JP2016116104A - Program, server, system, and method for measuring communication band using user data - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a program or the like in which a communication band of a network can be measured using user data even when the communication band in a terminal largely varies.SOLUTION: A program for functioning a computer installed in a server which transmits so as to measure a communication band of a network by a terminal using user data to be replied to the terminal allows the computer to execute: a first step in which the user data to be replied is buffered in a predetermined unit time and it is determined whether or not the buffer amount is a predetermined threshold or less; and a second step in which, when it is determined as true in the first step, the user data is divided, divided each data is configured in a UDP packet, each UDP packet is transmitted on the basis of an available band estimation, the user data is configured in a TCP packet when it is determined as false in the first step, and each TCP packet is transmitted on the basis of a throughput measurement.SELECTED DRAWING: Figure 2

Description

  The present invention relates to a technique for executing an active measurement method for a network.

  It is important for service providers of the Internet and mobile communications to monitor the communication bandwidth of the network that they operate. When the degradation of the communication band is detected, it is necessary to improve the communication band by installing a wireless base station or controlling / regulating the communication band. Specifically, there is an “active measurement method” that measures the communication bandwidth by transferring test packets to the network and a “passive measurement method” that measures the communication bandwidth from the actual transmission / reception state behavior of user data. is there.

Conventionally, as an active measurement method, for example, there are the following methods.
There is a technique for transmitting and receiving test data between a measurement server and a client and estimating TCP (Transmission Control Protocol) throughput from the amount of transfer data / transfer time (see, for example, Non-Patent Documents 1 and 2). In addition, the time interval of multiple test packets is controlled and transmitted between the measurement server and the client (packet train), and the available bandwidth is estimated by analyzing the increasing tendency of the queuing delay of each test packet on the receiving side. There is also a technique (see Non-Patent Document 3, for example). Here, the (maximum) TCP throughput is adjusted near the upper limit in accordance with the free capacity of the network line by the TCP congestion control function installed in the transmission / reception apparatus. Therefore, in an ideal environment, it is considered that there is a strong correlation with the available bandwidth between the transmitting and receiving apparatuses.

  On the other hand, there is a technique for measuring a communication band between transmitting and receiving apparatuses without transmitting a test packet for measurement to a network (see, for example, Non-Patent Documents 4 and 5). According to this technology, a video streaming application or a voice call application itself that operates on a terminal dynamically changes a transmission rate of transmission data. Then, a successful transfer rate that does not cause an increase in packet loss or packet delay is searched.

  Further, as an active measurement method using user data, there is a technique for transmitting and receiving a test packet in which user data is embedded in a payload portion between transmitting and receiving apparatuses and estimating an available bandwidth (for example, see Non-Patent Document 6). This technology is implemented in Linux (registered trademark).

  Furthermore, there is a TCP proxy technology in which equipment devices in a network divide the connection into a plurality of communication connections and relay packets in order to increase the speed of communication connections set between transmitting and receiving devices (for example, non-patent literature). 7).

  Further, in order to improve security, there is also a web proxy technology in which an HTTP connection from a terminal is once terminated at a communication equipment device on a route, and the equipment device acts as a proxy to access a web server (for example, refer to Non-Patent Document 8). ).

SPEEDTEST, [online], [Search November 10, 2014], Internet <URL: http://www.speedtest.net/> iperf, [online], [November 10, 2014 search], Internet <URL: http://code.google.com/p/iperf/> Vinay J. Ribeiro, Rudolf H. Riedi, Richard G. Baraniuk, Jiri Navratil, Les Cottrell, “pathChirp: Efficient Available Bandwidth Estimation for Network Paths,” (2003), [online], [November 10, 2014 search ] Internet <URL: http://www.slac.stanford.edu/cgi-wrap/getdoc/slac-pub-9732.ps.gz> David Hassoun, “Dynamic stream switching with Flash Media Server 3,” April 2008, [online], [searched on November 10, 2014], Internet <URL: http://www.adobe.com/devnet/adobe- media-server / articles / dynamic_stream_switching.html> L. De Cicco, S. Mascolo, and V. Palmisano, “Skype Video Responsiveness to Bandwidth Variations,” NOSSDAV, 2008, [online], [searched November 10, 2014], Internet <URL: http: // c3lab.poliba.it/index.php/MultimediaCC> Tsutomu Murase, Hideyuki Shimonishi, Yohei Hasegawa, “Proposal of TCP Overlay Network,” IEICE Communication Society Conference (B-7-49), Sept. 2002. Squid (Software), [online], [Searched on November 10, 2014], Internet <URL: http://en.wikipedia.org/wiki/Squid_(%E3%82%BD%E3%83%95 % E3% 83% 88% E3% 82% A6% E3% 82% A7% E3% 82% A2) ＞

