WO2011150651A1

WO2011150651A1 - Congestion control method and device for data transmission

Info

Publication number: WO2011150651A1
Application number: PCT/CN2010/080436
Authority: WO
Inventors: 肖文曙; 尹月静; 许志勇
Original assignee: 华为技术有限公司
Priority date: 2010-12-29
Filing date: 2010-12-29
Publication date: 2011-12-08
Also published as: CN102204182B; CN102204182A

Abstract

A congestion control method for data transmission is provided. The packet loss rate and the change of the packet loss rate are counted during the data transmission, and the congestion control is performed by using the packet loss rate and the change of the packet loss rate as adjustment factors. In the embodiments of the present invention, a congestion control device for data transmission is also provided. By using the packet loss rate and the change of the packet loss rate as adjustment factors of the congestion control, the embodiments of the present invention enable the data transmission rate to rapidly converge at the vicinity of the best value matching with the network available bandwidth, and thus the effective bandwidth is utilized sufficiently.

Description

Congestion control method and device for data transmission

Technical field

The present invention relates to the field of network communication technologies, and in particular, to a congestion control method and apparatus for data transmission.

Background of the invention

In the IP network, TCP (Transport Control Protocol) or UDP (User The Datagram Protocol, a user datagram protocol, is widely used as a standard transport protocol.

Among them, TCP is a reliable transmission protocol, using congestion control mechanism to ensure reliable transmission of data packets, while ensuring the fairness of network resource usage. The focus of the TCP congestion control algorithm is on the window control management. From the congestion control algorithm that has been disclosed, the congestion control algorithm can be divided into: 1) by the congestion control algorithm. The congestion control algorithm with the adjustment factor is used to improve the traditional TCP for the packet loss information of the ACK feedback. The representative algorithms are Reno, Highspeed, Bic, Cubic and SCTP. 2) The congestion control algorithm with delay as the adjustment factor. Taking RTT delay as the adjustment factor, the representative algorithms are: Vegas, FastTCP; 3) congestion control algorithm with packet loss + delay as the adjustment factor, representing algorithm: Ilinois. The core of various algorithms with packet loss as the adjustment factor is that after the packet loss is perceived, the network is considered to be congested, and then various window adjustment mechanisms are adopted, the difference being mainly in the amplitude of the adjustment.

TCP For example, reno is the process of the TCP congestion control algorithm. As shown in Figure 1, the slow start mechanism is used at the beginning of the transmission. During the slow start, the TCP sender first sends a datagram, then waits for the response from the other party and gets a response. After that, the size of the send window is doubled, and then two datagrams are sent continuously, and after the other party responds, the send window is doubled. Until the congestion window cwnd exceeds the slow start threshold ssthresh, the slow start process is stopped and the congestion avoidance phase is entered. The algorithm for performing the congestion avoidance phase in the congestion avoidance phase is specifically: when all the messages in the transmission window are acknowledged, cwnd adds a segment. The algorithm for this congestion avoidance phase is maintained until congestion is detected. Congestion detection includes two cases: one is that the TCP sender timer expires, that is, the packet sent by the sender does not receive the acknowledgement from the receiver at the specified time, then ssthresh is adjusted to half of the current congestion window, and the congestion window is added. Adjusted to 1, And restart slowly. The other case is to receive 3 duplicate ACKs, then adjust ssthresh to half of the current congestion window and adjust the congestion window to the adjusted ssthresh.

Existing TCP-based congestion control algorithms have at least the following disadvantages:

When a packet loss occurs on the network, it does not necessarily mean that the network is congested. For example, in a high-bandwidth delay product network, there are various non-congestion factors, such as link state, error, and the like, and in such a network environment, a congestion algorithm using packet loss as an adjustment factor is used as long as the loss occurs. The packet is considered to be congested on the network, which will reduce the transmission rate; if the timer expires, the slow start threshold will be lowered and restarted from the slow start, so that the effective bandwidth cannot be fully utilized. The road is fatal.

For the congestion algorithm with delay as the adjustment factor, due to the variability of the router cache and the dependence of the RTT delay sampling rate, the increasing delay and packet loss in the network do not directly correspond, resulting in delay as the adjustment factor. The congestion algorithm cannot control the window accurately.

