CN113891172B - Adaptive code rate control method based on RTT (round trip time) and suitable for wireless Mesh network - Google Patents
Adaptive code rate control method based on RTT (round trip time) and suitable for wireless Mesh network Download PDFInfo
- Publication number
- CN113891172B CN113891172B CN202111023201.XA CN202111023201A CN113891172B CN 113891172 B CN113891172 B CN 113891172B CN 202111023201 A CN202111023201 A CN 202111023201A CN 113891172 B CN113891172 B CN 113891172B
- Authority
- CN
- China
- Prior art keywords
- network
- code rate
- rate
- value
- state
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/60—Network structure or processes for video distribution between server and client or between remote clients; Control signalling between clients, server and network components; Transmission of management data between server and client, e.g. sending from server to client commands for recording incoming content stream; Communication details between server and client
- H04N21/63—Control signaling related to video distribution between client, server and network components; Network processes for video distribution between server and clients or between remote clients, e.g. transmitting basic layer and enhancement layers over different transmission paths, setting up a peer-to-peer communication via Internet between remote STB's; Communication protocols; Addressing
- H04N21/647—Control signaling between network components and server or clients; Network processes for video distribution between server and clients, e.g. controlling the quality of the video stream, by dropping packets, protecting content from unauthorised alteration within the network, monitoring of network load, bridging between two different networks, e.g. between IP and wireless
- H04N21/64784—Data processing by the network
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/10—Flow control; Congestion control
- H04L47/28—Flow control; Congestion control in relation to timing considerations
- H04L47/283—Flow control; Congestion control in relation to timing considerations in response to processing delays, e.g. caused by jitter or round trip time [RTT]
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/60—Network structure or processes for video distribution between server and client or between remote clients; Control signalling between clients, server and network components; Transmission of management data between server and client, e.g. sending from server to client commands for recording incoming content stream; Communication details between server and client
- H04N21/63—Control signaling related to video distribution between client, server and network components; Network processes for video distribution between server and clients or between remote clients, e.g. transmitting basic layer and enhancement layers over different transmission paths, setting up a peer-to-peer communication via Internet between remote STB's; Communication protocols; Addressing
- H04N21/647—Control signaling between network components and server or clients; Network processes for video distribution between server and clients, e.g. controlling the quality of the video stream, by dropping packets, protecting content from unauthorised alteration within the network, monitoring of network load, bridging between two different networks, e.g. between IP and wireless
- H04N21/64723—Monitoring of network processes or resources, e.g. monitoring of network load
- H04N21/64738—Monitoring network characteristics, e.g. bandwidth, congestion level
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks
Abstract
The invention discloses a self-adaptive code rate control method based on RTT (round trip time) suitable for a wireless Mesh network, which comprises the steps of firstly obtaining an index parameter through a network loop delay predicted value to carry out quantization processing on the congestion condition of the current network; then, after the network congestion situation is subjected to quantization processing, the network state is divided into four different states according to the estimated value, and the code rate is subjected to different strategy adjustment; when the network is in an idle state, the control algorithm increases the code rate in an additive mode to increase the effective utilization rate of the network, and when the network is in a normal state, the video code rate injected into the network is not adjusted; when the network is in a growth state, reducing the code rate in an additive reduction mode; when the network is in a congestion state, the code rate is reduced in a multiplicative reduction mode; and finally, introducing emergency control based on the packet loss rate, and directly reducing the code rate to the lowest transmission code rate. The invention can improve the video quality, reduce the transmission risk, and ensure that the video delay and the buffering state are stable and the code rate is smoothly adjusted.
Description
Technical Field
The invention belongs to the technical field of communication, and particularly relates to a self-adaptive code rate control method.