  According to the techniques described in Non-Patent Documents 1 and 2, in order to measure the TCP throughput, a test packet for measurement must be transmitted to the network separately from the user data communication. Therefore, a high load is applied to the communication band of the network.

  According to the technique described in Non-Patent Document 3, the available bandwidth can be estimated with a smaller amount of test packets than the techniques described in Non-Patent Documents 1 and 2. Since the load is not actually applied to the network for measurement, the accuracy of the estimation result is relatively low. In particular, although it is a smaller amount than the techniques described in Non-Patent Documents 1 and 2, it is necessary to allow the test packet to flow into the network.

  According to the techniques described in Non-Patent Documents 4 and 5, the network load due to the measurement test packet does not occur in order to estimate the available bandwidth from the data transfer rate actually used by the application. However, since actual user data is used, if the transfer rate is erroneously set higher than the available bandwidth, the communication bandwidth of the application itself is degraded. For this reason, it is easy for the target application to adjust the transfer rate in the direction of decreasing the transfer rate, but it is difficult to adjust the transfer rate in the direction of increasing the transfer rate, and the communication band cannot be accurately measured. According to this technique, it is necessary to activate the target application in advance in order to measure the communication band.

  Note that the techniques described in Non-Patent Documents 6 and 7 are mainly aimed at speeding up user communication and improving security, and do not implement an active measurement function of a communication band.

When a portable terminal such as a smartphone is assumed, generally, the data communication amount that can be used with a one-month contract with a communication carrier is limited in advance, or the usage fee increases according to the data communication amount. For this reason, transferring a test packet for measurement using such a portable terminal is costly disadvantageous for the user (erosion of contract fee) and is not actually possible.
Further, when user data is applied to an active measurement method having a low data transfer rate (a small amount of data transfer per measurement) as in the technique described in Non-Patent Document 3, arrival of user data at a reception difference may occur. This greatly delays the communication bandwidth experienced by the user.

  Therefore, the present invention provides a program, a server, a system, and a method capable of measuring a network communication band using user data even when the amount of data communication used by the user varies greatly. Objective.

According to the present invention, a program for causing a computer mounted on a server to function so as to be able to measure a network communication band by the terminal using user data returned to the terminal,
A first step of buffering user data to be returned in a predetermined unit time and determining whether the buffer amount is equal to or less than a predetermined threshold;
If it is determined to be true by the first step, the user data is divided, each divided data is formed into a UDP (User Datagram Protocol) packet, and each UDP packet is transmitted based on the available bandwidth estimation. If it is determined to be false in this step, the user data is configured as a TCP (Transmission Control Protocol) packet, and the second step of transmitting each TCP packet based on the throughput measurement is executed.

According to another embodiment of the server program of the present invention,
Regarding the second step, it is also preferable that when the server constructs a UDP packet or a TCP packet, it is executed so that pseudo bytes are added to the user data and padded to the specified data size for each packet.

According to another embodiment of the server program of the present invention,
For the first step, the server activates a standby time timer each time the user data packet to be returned is buffered, and when the standby time timer exceeds a predetermined time value, the determination result is set to true and the second step is performed. It is also preferable to execute the steps.

According to another embodiment of the server program of the present invention,
About a 1st step, it is also preferable that a server is performed so that a packet with a high transmission priority may not be buffered in user data.

According to another embodiment of the server program of the present invention,
User data is based on HTTP (HyperText Transport Protocol),
Regarding the first step, it is also preferable that the server execute user data based on HTML (HyperText Markup Language) text so as not to be buffered as a packet having a high transmission priority.

According to the present invention, there is provided a program for a terminal that causes a computer mounted on a “terminal” to communicate with a server that executes the above-described server program by a computer,
When the terminal receives user data of the UDP packet, the terminal measures the transmission rate based on the available bandwidth estimation, and when receiving the user data of the TCP packet, the terminal executes the transmission rate based on the throughput measurement. It is characterized by.

