowledgment data with respect to the data and the response data,
The quality monitoring apparatus according to any one of claims 1 to 3, wherein The cause identifying processing unit divides the cause of quality degradation in the network into at least one of those caused by the radio network or a packet network that aggregates the radio network,
The quality monitoring apparatus according to claim 3 or 4, characterized in that: The cause identification processing unit determines that there is no abnormality when the throughput deterioration rate is small compared to the first threshold, the throughput deterioration rate is large compared to the first threshold, and the first threshold The main cause of quality degradation in the network is attributed to a packet network in which radio networks included in the network are aggregated, and the throughput degradation rate is the first Determining that the main cause of quality degradation in the network is caused by the wireless network when it is larger than the threshold value of 2;
The quality monitoring apparatus according to claim 3 or 4, characterized in that: The cause identification processing unit may cause at least one of causes of quality degradation in the network due to resource shortage in the wireless network, or due to handover and reconnection involving packet non-arrival or interruption in the wireless network. Carved into crab,
The quality monitoring apparatus according to claim 3, wherein the quality monitoring apparatus is characterized in that: The cause identification processing unit determines that the main cause of quality degradation in the network is caused by a shortage of resources in the wireless network when the throughput degradation rate is smaller than a third threshold, and the throughput degradation rate Determining that the main cause of quality degradation in the network is due to handover and reconnection with packet non-arrival or interruption in the wireless network when is greater than the third threshold;
The quality monitoring apparatus according to claim 3 or 4, characterized in that: Calculating a throughput value for each data communication in a session in a network including at least a wireless network;
Calculating a throughput deterioration rate from each throughput value;
Identifying a cause of quality degradation in the network based on the degradation rate;
Quality monitoring method consisting of A wireless network including wireless terminals;
A base station that transmits and receives wireless data to and from the wireless network;
A packet network that communicates with and aggregates the base stations;
A server communicating with the wireless terminal via a packet network that aggregates the base stations;
The base station is connected to a network capture point, a throughput deterioration rate is calculated from a throughput value in the communication session, and the result is collated with cause setting information stored in advance in a storage unit. A quality monitoring device for setting the cause of the quality degradation of the communication,
System with.