Background
With the application of wireless video transmission becoming more and more extensive, disaster areas where wireless Mesh communication networks are deployed are possible to carry out disaster area live broadcast and monitoring, however, real-time video services such as video conferences have strict requirements on time delay, and main factors causing the video transmission time delay are network congestion and bit error rate. Therefore, a video transmission mechanism of the wireless mesh network is different from a wired video transmission mechanism, and how to realize efficient and stable wireless video transmission control becomes a research hotspot aiming at various characteristics of the wireless mesh network. When network congestion occurs, the data flow uploaded in the video link is necessarily controlled in the wireless video transmission system, that is, the code rate of the wireless video needs to be dynamically changed according to the network condition, so as to avoid the occurrence of network congestion and ensure the stability of the operation of the wireless video transmission link. Most of existing methods related to code rate control of wireless video transmission perform adaptive adjustment of code rate based on packet loss rate, and for a wireless mesh network with large dynamic mobility and strong expandability, these methods cannot continuously meet the requirements, and in the prior art, the method for realizing adaptive adjustment of code rate includes the following methods:
application number 201610409956.6, "video stream service code rate adaptive method based on online learning", the invention first establishes a wireless video service system framework, and generates different selectable code rate levels with scalable video coding as a source end coding mode; the QoE (Quality of Experience) is characterized by an MOS (Mean Opinion Score) value by establishing a utility function so as to optimize the target; meanwhile, a learning algorithm is provided for iterative operation, an action value function space and a confidence coefficient distribution function are updated through an instant MOS value and packet loss rate information fed back by a terminal in each period, and finally a proper code rate is selected for transmission, so that the self-adaptive control of the code rate is realized. However, in this way, only the packet loss rate in the network state is considered, and no other network state index is considered, because the packet loss rate information is a delayed network state indication, when packet loss occurs, the network has already suffered from a light or heavy network congestion, so it is difficult to cope with the video live broadcast service with a constantly changing network environment and a high real-time requirement.
Disclosure of Invention
In order to overcome the defects of the prior art, the invention provides an adaptive code rate control method based on RTT (round trip time) suitable for a wireless Mesh network, which comprises the steps of firstly obtaining an index parameter through a network loop delay predicted value to carry out quantitative processing on the congestion condition of the current network; then, after the network congestion situation is subjected to quantitative processing, the network state is divided into four different states according to the estimated value, and the code rate is adjusted by different strategies; when the network is in an idle state, the control algorithm increases the code rate in an additive mode to increase the effective utilization rate of the network, and when the network is in a normal state, the video code rate injected into the network is not adjusted; when the network is in a growth state, the code rate is reduced in an additive reduction mode; when the network is in a congestion state, the code rate is reduced in a multiplicative reduction mode; and finally, introducing emergency control based on the packet loss rate, and directly reducing the code rate to the lowest transmission code rate. The invention can improve the video quality, reduce the transmission risk, and ensure that the video delay and the buffering state are stable and the code rate is smoothly adjusted.
The technical scheme adopted by the invention for solving the technical problem comprises the following steps:
step 1: establishing a streaming media server and a client through RTCP and RTP;
and 2, step: the streaming media sending end calculates RTT time through RTCP packet information fed back by the receiving end:
RTT i ＝T now -T start -T delay (1) Wherein: RTT i Is the ith measurement of RTT, T now For the current time of RR received by the acquisition terminal, T start NTP timestamp in SR received by client; t is a unit of delay The time delay from the time when the client receives the SR packet from the server to the time when the RR packet is sent is provided;
and 3, step 3: predicting the network delay of the current RTP session by adopting an Exponential Weighted Moving Average (EWMA) -based prediction algorithm so as to obtain the smooth round-trip delay of the current network:
RTT′ i+1 ＝(1-α)RTT′ i +αRTT i (2) Wherein: RTT' i+1 Is a predicted value, RTT ', of an RTT value at the i +1 th time based on the RTT value at the i th time' i Is a predicted value based on the ith-1 st RTT value to the ith RTT value; alpha is a weighting coefficient, the value of the alpha determines the proportion of the ith measurement value of the RTT in the final predicted value, the proportion represents the reaction speed and the degree of the prediction model to the network change, and the value of the alpha is between 0 and 1;