According to the present invention, a server that transmits user data sent back to a terminal so that the communication bandwidth of the network can be measured by the terminal,
Buffer determination means for buffering user data to be returned in a predetermined unit time, and determining whether the buffer amount is equal to or less than a predetermined threshold;
When it is determined to be true by the buffer determination unit, the user data is divided, each divided data is configured into a UDP packet, and each UDP packet is transmitted based on the available band estimation; and
When it is determined that the buffer determination means is false, it has a throughput measurement transmission means for configuring user data into TCP packets and transmitting each TCP packet based on the throughput measurement.

According to the present invention, there is provided a system having the above-described server and a terminal that communicates with the server,
The terminal
When user data of a UDP packet is received, an available bandwidth estimation unit that measures a transmission rate based on the available bandwidth estimation;
It is also preferable to have a throughput measuring means for measuring the transmission rate based on the throughput measurement when the user data of the TCP packet is received.

According to another embodiment of the system of the present invention,
The server is a relay server,
The terminal sends a content acquisition request to the relay server,
The relay server sends a content acquisition request to the Web server,
The Web server returns user data based on the content response to the relay server,
The relay server uses the user data of the content response sent back to the terminal, and transmits it so that the communication band of the network can be measured by the terminal.

According to the present invention, in a method for measuring a network communication band using user data returned from a server to a terminal,
A server for buffering user data to be returned in a predetermined unit time, and determining whether the buffer amount is equal to or less than a predetermined threshold;
If the server determines true in the first step, the server divides the user data, configures each divided data into UDP packets, transmits each UDP packet based on the available bandwidth estimation, and the first step A second step of configuring user data into TCP packets and transmitting each TCP packet based on throughput measurement;
The terminal measures the transmission rate based on the available bandwidth estimation when receiving the user data of the UDP packet, and measures the transmission rate based on the throughput measurement when receiving the user data of the TCP packet; It is characterized by having.

  According to the program, server, system, and method of the present invention, the network communication band can be measured using user data even when the amount of data communication used by the user varies greatly.

It is a sequence diagram between a server and a terminal in the present invention. It is a sequence diagram showing the difference of an available band estimation and a throughput estimation. It is a sequence diagram between the server-relay server-terminal in this invention. It is a functional block diagram of the server in this invention.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

  FIG. 1 is a sequence diagram between a server and a terminal in the present invention.

  According to FIG. 1, the server 1 and the terminal 2 are connected via a network. It is assumed that a TCP connection is established between the server 1 and the terminal 2 and that UDP stream communication is possible.

First, the terminal 2 transmits an HTTP GET request (content request) to the server 1 using a TCP connection (which may be a UDP packet).
Next, the server 1 returns user data to be returned by an HTTP GET response corresponding to the HTTP GET request. At this time, the server 1 transmits user data to be returned to the terminal 2 so that the terminal 2 can measure the communication band of the network.
On the other hand, the terminal 2 measures the communication band of the network using user data continuously received from the server 1.

According to the present invention, the server 1 returns user data to the terminal 2 in the following two steps.
[Step 1: S1] The server 1 buffers user data (HTTP content) to be returned for a predetermined unit time, and determines whether the buffer amount is equal to or less than a predetermined threshold (true / false).
[Step 2: S2]
S21: The server 1 divides user data, configures each divided data into a UDP packet, and determines each based on “Available Bandwidth Estimation” when it is determined to be true by S1 (buffer amount is equal to or less than a predetermined threshold). Send a UDP packet.
S22: Conversely, if the server 1 determines that it is false by S1 (the buffer amount is larger than a predetermined threshold), the server 1 configures user data into TCP packets, and transmits each TCP packet based on “throughput measurement”. .

On the other hand, the terminal 2 operates as follows.
S21: When user data of a UDP packet is received, the transmission rate is measured based on “available bandwidth estimation”.
S22: When user data of a TCP packet is received, the transmission rate is measured based on “throughput measurement”.
The payload of the packet used for measurement includes user data. Therefore, the terminal 2 outputs these user data to the application. If the user data is HTTP content, it is output to a web browser.

  FIG. 2 is a sequence diagram showing the difference between the available bandwidth estimation and the throughput estimation.