For the connectionless UDP protocol, no transmission reliability is considered, and there is no end-to-end congestion control mechanism. Even if the network sends a congestion indication, UDP does not reduce the amount of data sent to the network. However, since the transmission efficiency of UDP is much higher than that of the TCP protocol, congestion control is introduced based on UDP to provide an efficient and reliable transmission scheme. Among them, UDT (UDP-based The Data Transfer Protocol (UDP-based data transfer protocol) is a protocol that introduces congestion control mechanisms based on UDP.

UDT introduces new congestion control and data reliability control mechanisms. Congestion control mechanism in UDT adopts progressively increasing AIMD (additive Increase Multiplicative-decrease, and increase the product type): The rate control is used to adjust the transmission interval; supplemented by the window control, the window size is dynamically adjusted to limit the number of unresponsive packets.

The main disadvantage of UDT is that packet loss is still used as an adjustment factor for congestion control. Therefore, in a high bandwidth delay product network, there is also a problem that the effective bandwidth cannot be fully utilized.

In summary, the existing congestion control mechanism for data transmission cannot fully utilize the available bandwidth in a high bandwidth delay product network.

Summary of the invention

Embodiments of the present invention provide a congestion control method and apparatus for data transmission, which can fully utilize available bandwidth.

The embodiments of the present invention are implemented by the following technical solutions:

An embodiment of the present invention provides a congestion control method for data transmission, including:

Statistics of packet loss rate and packet loss rate during data transmission;

Congestion control is performed by using the change of the packet loss rate and the packet loss rate as an adjustment factor.

An embodiment of the present invention provides a congestion control apparatus for data transmission, including:

Data transmission unit for data transmission;

a statistical unit, configured to count changes in packet loss rate and packet loss rate during data transmission;

The congestion control unit is configured to perform congestion control by using the change of the packet loss rate and the packet loss rate as an adjustment factor.

It can be seen from the technical solution provided by the foregoing embodiments of the present invention that the change of the packet loss rate and the packet loss rate is used as the congestion control adjustment factor, so that the data transmission rate quickly converges to the optimal value matching the available bandwidth of the network. Nearby to take advantage of effective bandwidth.

BRIEF DESCRIPTION OF THE DRAWINGS

1 is a schematic diagram of an execution process of a prior art TCP congestion control algorithm;

2 is a flowchart of a congestion control method for data transmission according to an embodiment of the present invention;

3 is a flowchart of an application scenario operation of a congestion control method for data transmission according to an embodiment of the present invention;

4 is a flowchart of another application scenario operation of a congestion control method for data transmission according to an embodiment of the present invention;

FIG. 5 is a schematic structural diagram of a congestion control apparatus for data transmission according to an embodiment of the present invention.

Mode for carrying out the invention

The technical solutions in the embodiments of the present invention are clearly and completely described in the following with reference to the accompanying drawings in the embodiments of the present invention. It is obvious that the described embodiments are only a part of the embodiments of the present invention, but not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative efforts are within the scope of the present invention.

An embodiment of the present invention provides a congestion control method for data transmission. As shown in FIG. 2, the method includes the following steps:

Step 20: Calculate the change of the packet loss rate and the packet loss rate during the data transmission process;

The data transmission in the embodiment of the present invention includes data transmission using UDP and data transmission using TCP.

Step 21: Perform congestion control by using the change of the packet loss rate and the packet loss rate as an adjustment factor.

The congestion control using the change of the packet loss rate and the packet loss rate as an adjustment factor includes:

1) Comparing the packet loss rate with the set upper threshold and the lower threshold, and adjusting the data transmission rate according to the comparison result; the upper threshold and the lower threshold described in the embodiment of the present invention may be obtained according to an estimation of the network link status. Specifically include:

If the packet loss rate is greater than the upper threshold, the data transmission rate is decreased, and the adjusted data transmission rate is a product of the pre-adjustment data transmission rate and the first adjustment factor, and the first adjustment factor=1.0-packet rate+first step The long factor w1; the step factor here and the step factor mentioned below is an influence factor acting on the rate adjustment amplitude, which is based on environmental conditions and experience.