and 4, step 4: defining a parameter T as a network status index:
T i ＝(RTT′ i+1 -D)/D (3) wherein: t is i D is the minimum round trip time predicted value RTT of the network at the lowest peak time of the network service as the estimated value of the current network state min ；
And 5: quantitatively dividing the network state into four states: idle, normal, growth, congestion, as in table 1:
TABLE 1 four states of the network
Wherein Q unloaded Network loop-back delay critical value, Q, for idle state loaded A network loop back delay critical value of a network congestion state;
and 6: introducing an index parameter U i Recording network state index parameter T i As shown in equation (4):
U i ＝T i -T i-1 (4) Wherein: t is i-1 Is an estimated value of the last network state;
if Q unloaded <T i <＝Q loaded If the following three judgment conditions are not satisfied, judging the current network state as a normal state; when any one of the following judgment conditions is satisfied, the network service is judged to be continuously increased, and the current network state is judged to be an increased state:
1) If U appears n times continuously i >0, and a rising rate U i Are all greater than a set value U change ；
2) If T i >T i-1 And the rising rate U i Greater than a set value U step ；
3) If T i Reaches and exceeds the set value of Q increase Value, but not yet Q loaded A value;
wherein, U change For trend controlThreshold value, U step For sudden variation of the threshold, Q, of RTT increase Indicating that the network state enters a growth state critical state, Q loaded Indicating that the network state enters a congestion state;
and 7: filtering and predicting the network packet loss rate by adopting an EWMA algorithm:
P′ i+1 ＝(1-γ)P′ i +γP i (5) Wherein, P' i+1 Predicted value, P ', representing current network packet loss rate' i Predicted value, P, representing last network packet loss rate i Representing the packet loss rate measured in the current measurement window, and gamma represents the weight;
and step 8: the code rate injected into the network is constrained as follows: current _ rate ∈ [ Minrate, maxrate ], where: current _ rate is the code rate of the Current injection network, minrate is the lowest code rate for finishing the basic display of the video, and Maxrate is the maximum code rate which can be loaded by the Current network; the code rate adaptive control algorithm of different RTTs is adopted for the following 5 different network states, which specifically comprises the following steps:
1) When the network is in an idle state, the code rate value injected into the network is increased in an additive mode, so that the effective utilization rate of the network is increased, and a code rate control algorithm is shown as a formula (6):
Current_rate n ＝min(Current_rate n-1 + I, maxrate) (6) wherein, current _ rate n-1 Current _ rate, the code rate injected into the network for the last time instant n I is an additive increase and decrease factor in a code rate control algorithm, and is the adjusted code rate injected into the network;
2) When the network is in a normal state, the video code rate injected into the network is not adjusted;
3) When the network is in a growth state, the code rate needs to be reduced, and a code rate control algorithm is shown as formula (7):
Current_rate n ＝max(Current_rate n-1 -I，Minrate) (7)
4) When the network is in a congestion state, the network delay needs to be reduced by reducing the code rate, and the code rate control algorithm is shown in formula (8):
Current_rate n ＝max(Current_rate n-1 d, minrate) (8) wherein: d is a multiplicative subtraction factor;
5) When P' i+1 >And = β, which indicates that a packet loss phenomenon has occurred in the current network, and at this time, the code rate should be adjusted to the lowest point, and the code rate control algorithm is shown in equation (9):
Current_rate n+1 = min rate (9) where β is the packet loss rate threshold, current _ rate n+1 Is the code rate value expected to be injected at the next time.
Preferably, α =0.8, γ =0.8, and β =0.5%.
Preferably, said Q unloeded Value of 5%, Q loeded A value of 30%, U change Value of 6%, U step Value of 10%, Q increase The value is 20%.
Preferably, the value of I is 35Kbps, the value of n is 5, and the value of d is 2.
The invention has the following beneficial effects:
the invention realizes the adjustment and control of the code rate injected into the network, completes the congestion control and the real-time adjustment of the video quality, and combines various information to carry out the self-adaptive adjustment of the code rate, rather than the traditional single consideration based on the packet loss rate; the effects of improving the video quality and reducing the transmission risk are achieved; finally, the video time delay and the buffer state are stable, the code rate adjustment is smooth, and the wide area network client has good watching experience.
Drawings
FIG. 1 is a flowchart of a method for controlling a code rate according to the present invention.
Detailed Description
The invention is further illustrated with reference to the following figures and examples.
The traditional methods such as a video streaming service code rate self-adaption method based on online learning and the like are all used for carrying out code rate self-adaption control based on packet loss rate, and because the packet loss rate is a representation of a delayed system state and the characteristic of dynamic mobility of a wireless Mesh network is not considered, the traditional code rate control method is not suitable for a wireless Mesh networking scene with high delay requirement, such as live video.