  According to the present invention, “usable bandwidth estimation” based on estimation and “throughput measurement” with relatively high accuracy are executed while adaptively switching according to the communication amount of user data in a predetermined unit unit time. .

[S21: Available Bandwidth Estimation]
When the buffer amount for a predetermined unit time is small (when the user data to be transferred is small), the delay time is shortened by the UDP packet in consideration of the influence on the user's experience, and the communication band is set by “available bandwidth estimation”. presume.

  “Available bandwidth estimation” refers to a technique of transmitting a test packet after adjusting its transmission interval to be exponentially shortened, and typically includes pathChirp (see Non-Patent Document 3, for example). According to the present invention, a packet including user data is transmitted instead of a test packet. The payload of the packet to be transmitted includes time information on the transmission timing of each packet. Further, the packet based on the available bandwidth estimation is composed of UDP because continuous real-time transfer is required. The terminal 2 that receives the continuous UDP packet records the reception time. The terminal 2 analyzes the fluctuation of the delay time that is the difference between the transmission time and the reception time for each packet, and estimates the available bandwidth.

  Here, pathChirp will be specifically described. A packet pair is composed of adjacent packets in the train. And it transmits, decreasing the probe transfer interval exponentially. A phenomenon is detected in which the variation in delay time increases when the packet transmission rate exceeds the available bandwidth. At first, since the transmission interval is wide, there is no fluctuation in the delay time. Thereafter, when the transmission interval between packets becomes narrower, the variation in delay time starts to increase from a certain pair. The available bandwidth is estimated based on the packet transmission rate when the delay time starts to increase.

[S22: Throughput measurement]
When the buffer amount for a predetermined unit time is large (when there is a lot of user data to be transferred), the payload is lengthened by the TCP packet, and the communication band is measured with high accuracy by “throughput measurement”.

“Throughput measurement” calculates the TCP throughput as follows based on the data amount of packets received by the receiving terminal per unit time.
TCP throughput = reception data amount / delay time delay time = reception time-transmission time The delay time is preferably an average value or a maximum value based on a plurality of packets.

<Embodiment 1: Use of pseudo-padding for packet>
Regarding S21 and S22, when constructing a UDP packet or a TCP packet, the server 1 “paddings” by embedding pseudo bytes up to the designated data size for each packet. That is, when the user data to be embedded in one packet does not reach the specified data size, the remaining portion is embedded with, for example, a null byte string. The designated data size may be designated by a UDP packet or a TCP packet. It is configured in the same way as a test packet using user data.

<Embodiment 2: Use of Wait Time Timer for Buffer>
As for S1, the server 1 activates a “waiting time timer” (buffer restart) every time the user data packet to be returned is buffered. When the standby time timer becomes equal to or greater than a predetermined time value, S21 is executed with the determination result as true. That is, “available bandwidth estimation” is executed. As a result, an increase in the arrival delay of the user data in the measurement of the communication band can be suppressed with respect to the experience of the user who operates the terminal.

<Embodiment 3: Exclude buffer of high priority packet>
About S1, the server 1 is made to perform so that a packet with a high transmission priority may not be buffered in user data. The priority is predetermined according to the characteristics of user data.

  Here, it is assumed that the user data is based on HTTP. At this time, for S1, the server 1 executes user data based on the HTML text so as not to be buffered as a packet having a high transmission priority. The HTML text as a response is user data that the terminal 2 first downloads from the Web server 1. By not buffering this packet as a packet having a high transmission priority, the delay of display on the Web browser of the terminal visually recognized by the user can be minimized.

The reason is that general HTML text is based on a small data size. Therefore, when HTML text is buffered, there is a high possibility that transmission of user data will stop until a predetermined timeout without accumulating up to a predetermined threshold of the buffer. The terminal 2 does not start receiving the next content data unless the HTML text arrives. Such a transfer stop of HTML text leads to a delay in display on the Web browser of the terminal visually recognized by the user.
On the other hand, content data other than HTML text has a large data size, and therefore is stored in a short time up to a predetermined threshold value of the buffer. According to this embodiment, it is attempted to measure the communication band using large user data such as content without measuring the communication band using small user data such as HTML text.

  In order to determine whether the text is HTML text or other content data, it is also preferable to refer to, for example, a Content-Type entity header field included in the HTML header. This field is generally given text / html header information in the case of HTML text, and header information such as image / jpeg or video / mpeg in the case of content data such as an image or video. .