If the packet loss rate is between the upper threshold and the lower threshold, it is determined whether the change of the packet loss rate is rising or decreasing; if the packet loss rate is increased, the data transmission rate is lowered, and the adjusted data transmission rate is the data transmission rate before the adjustment. The product of the second adjustment factor, the second adjustment factor = 1.0 - the packet loss rate + the second step factor w2; if the packet loss rate is decreased, the data transmission rate is adjusted, and the adjusted data transmission rate is the data transmission before the adjustment. The product of the rate and the reciprocal of the third adjustment factor, the third adjustment factor = 1.0 - the packet loss rate - the third step factor w3; it can be understood that if the packet loss rate is measured for the first time, the packet loss rate is determined. When the upper threshold is lower than the lower threshold, the rate adjustment operation is not performed first, and the packet loss rate is continuously counted, so as to determine the change of the packet loss rate and then perform the rate adjustment operation.

If the packet loss rate is lower than the lower threshold, the data transmission rate is adjusted, and the adjusted data transmission rate is the product of the pre-adjustment data transmission rate and the reciprocal of the fourth adjustment factor, and the fourth adjustment factor=1.0-packet rate- Four step factor w4.

It can be seen from the foregoing adjustment process that the embodiment of the present invention adjusts the data transmission rate according to the change of the packet loss rate and the packet loss rate. When the packet loss rate is between the upper threshold and the lower threshold, the direction of the packet loss rate changes. The direction of data transmission rate adjustment, the increase of the packet loss rate indicates that the data transmission rate is too large compared to the available bandwidth matching value, and the packet loss rate decreases, indicating that the data transmission rate is less than the available bandwidth matching value, when the rate is near the value matching the available bandwidth. The packet loss rate will tend to be stable. When the packet loss rate and the packet loss rate are different, the data transmission rate is adjusted differently, so that the adjusted data transmission rate can gradually converge to a value that matches the available bandwidth, thereby realizing the full utilization of the available bandwidth.

2) Comparing the packet loss rate with the set upper threshold and the lower threshold, and adjusting the slow start threshold and the congestion window according to the comparison result, specifically including:

If the packet loss rate is greater than the upper threshold, modify the slow start threshold and enter the slow start phase. The modified slow start threshold is the product of the current congestion window size and the first adjustment value, and the congestion window is reduced to one segment; That is, when the packet loss rate is greater than the upper threshold, whether the slow start phase or the congestion avoidance phase is performed, the operation of modifying the slow start threshold is performed, and then the slow start phase is re-entered; wherein the first adjustment value is The size is determined by the packet loss rate, for example, the first adjustment value = 1.0 - packet loss rate + fifth step factor w5.

If the packet loss rate is between the upper threshold and the lower threshold, enter the congestion avoidance phase to determine whether the packet loss rate changes or decreases. If the packet loss rate increases, modify the slow start threshold and re-enter congestion avoidance. The slow start threshold is the product of the current congestion window size and the second adjustment value, and reduces the congestion window to a slow start threshold (a reduced slow start threshold), and the second adjustment value is determined by the packet loss rate and the sixth The step factor determines, for example, the second adjustment value = 1.0 - the packet loss rate + the sixth step factor w6; if the packet loss rate decreases, the congestion avoidance is continued without changing the slow start threshold and the congestion window size; that is, if the packet loss is lost; If the packet rate is between the upper threshold and the lower threshold, if it is in the slow start phase, it immediately enters the congestion avoidance phase. If it is in the congestion avoidance phase, the congestion avoidance process is continued, and the packet loss rate is directly changed. Judge. It can be understood that if the packet loss rate is determined to be between the upper threshold and the lower threshold, the slow start threshold and the congestion window size adjustment operation are not performed first, but the packet loss is continued. Rate, thereby determining the change of the packet loss rate and then performing the slow start threshold and the congestion window size adjustment operation.

If the packet loss rate is lower than the lower threshold, the slow start is continued in the slow start phase. If the slow start threshold and the congestion window are not changed during the congestion avoidance phase, the congestion avoidance is continued.

The method in the embodiment of the present invention can be applied to a high bandwidth delay product network, but is not limited to a high bandwidth delay product network.

In the embodiment of the present invention, the slow start threshold and the congestion window are adjusted according to the change of the packet loss rate and the packet loss rate, and the slow start threshold and the congestion window are adjusted when different packet loss rates and packet loss rates are different. Different from the adjustment range, the adjusted data transmission rate converges to the vicinity of the optimal value matching the available bandwidth, so as to make full use of the available bandwidth.