Aiming at the problems existing in the traditional mode, the invention improves the traditional mode and provides a code rate self-adaptive control method based on a plurality of comprehensive factors of RTT (Round-Trip Time Round-Trip delay) and packet loss rate. Unlike the traditional method, the adjustment of the code rate is not performed until the time delay occurs for a long time, but the code rate control is performed according to the round trip time RTT of each time, the current network state is subdivided into different situations according to the difference of the RTT values, different code rate adjustments are performed according to each situation, obviously, the control of the code rate after the network state is finely adjusted is more smooth, and the code rate control is used as a user of a video client side, so that the watching experience is smoother, large pause cannot occur, and the code rate control also meets the requirement of real-time live video. Meanwhile, the invention also provides an emergency control algorithm based on the packet loss rate to adapt to a series of sudden situations caused by node movement and routing multipath of the wireless Mesh network, and the method can be also suitable even in the limit scene of temporary emergency rescue when a disaster comes.
Step 1: the stream media server and the client are established by RTCP (Real-Time Transport Control Protocol) and RTP (Real-Time Transport Protocol implementation Transport Protocol) protocols. The RTP protocol is constructed on the UDP protocol, receives streaming media code stream from an application layer, encapsulates the streaming media code stream into RTP data packets, and then delivers the RTP data packets to a lower layer UDP. The RTP protocol provides sequence numbers to recover the sequence of data packets, realize packet loss detection and provide information such as network congestion for real-time transmission; providing a time stamp for media synchronization, so that a receiving end plays back data at a correct rate; providing a synchronization source mark to enable a receiving end to obtain information related to a sending end;
step 2: the streaming media sending end calculates RTT time through RTCP packet information fed back by the receiving end:
RTT i ＝T now -T start -T delay (1) Wherein: RTT i Is the measurement of the ith time of RTT, T now For the current time of RR (receiver report) received by the collecting terminal, T start A Network Time Protocol (NTP) timestamp in a Sender Report (SR) received by a client; t is delay The time delay from the time when the client receives the SR packet from the server to the time when the RR packet is sent is provided;
and step 3: predicting the network delay of the current RTP session by adopting an Exponential Weighted Moving Average (EWMA) based prediction algorithm so as to obtain the smooth round-trip delay of the current network:
RTT′ i+1 ＝(1-α)RTT′ i +αRTT i (2) Wherein: RTT' i+1 Is a predicted value, RTT ', of an RTT value at the i +1 th time based on the RTT value at the i th time' i Is a predicted value based on the ith-1 st RTT value to the ith RTT value; alpha is a weighting coefficient, the value of the alpha determines the proportion of the ith measurement value of RTT in the final predicted value, the response speed and degree of the prediction model to network change and the error correction capability of the prediction model are represented, the value of the alpha can be tentatively valued according to the local network condition to obtain the best effect, the values of different network conditions are not unique, the value of alpha is between 0 and 1, and the value of alpha in the EWMA prediction algorithm is 0.8;
and 4, step 4: defining a parameter T as a network state index:
T i ＝(RTT′ i+1 -D)/D (3) wherein: t is a unit of i D is the minimum return loop delay time predicted value RTT of the network at the lowest peak time of the network service for the estimated value of the current network state min ；
And 5: quantitatively dividing the network state into four states: idle, normal, growth, congestion, as in table 1:
TABLE 1 four states of the network
Wherein Q unloaded Critical value of network loop delay in idle state, Q unloeded Value of 5%, Q loaded Critical value of network loop delay, Q, for network congestion status loeded The value is 30%;
and 6: q unloaded <T i <＝Q loaded The network state is shown to be possibly in a normal state or in a load accumulation state caused by continuous increase of video traffic, the accumulation can be represented as continuous increase of network loop delay or reflected on the increase rate of the network loop delay, and the network state index parameter T can be used for indicating that the network state is possibly in a normal state or in a load accumulation state caused by continuous increase of the video traffic, and the accumulation can be represented as continuous increase of the network loop delay or reflected on the increase rate of the network loop delay and can be used for indicating the network state index parameter T i The amount of change of (c) is indirectly reflected; thus, the index parameter U is introduced i Recording network state index parameter T i As shown in equation (4):
U i ＝T i -T i-1 (4) Wherein: t is i-1 Is an estimated value of the last network state;
if Q unloaded <T i <＝Q loaded If the following three judgment conditions are not satisfied, judging the current network state as a normal state; and when any one of the following judgment conditions is satisfied, judging that the network service is continuously increased, and judging the current network state as an increased state:
1) If U appears n times in succession i >0, and a rising rate U i Are all greater than a set value U change ；
2) If T is i >T i-1 And the rising rate U i Greater than a set value U step ；
3) If T i Reaches and exceeds the set value of Q increase Value, but not yet Q loaded A value;
wherein, U change For trend control thresholds, U step Abruptly changing a control threshold for RTT; u shape change 、U step 、Q increase The values of (A) are respectively 6%, 10% and 20%, and are critical values obtained by statistical learning under different network states, U change Representing the transition of the network state from the growth state to the congestion stateThe critical value of the network loop delay difference value, the fact that the delay of the network at the later moment is larger than that of the network at the former moment after n times of continuous network means that the network evolves towards a congestion state, and the value of n is 5.