  As another embodiment, it is also preferable to use a TOS (Type Of Service) value of the IP header in addition to the HTML text as a method for selecting whether or not the buffer is available. The TOS value (8 bits) is used for the purpose of indicating the transfer priority based on the type of communication service.

  FIG. 3 is a sequence diagram between a server, a relay server, and a terminal in the present invention.

  According to FIG. 3, compared with FIG. 1, the relay server 1 that is a feature of the present invention is arranged between the existing Web server (application server) and the terminal 2. According to the system of FIG. 3, it is possible to measure the communication band for any Web server as viewed from the user terminal 2 without mounting the function of the present invention on the Web server. Further, the server function of the present invention can be realized via the relay server 1 without being implemented in an existing Web server.

Here, there are the following two reasons for installing the present invention in the relay server.
First, it is the section of the radio link that requires the most measurement of the communication band. It is preferable to measure the communication band between the relay server and the terminal via a wireless link.
Secondly, in order to measure a communication band with any Web server, it is preferable to arrange a relay server in the communication carrier equipment.

The relay server 1 establishes a TCP connection with the terminal and a TCP connection with the Web server separately.
(S31) The terminal 2 adds a UDP or TCP header to the HTTP GET request to be transmitted to the Web server, and transmits a packet specifying the relay server 1 as the destination IP address to the relay server 1.
(S32) Next, the relay server 1 removes the UDP or TCP header of the received packet and extracts an HTTP GET request. Information such as the IP address of the communication destination Web server is acquired from the HTTP GET request. Then, the relay server 1 adds a UDP or TCP header to the HTTP GET request and transmits a packet specifying the Web server as the destination IP address to the Web server.
(S33) On the other hand, the Web server removes the UDP or TCP header of the received packet and extracts an HTTP GET request. Then, the Web server adds a UDP or TCP header to the HTML text and content data packet corresponding to the request, and transmits the packet specifying the relay server 1 as the destination IP address to the relay server 1.

  Then, the relay server 1 inputs the packet received from the Web server to the buffer of the present invention. Thereafter, the relay server 1 executes the processes of S1 and S2 described above. Note that the relay server 1 may perform communication band measurement only for the HTTP GET response. In that case, packets may be distinguished using an iptables function implemented in an OS (for example, Linux (registered trademark) or Android (registered trademark)).

  As a result, the terminal 2 can measure the communication band simultaneously with the reception of user data from the relay server 1.

  FIG. 4 is a functional configuration diagram of the server in the present invention.

  The server in FIG. 4 may be the server 1 in FIG. 1 or the relay server 1 in FIG. According to FIG. 4, the server 1 includes a buffer determination unit 11, an available bandwidth estimation transmission unit 12, and a throughput measurement unit 13. These functional components are realized by executing a program that causes a computer mounted on the server to function. The processing flow of these functional components can also be understood as a packet transfer method.

The buffer determination unit 11 buffers user data to be returned for a predetermined unit time, and determines whether the buffer amount is equal to or less than a predetermined threshold (see S1 in FIG. 1).
The usable bandwidth estimation transmission unit 12 divides user data when the buffer determination unit 11 determines that the buffer is true (the buffer amount is equal to or less than a predetermined threshold), configures each divided data into a UDP packet, and estimates the usable bandwidth. Based on this, each UDP packet is transmitted (see S21 in FIG. 1).
If the buffer determination unit 11 determines that the buffer determination unit 11 is false (the buffer amount is greater than the predetermined threshold), the throughput measurement transmission unit 13 configures user data into TCP packets and transmits each TCP packet based on the throughput measurement ( (See S22 in FIG. 1).

  On the other hand, when the user data of the UDP packet is received, the terminal 2 measures the transmission rate based on the available bandwidth estimation. When the terminal 2 receives the user data of the TCP packet, the terminal 2 measures the transmission rate based on the throughput measurement. . For example, user data of HTTP response is output to the Web browser.

  As described above in detail, according to the program, server, system, and method of the present invention, even if the amount of data communication used by the user fluctuates greatly, the communication bandwidth of the network can be reduced using user data. It can be measured. In order to use user data, it is not necessary to transfer a specific test packet, and the amount of data communication contracted for each user is not eroded. In addition, when the amount of user data to be transferred is small, in order to measure the communication bandwidth by estimating the available bandwidth configured in the UDP packet, it is possible to reduce the arrival delay of the user data without using TCP with a long payload and window control. Can be small.