In order to further understand the above embodiments, the method of the foregoing embodiment is described in detail below by taking the above method as a specific example.

Scenario 1: Data transmission through UDP, TCP transmission signaling; as shown in FIG. 3, the congestion control method in the data transmission and data transmission process specifically includes the following steps:

Step 30: The transmitting end transmits the first block of data through UDP.

In this embodiment, the data is transmitted and fed back in units of blocks, and each piece of data includes multiple data packets, and the fixed size of the data packet is the maximum MTU of the network (Maximum Transmission) Unit, maximum transfer unit), this block-based data transfer method avoids complex sliding window control operations based on each packet.

Step 31: The transmitting end completes signaling through TCP transmission, and notifies the receiving end that the first block of data has been sent.

Step 32: The receiving end returns ACK signaling, which includes a view of the correctly received data packet.

If it is determined according to the view that there is a lost data packet, step 33 is performed;

Step 33: The sender retransmits the lost data packet through UDP.

Step 34: The transmitting end completes the signaling by using the TCP transmission, and notifies the receiving end that the lost data packet has been sent.

Step 35: The receiving end returns ACK signaling, which includes a view of the data correctly received after the last retransmission;

If it is determined according to the display in the view that there is still a lost data packet, proceed to step 36;

Step 36: The transmitting end retransmits the still lost data packet until the view shows that the first block of data is all correctly received by the receiving end;

Step 37: Statistics the packet loss rate, and adjust the data transmission rate according to the packet loss rate.

In the embodiment of the present invention, the time when the piece of data is transmitted and all received correctly is used as the time unit of the statistical packet loss rate. The operation of the statistical packet loss rate starts from the start of sending the piece of data, and the specific statistical methods include but are not limited to: In a statistical time unit of packet loss rate, the proportion of lost data packets in each data packet is counted, and the average value of multiple calculation results is used as the packet loss rate in the statistical time; or, in a packet loss During the statistical time of the rate, the proportion of the lost data packets in the first transmission packet is taken as the packet loss rate during the statistical time.

The method for adjusting the data transmission rate according to the packet loss rate includes: comparing the packet loss rate with the set upper threshold value and the lower threshold value, in this embodiment, the upper threshold is set to 1%, and the lower threshold is 0.01%;

If the packet loss rate is greater than the upper threshold by 1%, the data transmission rate is lowered, and the adjusted data transmission rate is a product of the pre-adjustment data transmission rate and the first adjustment factor, and the first adjustment factor is 1.0-packet rate+ One step length factor w1;

If the packet loss rate is less than the lower threshold of 0.01%, the data transmission rate is adjusted, and the adjusted data transmission rate is the product of the pre-adjustment data transmission rate and the reciprocal of the fourth adjustment factor, and the fourth adjustment factor=1.0-loss rate - The fourth step factor w4.

If the packet loss rate is between the upper threshold and the lower threshold, the data loss rate is not adjusted temporarily because the packet loss rate is not detected for the first time.

Step 38: Complete the transmission of the second block of data according to the method of steps 30-36;

Step 39: The packet loss rate is measured from the second block of data, and the data transmission rate is adjusted according to the change of the packet loss rate and the packet loss rate.

The specific adjustment method is the same as that in step 37. When it is determined that the packet loss rate is between the upper threshold and the lower threshold, the packet loss rate of the current statistics and the packet loss rate of the last statistics (statistics in step 37) are determined. Whether the change is rising or decreasing; if the packet loss rate is increased, the data transmission rate is lowered, and the adjusted data transmission rate is the product of the pre-adjustment data transmission rate and the second adjustment factor, and the second adjustment factor=1.0-packet loss rate + second step factor w2; if the packet loss rate is decreased, the data transmission rate is adjusted, and the adjusted data transmission rate is the product of the pre-adjustment data transmission rate and the reciprocal of the third adjustment factor, the third adjustment factor=1.0 - packet loss rate - third step factor w3;

Step 310: Continue the next data block transmission.

Subsequently transmitted data blocks are sent at an adjusted rate.