And 7: an emergency control algorithm is defined to deal with dynamic changes of the wireless Mesh network, and the instability of the wireless Mesh network is prominent due to the specific multi-hop of the wireless Mesh network, the node mobility, the weak factors of the multi-path effect and the like. Therefore, each time, due to the movement of the network node or the exhaustion of the power of the node, there is a temporary sudden change in the time delay, and the sudden change may cause a temporary packet loss of the network. Therefore, it is necessary to introduce an emergency control algorithm based on the packet loss rate into the congestion control algorithm, so that the entire control system adjusts the code rate to the lowest point when the packet loss rate abnormally increases in the network.
Packet loss ratio refers to the ratio of the number of packets lost in transmission to the number of packets transmitted, and is typically measured in throughput, which is related to the packet length and the packet transmission frequency. Therefore, the statistical method of the packet loss rate is as follows:
wherein: curSeqNo is the sequence number of the latest RTP arriving when the packet loss rate is calculated, lastSeqNo is the sequence number of the RTP arriving when the packet loss rate is calculated last time, and PktRCv is the total number of all RTP packets received during the period, and the data can be obtained from data frames of RTP and RTCP protocol packets.
Through the above calculation, the network packet loss rate of the last measurement window can be obtained. However, because the network has strong dynamic characteristics, the state of the network changes at any time, and the packet loss rate of the current measurement window cannot be directly used as the packet loss rate prediction value of the next measurement window. Therefore, an EWMA algorithm is adopted to filter and predict the network packet loss rate:
P′ i+1 ＝(1-γ)P′ i +γP i (5) Wherein, P' i+1 Predicted value, P ', representing current network packet loss rate' i Predicted value, P, representing last network packet loss rate i Representing the packet loss rate measured in the current measurement window, and gamma represents the weight;
and step 8: the code rate injected into the network is constrained as follows: current _ rate ∈ [ Minrate, maxrate ], where: current _ rate is the code rate of the Current injection network, minrate is the lowest code rate for finishing the basic display of the video, and Maxrate is the maximum code rate which can be loaded by the Current network;
the code rate adaptive control algorithm of different RTTs is adopted for the following 5 different network states, which specifically includes:
1) When the network is in an idle state, the code rate value injected into the network is increased in an additive mode, so that the effective utilization rate of the network is increased, and a code rate control algorithm is shown as a formula (6):
Current_rate n ＝min(Current_rate n-1 + I, maxrate) (6) wherein Current _ rate n-1 The code rate, current _ Rate, for the last time injected into the network n I is an additive increase and decrease factor in a code rate control algorithm, training setting is carried out according to the state of a local network, and the value is 35Kbps;
2) When the network is in a normal state, the current network load state is good, no congestion tendency exists, and the video code rate injected into the network is not adjusted;
3) When the network is in a growth state, it means that the current network load is large, and there is a tendency to enter a congestion state, and it is necessary to reduce the code rate, and the code rate control algorithm is shown in formula (7):
Current_rate n ＝max(Current_rate n-1 -I，Minrate) (7)
4) When the network is in a congestion state, the network delay cannot meet the video delay requirement, the network delay needs to be reduced by reducing the code rate, and the code rate control algorithm is shown as formula (8):
Current_rate n ＝max(Current_rate n-1 d, minrate) (8) wherein: d is a multiplicative subtraction factor; network of operationWhen congestion occurs, it is implicitly assumed that a new application in the network shares the bandwidth resource of the application, and the traffic load should be halved. Therefore, take d =2.