  Various changes, modifications, and omissions of the above-described various embodiments of the present invention can be easily made by those skilled in the art. The above description is merely an example, and is not intended to be restrictive. The invention is limited only as defined in the following claims and the equivalents thereto.

DESCRIPTION OF SYMBOLS 1 Server, relay server 11 Buffer determination part 12 Available bandwidth estimation transmission part 13 Throughput measurement transmission part 2 Terminal 3 Web server

Claims (10)

A program that causes a computer mounted on a server to function so that the communication bandwidth of the network can be measured by the terminal using user data returned to the terminal,
A first step of buffering user data to be returned in a predetermined unit time and determining whether the buffer amount is equal to or less than a predetermined threshold;
If it is determined to be true by the first step, the user data is divided, each divided data is formed into a UDP (User Datagram Protocol) packet, and each UDP packet is transmitted based on the available bandwidth estimation. For the server, the user data is configured in a TCP (Transmission Control Protocol) packet and the second step of transmitting each TCP packet based on the throughput measurement is executed when it is determined to be false in the step Program. Regarding the second step, when the server constructs a UDP packet or a TCP packet, the server performs execution so as to add up to the specified data size for each packet and fill the pseudo bytes with padding. The server program according to claim 1. With respect to the first step, the server activates a standby time timer each time a user data packet to be returned is buffered, and when the standby time timer exceeds a predetermined time value, the server sets the determination result to true. The server program according to claim 1 or 2, wherein the steps are executed. 4. The server for the server according to claim 1, wherein the server executes the server so that a packet having a high transmission priority is not buffered in the user data. 5. program. The user data is based on HTTP (HyperText Transport Protocol),
5. The server according to claim 4, wherein the server causes the user data based on HTML (HyperText Markup Language) text to be executed so as not to be buffered as a packet having a high transmission priority. 6. Program. A program for a terminal that causes a computer mounted on a "terminal" that communicates with a server that executes the program for a server according to any one of claims 1 to 5 to be executed by a computer,
When receiving the user data of the UDP packet, the terminal measures the transmission rate based on the available bandwidth estimation. When receiving the user data of the TCP packet, the terminal executes the transmission rate based on the throughput measurement. A program for a terminal characterized by the above. A server that uses user data sent back to the terminal to transmit the network communication band so that the terminal can measure the network,
Buffer determination means for buffering user data to be returned in a predetermined unit time, and determining whether the buffer amount is equal to or less than a predetermined threshold;
When it is determined to be true by the buffer determination unit, the user data is divided, each divided data is configured into a UDP packet, and each UDP packet is transmitted based on the available band estimation; and
And a throughput measurement transmission unit configured to configure user data in a TCP packet and transmit each TCP packet based on the throughput measurement when the buffer determination unit determines to be false. A system comprising the server according to claim 7 and a terminal that communicates with the server,
The terminal
When user data of a UDP packet is received, an available bandwidth estimation unit that measures a transmission rate based on the available bandwidth estimation;
A system comprising throughput measuring means for measuring a transmission rate based on throughput measurement when user data of a TCP packet is received. The server is a relay server,
The terminal transmits a content acquisition request to the relay server,
The relay server transmits the content acquisition request to a Web server;
The Web server returns user data based on the content response to the relay server,
9. The system according to claim 8, wherein the relay server transmits the content response user data sent back to the terminal so that the network communication band can be measured by the terminal. In the method of measuring the communication bandwidth of the network using user data returned from the server to the terminal,
A first step in which the server buffers user data to be returned in a predetermined unit time, and determines whether or not the buffer amount is equal to or less than a predetermined threshold;
If the server determines true in the first step, the server divides user data, configures each divided data into UDP packets, transmits each UDP packet based on the available bandwidth estimation, A second step of configuring user data into TCP packets and transmitting each TCP packet based on throughput measurement if determined to be false in the step;
The terminal measures the transmission rate based on the available bandwidth estimation when receiving user data of the UDP packet, and measures the transmission rate based on throughput measurement when receiving the user data of the TCP packet. A method characterized by comprising:

Images