In this embodiment, the data transmission rate is adjusted according to the change of the packet loss rate and the packet loss rate. When the packet loss rate and the packet loss rate change are different, the adjustment range of the data transmission rate is different, so that the adjusted data transmission rate can converge to The available bandwidth is matched to take advantage of the available bandwidth. In addition, the present embodiment transmits data in units of blocks to avoid complicated sliding window control operations based on each data packet.

Scenario 2: Transmitting data through TCP; as shown in FIG. 4, the congestion control method in the data transmission and data transmission process specifically includes the following steps:

Step 40: The transmitting end transmits data through TCP, and undergoes a slow start phase;

Step 41: After the congestion window reaches the slow start threshold, enter the congestion avoidance phase.

Step 42: In the congestion avoidance phase, when all the packets in the congestion window are acknowledged, the congestion window adds a segment;

Step 43: Statistics of packet loss rate and packet loss rate change;

In this embodiment, starting from the transmission of data, the packet loss rate is counted in units of a fixed time, that is, the number of times of retransmitting the data packet in the fixed time is counted as a numerator, and the number of times of retransmission of the data packet is received and received. The number of ACKs obtained is used as the denominator, and the score obtained is the packet loss rate during the fixed time period;

Step 44: Adjust the slow start threshold and the congestion window according to the packet loss rate and the packet loss rate change.

The specific adjustment strategy is as follows:

If the packet loss rate is greater than the upper threshold, whether it is in the slow start phase or the congestion avoidance phase, modify the slow start threshold and re-enter the slow start phase. The modified slow start threshold is the current congestion window size and the first adjustment value. The product is reduced, and the congestion window is reduced to one segment; the size of the first adjustment value is determined by the packet loss rate and the fifth step factor.

If the packet loss rate is between the upper threshold and the lower threshold, enter the congestion avoidance phase to determine whether the packet loss rate changes or decreases. If the packet loss rate increases, modify the slow start threshold and re-enter congestion avoidance. The slow start threshold is the product of the current congestion window size and the second adjustment value, and reduces the congestion window to a slow start threshold (a reduced slow start threshold), and the second adjustment value is determined by the packet loss rate and the sixth The step size factor is determined. If the packet loss rate is reduced, the congestion avoidance threshold and the congestion window size are not changed to continue to perform congestion avoidance. It can be understood that if the packet loss rate is measured for the first time, the packet loss rate is determined to be at the upper threshold and the lower limit. Between the thresholds, the slow start threshold and the congestion window size adjustment operation are not performed first, but the packet loss rate is continuously counted, thereby determining the change of the packet loss rate and then performing the slow start threshold and the congestion window size adjustment operation.

Step 45: Continue data transmission with the adjusted congestion window and the slow start threshold.

The invention provides a change of the packet loss rate and the packet loss rate as a congestion control adjustment factor, so that the data transmission rate quickly converges to the vicinity of the optimal value matching the available bandwidth of the network, thereby fully utilizing the effective bandwidth, and thus can be applied to high bandwidth. Delayed product and other network environments.

A further embodiment of the present invention provides a congestion control apparatus for data transmission. The congestion control apparatus for data transmission according to this embodiment is a device corresponding to the method described in the foregoing embodiment, and the method of the foregoing embodiment can be implemented by the apparatus. , as shown in Figure 5, including:

The data transmission unit 50 is for data transmission; the data transmission unit 50 can perform data transmission in units of blocks.

The statistics unit 51 is configured to count the change of the packet loss rate and the packet loss rate during the data transmission process;

The congestion control unit 52 is configured to perform congestion control by using the change of the packet loss rate and the packet loss rate as an adjustment factor. The congestion control unit 52 can further include:

The first adjustment subunit 521 (not shown) is configured to compare the packet loss rate with the set upper threshold and the lower threshold, and adjust the data transmission rate according to the comparison result. The specific adjustment method includes: comparing the packet loss rate with the set upper threshold and the lower threshold, and if the packet loss rate is greater than the upper threshold, the data transmission rate is lowered, and the adjusted data transmission rate is the pre-adjustment data transmission rate and the first adjustment. The product of the factor, the first adjustment factor is determined by the packet loss rate, specifically, the first adjustment factor = 1.0 - the packet loss rate + the first step factor w1; if the packet loss rate is between the upper threshold and the lower threshold And determining whether the change of the packet loss rate is rising or decreasing; if the packet loss rate is rising, the data transmission rate is lowered, and the adjusted data transmission rate is a product of the pre-adjustment data transmission rate and the second adjustment factor, the second adjustment Factor=1.0-packet loss rate+second step factor w2; if the packet loss rate decreases, the data transmission rate is adjusted, and the adjusted data transmission rate is the product of the pre-adjustment data transmission rate and the reciprocal of the third adjustment factor. The third adjustment factor=1.0-packet loss rate-third step factor w3; it can be understood that if the packet loss rate is measured for the first time, it is determined that the packet loss rate is at the upper threshold and Between the threshold limit, the rate adjustment is not performed before the operation, but continues statistical packet loss ratio, to determine a change in the rate of loss ratio and then performing adjustment operation. If the packet loss rate is lower than the lower threshold, the data transmission rate is adjusted, and the adjusted data transmission rate is the product of the pre-adjustment data transmission rate and the reciprocal of the fourth adjustment factor, and the fourth adjustment factor=1.0-packet rate- Four step factor w4.

Or the congestion control unit 52 further includes:

The second adjustment sub-unit 522 (not shown) is configured to compare the packet loss rate with the set upper and lower thresholds, and adjust the slow start threshold and the congestion window according to the comparison result. The specific adjustment method includes: comparing the packet loss rate with the set upper threshold and the lower threshold, and if the packet loss rate is greater than the upper threshold, modifying the slow start threshold and entering the slow start phase, the modified slow start threshold is the current congestion window size and The product of the first adjustment value, and reduces the congestion window to one segment; that is, when the packet loss rate is greater than the upper threshold, whether the slow start phase or the congestion avoidance phase is performed, the modified slow start is performed. The operation of the threshold, and then re-enter the slow start phase; wherein the first adjustment value = 1.0 - packet loss rate + fifth step factor. If the packet loss rate is between the upper threshold and the lower threshold, enter the congestion avoidance phase to determine whether the packet loss rate changes or decreases. If the packet loss rate increases, modify the slow start threshold and re-enter congestion avoidance. The slow start threshold is the product of the current congestion window size and the second adjustment value, and reduces the congestion window to a slow start threshold (a reduced slow start threshold), the second adjustment value = 1.0 - packet loss rate + sixth Step factor; if the packet loss rate is reduced, the congestion avoidance is continued without changing the slow start threshold and the congestion window size; that is, if the packet loss rate is between the upper threshold and the lower threshold, then if the packet is in the slow start phase, Then, the congestion avoidance phase is immediately entered. If the congestion avoidance phase is here, the congestion avoidance is continued, and the change of the packet loss rate is directly performed. It can be understood that if the packet loss rate is determined to be between the upper threshold and the lower threshold, the slow start threshold and the congestion window size adjustment operation are not performed first, but the packet loss is continued. Rate, thereby determining the change of the packet loss rate and then performing the slow start threshold and the congestion window size adjustment operation. If the packet loss rate is lower than the lower threshold, the slow start is continued in the slow start phase. If the slow start threshold and the congestion window are not changed during the congestion avoidance phase, the congestion avoidance is continued.

The device in this embodiment is disposed on the data sending end, and may be, but is not limited to, applied to a high bandwidth delay product network environment. It will be understood that in order to highlight the inventive aspects of the apparatus of the embodiments of the present invention, an introduction to the existing functions of the apparatus is omitted.

The congestion control device for data transmission provided by the embodiment of the present invention uses the change of the packet loss rate and the packet loss rate as a congestion control adjustment factor, so that the data transmission rate quickly converges to the vicinity of the optimal value matching the available bandwidth of the network, thereby fully utilizing the effective bandwidth.

In summary, the embodiment of the present invention uses the change of the packet loss rate and the packet loss rate as a congestion control adjustment factor, so that the data transmission rate quickly converges to the vicinity of the optimal value matched with the available bandwidth of the network, thereby fully utilizing the effective bandwidth.

It will be understood by those skilled in the art that all or part of the steps of the foregoing embodiments may be implemented by a program to instruct related hardware, and the program may be stored in a computer readable storage medium, and the program is executed. At the time, the flow of the embodiment of each of the above methods may be included. The storage medium may be a magnetic disk, an optical disk, or a read-only storage memory (Read-Only) Memory, ROM) or Random Access Memory (RAM).

The above is only a preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art can easily think of changes or within the technical scope disclosed by the present invention. Alternatives are intended to be covered by the scope of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

A congestion control method for data transmission, comprising:

Statistics of packet loss rate and packet loss rate during data transmission;

Congestion control is performed by using the change of the packet loss rate and the packet loss rate as an adjustment factor.
The method of claim 1 wherein said data transmission is data transmission in units of blocks.
The method according to claim 1 or 2, wherein the performing congestion control by using the change of the packet loss rate and the packet loss rate as an adjustment factor comprises:

The packet loss rate is compared with the set upper threshold and the lower threshold, and the data transmission rate is adjusted according to the comparison result.
The method of claim 3, wherein the adjusting the data transmission rate according to the comparison result comprises:

If the packet loss rate is greater than the upper threshold, the data transmission rate is decreased, and the adjusted data transmission rate is a product of the pre-adjustment data transmission rate and the first adjustment factor, and the first adjustment factor=1.0-packet rate+first step Long factor w1;

If the packet loss rate is between the upper threshold and the lower threshold, it is determined whether the change of the packet loss rate is rising or decreasing; if the packet loss rate is increased, the data transmission rate is lowered, and the adjusted data transmission rate is the data transmission rate before the adjustment. The product of the second adjustment factor, the second adjustment factor = 1.0 - the packet loss rate + the second step factor w2; if the packet loss rate is decreased, the data transmission rate is adjusted, and the adjusted data transmission rate is the data transmission before the adjustment. The product of the rate and the reciprocal of the third adjustment factor, the third adjustment factor = 1.0 - the packet loss rate - the third step factor w3;

If the packet loss rate is lower than the lower threshold, the data transmission rate is adjusted, and the adjusted data transmission rate is the product of the pre-adjustment data transmission rate and the reciprocal of the fourth adjustment factor, and the fourth adjustment factor=1.0-packet rate- Four step factor w3.
The method according to claim 1, wherein the performing congestion control by using the change of the packet loss rate and the packet loss rate as an adjustment factor comprises:

The packet loss rate is compared with the set upper threshold and the lower threshold, and the slow start threshold and the congestion window are adjusted according to the comparison result.
The method according to claim 5, wherein the adjusting the slow start threshold and the congestion window according to the comparison result comprises:

If the packet loss rate is greater than the upper threshold, modify the slow start threshold and enter the slow start phase. The modified slow start threshold is the product of the current congestion window size and the first adjustment value, and the congestion window is reduced to one segment. First adjustment value = 1.0 - packet loss rate + fifth step factor w5;

If the packet loss rate is between the upper threshold and the lower threshold, enter the congestion avoidance phase to determine whether the packet loss rate changes or decreases. If the packet loss rate increases, modify the slow start threshold, re-enter congestion avoidance, and slow down. The startup threshold is the product of the current congestion window size and the second adjustment value, and the congestion window is reduced to a slow start threshold, and the second adjustment value is 1.0-packet loss rate + the sixth step factor w6; if the packet loss rate is decreased, Then, continue to perform congestion avoidance;

If the packet loss rate is lower than the lower threshold, the slow start is continued in the slow start phase, and the congestion avoidance is performed in the congestion avoidance phase.
A congestion control device for data transmission, comprising:

Data transmission unit for data transmission;

a statistical unit, configured to count changes in packet loss rate and packet loss rate during data transmission;

The congestion control unit is configured to perform congestion control by using the change of the packet loss rate and the packet loss rate as an adjustment factor.
The apparatus according to claim 7, wherein said data transmission unit performs data transmission in units of blocks.
The device according to claim 7 or 8, wherein the congestion control unit further comprises:

The first adjusting subunit is configured to compare the packet loss rate with the set upper threshold and the lower threshold, and adjust the data transmission rate according to the comparison result.
The device of claim 7, wherein the congestion control unit further comprises:

The second adjusting subunit is configured to compare the packet loss rate with the set upper threshold and the lower threshold, and adjust the slow start threshold and the congestion window according to the comparison result.