5) When P' i+1 >And = β, which indicates that an obvious packet loss phenomenon has occurred in the current network, and at this time, the code rate should be adjusted to the lowest point, and the code rate control algorithm is shown in formula (9):
Current_rate n+1 = min rate (9), where β is the packet loss threshold, and its value is 0.5%; current _ rate n+1 Is the code rate value expected to be injected at the next time.
The specific embodiment is as follows:
the invention has good improvement performance in both time delay and packet loss rate, and the following table 2 shows the comparison of the time delay test data of the video transmission system.
Table 2 video transmission system delay test data
Compared with the first three groups of experimental data, the code rate control algorithm has no great influence on the video transmission delay when the network is idle. By comparing the three groups of data 4-6 under the two conditions, the video transmission delay obviously rises by times under the condition of no code rate control, and the video transmission delay has no obvious change under the condition of a code rate control algorithm. After repeated tests, the change rule of the test data is basically consistent. It can be concluded that the rate control algorithm proposed herein can improve the network congestion situation well.
By analyzing the packet sequence of the RTP data packets and the information of the total number of the RTP packets received by the receiving end, the packet loss rate of the RTP data packets transmitted by the network within a period of time can be calculated. In the embodiment, each test time period is set to be 30s, and is set to be the first test time period, so that the network state is good, and the bandwidth is sufficient; in the second test period, a network congestion condition is artificially created. The test pairs are shown in table 3 below, comparing the impact of using the congestion control algorithm on the average packet loss rate.
Table 3 test results of network UDP packet loss rate in experimental environment
According to the test result, under the condition that the network bandwidth is sufficient, the network packet loss rates of the two video transmission modes are not greatly different. When the available bandwidth of the network is reduced, for the condition of using a congestion control algorithm, the sending rate can be dynamically adjusted in time according to the condition of the network by the sending end, the network load is reduced when the available bandwidth is insufficient, the probability of network router queue overflow is greatly reduced, the passive packet loss of the network caused by exceeding the maximum transmission capability of the network is avoided, the available bandwidth can be utilized as much as possible when the available bandwidth is abundant, and the total data packet loss rate is obviously lower than the loss rate without adopting a congestion control mode.
Claims (4)
1. An adaptive code rate control method based on RTT suitable for a wireless Mesh network is characterized by comprising the following steps:
step 1: establishing a streaming media server and a client through RTCP and RTP;
step 2: the streaming media sending end calculates RTT time through RTCP packet information fed back by the receiving end:
RTT i ＝T now -T start -T delay (1)
wherein: RTT (round trip time) i Is the measurement of the ith time of RTT, T now For the current time of RR received by the acquisition terminal, T start The NTP timestamp in the SR received by the client is shown; t is delay The time delay from the time when the client receives the SR packet from the server to the time when the RR packet is sent is obtained;
and 3, step 3: predicting the network delay of the current RTP session by adopting an Exponential Weighted Moving Average (EWMA) based prediction algorithm so as to obtain the smooth round-trip delay of the current network:
RTT′ i+1 ＝(1-α)RTT′ i +αRTT i (2)
wherein: RTT' i+1 Is a predicted value, RTT ', of an RTT value at an i +1 th time based on the RTT value at the i th time' i Is a predicted value based on the ith-1 st RTT value to the ith RTT value; alpha is a weighting coefficient, the value of alpha determines the proportion occupied by the measured value of the ith RTT in the final predicted value and represents the reaction speed and degree of the prediction model to the network change, and the value of alpha is between 0 and 1;
and 4, step 4: defining a parameter T as a network status index:
T i ＝(RTT′ i+1 -D)/D (3)
wherein: t is a unit of i D is the minimum round trip time predicted value RTT of the network at the lowest peak time of the network service as the estimated value of the current network state min ；
And 5: quantitatively dividing the network state into four states: idle, normal, growth, congestion:
when T is i ＜＝Q unloaded Defined as idle state; when Q is unloaded ＜T i ＜＝Q loaded Defined as a normal state or a growth state; when T is i ＞Q loaded Defined as a congestion state;
wherein Q unloaded Network loop-back delay critical value, Q, for idle state loaded A network loop delay critical value of a network congestion state;
step 6: introduction of index parameter U i Recording network state index parameter T i As shown in equation (4):
U i ＝T i -T i-1 (4)
wherein: t is i-1 Is an estimated value of the last network state;
if Q unloaded ＜T i ＜＝Q loaded And if the following three judgment conditions are not met, the current network state is determinedThe state is judged to be a normal state; and when any one of the following judgment conditions is satisfied, judging that the network service is continuously increased, and judging the current network state as an increased state:
1) If U appears n times continuously i > 0, and a rising rate U i Are all greater than a set value U change ；
2) If T i ＞T i-1 And the rising rate U i Greater than a set value U step ；
3) If T i Reaches and exceeds the set value of Q increase Value, but not yet reached Q loaded A value;
wherein, U change For trend control thresholds, U step For sudden variation of the threshold, Q, of RTT increase Indicating that the network state enters a growth state critical state, Q loaded Indicating that the network state enters a congestion state;
and 7: filtering and predicting the network packet loss rate by adopting an EWMA algorithm:
P′ i+1 ＝(1-γ)P′ i +γP i (5)
wherein, P' i+1 Predicted value, P ', representing current network packet loss rate' i Predicted value, P, representing last network packet loss rate i Representing the packet loss rate measured in the current measurement window, and gamma represents the weight;
and step 8: the code rate injected into the network is constrained as follows: current _ rate ∈ [ Minrate, maxrate ], where: current _ rate is the code rate of the Current injection network, minrate is the lowest code rate for finishing the basic display of the video, and Maxrate is the maximum code rate which can be loaded by the Current network; the code rate adaptive control algorithm of different RTTs is adopted for the following 5 different network states, which specifically comprises the following steps:
1) When the network is in an idle state, the code rate value injected into the network is increased in an additive mode, so that the effective utilization rate of the network is increased, and a code rate control algorithm is shown as a formula (6):
Current_rate n ＝min(Current_rate n-1 +I，Maxrate) (6)
wherein, current _ rate n-1 The code rate, current _ Rate, for the last time injected into the network n I is an additive increase and decrease factor in a code rate control algorithm for the adjusted code rate injected into the network;
2) When the network is in a normal state, the video code rate injected into the network is not adjusted;
3) When the network is in a growth state, the code rate needs to be reduced, and a code rate control algorithm is shown as formula (7):
Current_rate n ＝max(Current_rate n-1 -I，Minrate) (7)
4) When the network is in a congestion state, the network delay needs to be reduced by reducing the code rate, and the code rate control algorithm is shown as the formula (8):
Current_rate n ＝max(Current_rate n-1 /d，Minrate) (8)
wherein: d is a multiplicative subtraction factor;
5) When P' i+1 When > = beta, it indicates that the current network has packet loss, and at this time, the code rate should be adjusted to the lowest point, and the code rate control algorithm is shown in equation (9):
Current_rate n+1 ＝Minrate (9)
where β is the packet loss rate threshold, current _ rate n+1 Is the code rate value expected to be injected at the next time.
2. The RTT-based adaptive code rate control method for a wireless Mesh network as claimed in claim 1, wherein α =0.8, γ =0.8, and β =0.5%.
3. The method of claim 1, wherein Q is the Q for adaptive code rate control based on RTT for wireless Mesh network unloeded Value 5%, Q loeded A value of 30%, U change Value of 6%, U step Value of 10%, Q increase The value is 20%.
4. The adaptive code rate control method based on RTT for wireless Mesh networks as defined in claim 1, wherein I is 35kbps, n is 5, and d is 2.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111023201.XA CN113891172B (en) | 2021-09-02 | 2021-09-02 | Adaptive code rate control method based on RTT (round trip time) and suitable for wireless Mesh network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111023201.XA CN113891172B (en) | 2021-09-02 | 2021-09-02 | Adaptive code rate control method based on RTT (round trip time) and suitable for wireless Mesh network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113891172A CN113891172A (en) | 2022-01-04 |
| CN113891172B true CN113891172B (en) | 2022-10-04 |
Family
ID=79011704
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111023201.XA Active CN113891172B (en) | 2021-09-02 | 2021-09-02 | Adaptive code rate control method based on RTT (round trip time) and suitable for wireless Mesh network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113891172B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114513474B (en) * | 2022-02-08 | 2024-03-05 | 聚好看科技股份有限公司 | Video transmission method, video transmission terminal, media server and storage medium |
| CN115549860B (en) * | 2022-09-13 | 2023-08-29 | 北京天融信网络安全技术有限公司 | Data recovery method, device, equipment and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103200621A (en) * | 2013-03-14 | 2013-07-10 | 南京理工大学 | Self-adaptation flow control method suitable for wireless projection system real-time data transmission |
| CN103209435A (en) * | 2013-03-15 | 2013-07-17 | 河海大学 | Congestion control based wireless multi-hop network multicast method |
| CN107276910A (en) * | 2017-06-07 | 2017-10-20 | 上海迪爱斯通信设备有限公司 | The real-time adjusting apparatus of video code rate and system, video server |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4778453B2 (en) * | 2007-01-24 | 2011-09-21 | 株式会社エヌ・ティ・ティ・ドコモ | Communication terminal, congestion control method, and congestion control program |
| US9559972B2 (en) * | 2013-12-30 | 2017-01-31 | Comcast Cable Communications, Llc | Systems and methods for managing congestion |
-
2021
- 2021-09-02 CN CN202111023201.XA patent/CN113891172B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103200621A (en) * | 2013-03-14 | 2013-07-10 | 南京理工大学 | Self-adaptation flow control method suitable for wireless projection system real-time data transmission |
| CN103209435A (en) * | 2013-03-15 | 2013-07-17 | 河海大学 | Congestion control based wireless multi-hop network multicast method |
| CN107276910A (en) * | 2017-06-07 | 2017-10-20 | 上海迪爱斯通信设备有限公司 | The real-time adjusting apparatus of video code rate and system, video server |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113891172A (en) | 2022-01-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP2432175B1 (en) | Method, device and system for self-adaptively adjusting data transmission rate | |
| CN111615006B (en) | Video code conversion transmission control system based on network state self-evaluation | |
| RU2305908C2 (en) | Adaptive method for evaluating multimedia data transmission speed | |
| US8588071B2 (en) | Device and method for adaptation of target rate of video signals | |
| CN108024156B (en) | Partially reliable video transmission method based on hidden Markov model | |
| CN113891172B (en) | Adaptive code rate control method based on RTT (round trip time) and suitable for wireless Mesh network | |
| US20030198184A1 (en) | Method of dynamically determining real-time multimedia streaming rate over a communications networks | |
| CN108833930B (en) | Live broadcast data transmission control method and device, live broadcast equipment and storage medium | |
| CN108401128B (en) | Congestion control method in video call | |
| EP3103220A1 (en) | System and method for dynamic effective rate estimation for real-time video traffic | |
| CN111935441B (en) | Network state detection method and device | |
| KR100924309B1 (en) | Quality adaptive streaming method using temporal scalability and system thereof | |
| Zhang et al. | Network-adaptive rate control with TCP-friendly protocol for multiple video objects | |
| EP2485441B1 (en) | A video packet scheduling method for multimedia streaming | |
| CN107483990B (en) | Dynamic code rate adjusting method and device for streaming media transmission and transmission system | |
| Lu et al. | Context-adaptive cross-layer TCP optimization for Internet video streaming | |
| Zhang et al. | Resource allocation for audio and video streaming over the Internet | |
| KR100931375B1 (en) | Efficient data streaming method using efficien tparameters and data streaming server | |
| Jammeh et al. | Rate-adaptive video streaming through packet dispersion feedback | |
| CN113612649B (en) | Round trip estimation | |
| CN111212308B (en) | Method for self-adaptive adjustment of wireless network | |
| Yang et al. | Video transmission control for networked vision system based on RTCP | |
| Papadimitriou | An integrated smooth transmission control and temporal scaling scheme for MPEG-4 streaming video | |
| Wang et al. | A cross-layer based bandwidth and queue adaptations for wireless multimedia networks | |
| Wang et al. | A Two Time-Scales Network Bandwidth Measurement for Video Transmission |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |