CN102711172A - Modified TCPW congestion control method in wireless network - Google Patents

Modified TCPW congestion control method in wireless network Download PDF

Info

Publication number
CN102711172A
CN102711172A CN2012101717629A CN201210171762A CN102711172A CN 102711172 A CN102711172 A CN 102711172A CN 2012101717629 A CN2012101717629 A CN 2012101717629A CN 201210171762 A CN201210171762 A CN 201210171762A CN 102711172 A CN102711172 A CN 102711172A
Authority
CN
China
Prior art keywords
network
bandwidth
value
packet
tcpw
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.)
Pending
Application number
CN2012101717629A
Other languages
Chinese (zh)
Inventor
王万良
刘琳琅
岑跃峰
蒋一波
郑建炜
赵燕伟
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Zhejiang University of Technology ZJUT
Original Assignee
Zhejiang University of Technology ZJUT
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Zhejiang University of Technology ZJUT filed Critical Zhejiang University of Technology ZJUT
Priority to CN2012101717629A priority Critical patent/CN102711172A/en
Publication of CN102711172A publication Critical patent/CN102711172A/en
Pending legal-status Critical Current

Links

Images

Landscapes

  • Data Exchanges In Wide-Area Networks (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

The invention discloses a modified TCPW congestion control method in a wireless network. After a specified data receiving terminal receives a data packet each time, a timestamp is stamped in a reply packet corresponding to the data packet, after a data sending terminal receives the reply packet, the arrive time of the current reply packet is not recorded, but the timestamp on the reply packet is extracted, so as to obtain the timestamp difference, namely the sending time of the last data packet on the sending route, in the next arrived reply packet, the size of the arrived data packet and the sending time of the data packet on the sending route are respectively subjected to averaging at the sending terminal, and then a bandwidth sampling value is obtained according to the specific value of average values. Finally a bandwidth estimated value is smoothly obtained through an exponential weighting moving average filter according to the obtained bandwidth sampling value. The method is more accurate to estimate a current practical available bandwidth value and effectively improves network QoS.

Description

Improved TCPW jamming control method in a kind of wireless network
Technical field
The present invention relates to the congested control of Wi-Fi field, especially a kind of TCPW jamming control method.
Background technology
At present, the congestion control in wireless networks technology is faced with many difficulties.Because wireless network has the self characteristics of terminal mobility, wireless connections; Make that the reliability of link is not high; And want to postpone at high bit-error, greatly and the frequent mobile wireless network in terminal in the existing Transmission Control Protocol of operation; Become difficult relatively, transfer of data has its unsteadiness, and congestion phenomenon also occurs than being easier to.Therefore, want to make Transmission Control Protocol stably to operate in the wireless network, network throughput is preferably arranged, and have good robustness and extensibility, just must carry out corresponding modification original protocol.
TCPW is a kind of congestion avoidance algorithm end to end, meets the design criterion of Transmission Control Protocol fully.In the data transmission procedure of wireless network, it has been eliminated to a certain extent because the negative effect that the data random loss brings for network bandwidth utilance.The algorithm of TCPW is a kind of algorithm based on bandwidth estimation; It mainly is an available bandwidth value of estimating current network through the historical information on the wireless network links, and this historical information generally comprises the stream information of replying that message segment information and the recipient of transmission receive.TCPW checks that through transmit leg the stream information of replying in the historical information comes the message transmission rate of current connection is estimated, thereby the current network available bandwidth is estimated.It estimates the transmission present situation of the adjustment network that resulting available bandwidth value can be real-time; When appearance in the network is congested; Can adjust congestion window cwnd and begin threshold value ssthresh slowly through the available bandwidth value that estimates; The algorithm adjustment that last employing again and formula increase long-pending formula minimizing sends to the grouping size in the network, thereby reaches the purpose of controlling congestion phenomenon, improves the performance of TCP.
TCPW has adopted the thought that the network bandwidth is estimated in real time, and promptly the time interval of the echo reply stream that sends through continuous calculating recipient of transmit leg is estimated the available bandwidth on the link.Adopt ABSE (Adaptive Bandwidth Share Estimation) filtering mechanism to make it comparatively accurate to the estimation of bandwidth.
The TCPW basic idea is; In slow incipient stage and congestion avoidance phase; TCPW is the same with TCP, and the method for adjustment of send window does not change, when generation in the network is congested; Dedicated bandwidth estimated value BWE comes dynamically to adjust congestion window value cwnd and begins threshold value ssthresh slowly, and congestion control mechanism is AIAD.
(1) when transmitting terminal receives three repeat ACKs, makes ssthresh=(BWE*RTT Min)/seg_size (and being not the ssthresh=cwnd/2 of Reno).Wherein, if cwnd>ssthresh, then cwnd=ssthresh;
(2) when timer generation RTO is overtime, make ssthresh=(BWE*RTT Min)/seg_size, (and being not ssthresh=cwnd/2), cwnd=1.
Above congestion control mechanism all is based on the basis of bandwidth estimation, and the bandwidth estimation algorithm also becomes the key of this congestion control mechanism.
Wherein, the time interval of the ACK of sampling bandwidth and corresponding arrival:
b k=d kk (7)
Δ k=t k-t k-1 (8)
t kRepresent that one is replied stream ACK at t kConstantly arrive source end, Δ kExpression t K-1To t kThe time interval, d kBe illustrated in Δ kThe time interval in, the data character stream joint number that the TCP receiving terminal is received, b kBe Δ kSampling bandwidth in the time interval, adopt the EWMA technique computes to estimate the current bandwidth value that obtains again:
BWE k = βBWE K - 1 + ( 1 - β ) ( b k + b k - 1 2 ) - - - ( 9 )
β=(2τ-Δ k)/(2τ+Δ k) (10)
Wherein, BWE kBe t kThe estimation of available bandwidth value of filtering constantly, 1/ τ is the cut-off frequency of the low pass filter in the index filter, β is for filtering weights (general value 19/21).
When packet loss took place, TCP Reno reset to 1 with cwnd, when receiving three ACK streams, cwnd was reduced by half.And for fear of when receiving three repeat ACK streams; The excessive atrophy of cwnd; TCP Westwood has introduced a kind of mechanism of fast quick-recovery; It calculates threshold value ssthresh through estimating the estimation of available bandwidth value in the current connection of TCP, makes network take place when congested, and threshold value can be adjusted with available bandwidth synchronously.This congestion control policy based on estimation of available bandwidth has obtained good checking in the congested test of wireless network.But this algorithm also exists defective, and the network bandwidth value that it generally estimates is all high than the actual available bandwidth value of network, when having ACK compressive flow etc. to take place in the network, the estimation of bandwidth is also become very unstable.Simultaneously and since its too high estimation the available bandwidth of network, make transmit leg send to packet on the Radio Link, the congestion phenomenon on the contrary can emphasis network in fact greater than the available bandwidth of network.
2002; After TCPW is proposed first; On this basis; Increasing researcher begins one's study based on the congestion avoidance algorithm of bandwidth estimation, and has proposed a lot of improved algorithms, and wherein topmost improvement algorithm has following four kinds: TCPW BE (Bandwidth Estimation), TCPW RE (Rate Estimation), TCPW CRB (Combined Rate and Bandwidth) and TCPW ABE (Adaptive Bandwidth Estimation).
(1)TCPW?RE(Rate?Estimation)
The difference of it and early stage TCPW algorithm is exactly on the sampling mode of bandwidth.This algorithm will extend to T sample time, get interior all received ACK of TCP transmitting terminal of the T time interval and reply stream.Make that like this sampling bandwidth value of trying to achieve is an average on time interval T.To a certain extent, eliminated the problem of the too high estimation actual available bandwidth of TCPW.Specific algorithm is following:
b k = Σ t j > t k - T d j / T - - - ( 11 )
BWE k = β BWE K - 1 + ( 1 - β ) ( b k + b k - 1 2 ) - - - ( 12 )
β=(2τ-Δ k)/(2τ+Δ k) (13)
Wherein, the bandwidth sampling mode in the formula (11) is different, and it will extend to T sample time, and the T default value of this moment is RTT, and other parameters are the same.
In the wireless network environment, if congested only just because the data packet loss causes that the performance of TCPRE algorithm is better than TCPW.
(2)TCP?CRB(Combined?Rate?and?Bandwidth)
The main thought of TCP CRB algorithm is: under wireless network environment; Because TCP RE is relatively more effective in the network of the congested data-bag lost that causes; And that TCPW shows in the network of the packet loss that random error causes is better, and their boths can not adapt to all network environments.Based on this, TCP CBR has combined the advantage of TCPW and TCP RE, and at first (cwnd/ (BWERTTmin)/seg_size) judges network condition, and then selects to use which kind of mode to obtain the bandwidth value of sampling according to congestion detection mechanism.When (cwnd/ (BWERTTmin)/seg_size)>during θ (θ=1.4), use the congestion control mechanism of TCP BE, on the contrary then use the congestion control mechanism of TCP RE.This algorithm has used existing TCP RE and TCPW algorithm more flexibly, and the scope of application is wider than TCPW and two kinds of algorithms of TCP RE.
(3)TCP?ABE(Adaptive?Bandwidth?Estimation)
This algorithm is to time interval T kValue more flexible; It mainly is in time to adjust according to the congestion state of network to get; The sampling bandwidth value that obtains so also is that the congestion condition according to network carries out in good time adjustment ground; Thereby make algorithm can well estimate, and draw the controlling value that relatively meets network state, help control more the network congestion phenomenon to the state of network.The concrete bandwidth of TCP ABE sampling formula is following:
b k = Σ t j > t k - T d j / T k - - - ( 14 )
T k = RTT · ( cwnd RTT min - Th k ) / cwnd RTT min - - - ( 15 )
Th k=εTh k+(1-ε)Th (16)
Th = Σ t j > t k - RTT d j RTT - - - ( 17 )
The concrete implication of parameter also has brightly specifically with being arranged in the following formula among the TCPW RE, no longer details here.The tracking network state that TCPABE algorithm ability is real-time is adjusted the parameters in the formula according to network state, thereby can be good at improving the real-time and the accuracy of algorithm.But this algorithm also has apparent in view shortcoming, and the parameter that it relates to is many, calculates more complicated.
Summary of the invention
In order to overcome the defective of the too high estimation network available bandwidth value in the existing TCPW congestion avoidance algorithm; And the network asymmetry is to the influence of network congestion control, and the present invention provides a kind of and more accurately estimates current actual available bandwidth value, effectively improves improved TCPW jamming control method in the wireless network of network QoS.
The technical solution adopted for the present invention to solve the technical problems is:
Improved TCPW jamming control method in a kind of wireless network, said improved TCPW jamming control method may further comprise the steps:
Step 1 after each data receiver receives a packet, is all stamped a timestamp in its corresponding response packet, calculate the difference Δ t of adjacent timestamp then at data sending terminal i
Step 2, according to the definition of the network bandwidth, Δ t iThe sampling bandwidth value of network calculates as follows in period:
b i = d i Δ t i - - - ( 1 ) ;
In the formula (1), d iBe illustrated in time period Δ t iIn, institute's data packets for transmission size in the network;
Step 3 is asked sliding average
Figure BDA00001685799800062
and
Figure BDA00001685799800063
to the packet size that the difference and the receiving terminal of timestamp receives respectively in this time period
Step 4 based on step 1-3, can obtain the current sampling bandwidth value in the network:
b i = d i ‾ Δt i ‾ - - - ( 2 )
Step 5, the available bandwidth value that sampling bandwidth estimated value that utilization obtains and the estimation of available bandwidth value of last time calculate current network carries out level and smooth value with the exponentially weighted moving average (EWMA) filter, and is specific as follows:
B i = β B i - 1 + ( 1 - β ) ( b i + b i - 1 2 ) - - - ( 3 )
Wherein, β representes the mean value weights.
Further again, in step 1, after data sending terminal receives response packet, be not the time of advent of the current response packet of record, but extract the timestamp on its response packet, the difference Δ t of the timestamp in the response packet of adjacent arrival iIn fact be exactly that last data wrap in the transmitting time on the transmit path;
Δt i=t i-t i-1 (4)
In (4) formula, t iFor after receiving terminal receives i packet, the timestamp of in the echo reply bag, stamping, Δ t iBeing the difference between two timestamps of adjacent arrival, also is that i data wrap in the transmitting time on the transmit path in the network.
Further; In step 3; During the slip
Figure BDA00001685799800071
of difference sliding average that stabs computing time and the packet that receives in this time period size, the employing computing formula:
Δt i ‾ = α t i - 1 ‾ + ( 1 - α ) · t i - - - ( 5 )
d i ‾ = α d i - 1 ‾ + ( 1 - α ) · d i - - - ( 6 )
Wherein, α representes the mean value weights, for the transient change of reaction network to sampling bandwidth estimation effect degree, the b of this moment iBe the bandwidth sampling estimated value of being tried to achieve under the condition of the present invention.
Technical conceive of the present invention is: after the inventive method specified data receiving terminal receives a packet at every turn; All in its corresponding response packet, stamp a timestamp; Then after data sending terminal receives response packet; Be not the time of advent of the current response packet of record, but extract the timestamp on its response packet, in fact the difference of the timestamp in the response packet of the adjacent arrival that obtains like this is exactly that last data wrap in the transmitting time on the transmit path.Respectively the packet size of confirming to arrive is averaged with the transmitting time of this packet on transmit path at transmitting terminal then, the ratio according to them obtains the bandwidth sampled value again.According to resulting bandwidth sampled value, use the exponentially weighted moving average (EWMA) filter smoothly to obtain the bandwidth estimation value at last.
Beneficial effect of the present invention mainly shows: when having situation such as ACK compression generation, the sudden arrival of message segment, answering delay to take place in the network; The available bandwidth of the estimation network that the meeting of TCPW algorithm is too high; It is unstable that the bandwidth estimation value also can become; TCPW also can take too much bandwidth, the congestion phenomenon of emphasis network.The present invention is through asking for of timestamp and being used of moving average value; Well solved this problem, made the network bandwidth estimated value of trying to achieve well to follow the tracks of the network bandwidth available estimate on the transmit path, avoided the influence of return path and bursts greatly bandwidth estimation; Improved network throughput greatly; Occurring in the congestion phenomenon at network, can effectively adjust the network congestion state, is a kind of method of comparatively desirable raising network QoS.
Description of drawings
Fig. 1 is the timestamp t that transmitting terminal receives iThe sketch map of the packet size of the correspondence that receives with receiving terminal.
Embodiment
Below in conjunction with accompanying drawing the present invention is further described.
With reference to Fig. 1, though at present the congested control method of multiple network is arranged in wireless network, wherein the TCPW algorithm is more satisfactory aspect congested control, it also strictly observe TCP end to end design criterion design.But when in network, having bursts, delay compressed package etc. to take place, the network jitter under the TCPW is bigger, has influenced the recovery of network state.Method in the present embodiment just is being based on that above analysis result draws.
In the network, the data transmission rate of TCP source end is dynamically to adjust according to the state of network is next.The speed that is exactly according to the response packet that returns of improving one's methods of TCPW among the present invention measures and the current available bandwidth of estimation network, thereby adjusts the data transmission rate of source end.The network source end is according to estimating that the current network available bandwidth value that obtains is provided with congestion window size and starts threshold value slowly.
When the network connection is in slow incipient stage and congestion avoidance phase, there is not congestion phenomenon to take place in the network, this moment, TCPW was the same with TCP, and the method for adjustment of send window does not change.When generation in the network was congested, this moment, TCPW was different to the adjustment scheme and the TCP of network.TCPW dedicated bandwidth estimated value BWE comes dynamically to adjust congestion window value cwnd and begins threshold value ssthresh slowly, and congestion control mechanism is AIAD.
(1) when transmitting terminal receives three repeat ACKs, ssthresh=(BWE*RTT is set Min)/seg_size, wherein, if cwnd is>Ssthresh then is provided with cwnd=ssthresh;
(2) when timer generation RTO is overtime, ssthresh=(BWE*RTT is set Min)/seg_size, cwnd=1.
Wherein ssthresh is for start threshold value slowly, and cwnd is a congestion window size, RTT MinBe minimum round-trip delay value, seg_size is the size of tcp data bag.
Wherein the network availability bandwidth estimated value BWE method that just need use the bandwidth estimation of mentioning among the present invention is estimated to obtain.Shown in accompanying drawing 1, after data sending terminal receives response packet, note the timestamp t that echo reply is wrapped iBig or small d with the received packet of data receiver i, calculate the difference Δ t of the timestamp in the response packet of adjacent arrival again i
Δt i=t i-t i-1 (1)
In (1) formula, t iFor after receiving terminal receives i packet, the timestamp of in the echo reply bag, stamping, Δ t iBe the difference between two timestamps of adjacent arrival, and then the packet size that respectively difference and the receiving terminal of timestamp is received is asked sliding average in this time period
Figure BDA00001685799800091
With
Figure BDA00001685799800092
Adopt computing formula:
Δt i ‾ = α t i - 1 ‾ + ( 1 - α ) · t i - - - ( 2 )
d i ‾ = α d i - 1 ‾ + ( 1 - α ) · d i - - - ( 3 )
Wherein, α representes the mean value weights, for the transient change of reaction network to sampling bandwidth estimation effect degree, general value gets 0.9 here greater than 0.5.The b of this moment iBe the bandwidth sampling estimated value of being tried to achieve under the condition of the present invention
Based on above analysis, can obtain the current sampling bandwidth value in the network:
b i = d i ‾ Δt i ‾ - - - ( 4 )
Sampling bandwidth estimated value that utilization obtains and the estimation of available bandwidth value of last time calculate the available bandwidth value of current network; The jitter problem that network is caused to bursts; We carry out level and smooth value with the exponentially weighted moving average (EWMA) filter, and are specific as follows:
BWE i = β BWE i - 1 + ( 1 - β ) ( b i + b i - 1 2 ) - - - ( 5 )
Wherein β representes the mean value weights, gets 19/21 here, BWE iThe current available bandwidth value of required exactly network of trying to achieve.

Claims (3)

1. improved TCPW jamming control method in the wireless network, it is characterized in that: said improved TCPW jamming control method may further comprise the steps:
Step 1 after each data receiver receives a packet, is all stamped a timestamp in its corresponding response packet, calculate the difference Δ t of adjacent timestamp then at data sending terminal i
Step 2, according to the definition of the network bandwidth, Δ t iThe sampling bandwidth value of network calculates as follows in period:
b i = d i Δ t i - - - ( 1 ) ;
In the formula (1), d iBe illustrated in time period Δ t iIn, institute's data packets for transmission size in the network;
Step 3 is asked sliding average
Figure FDA00001685799700012
and
Figure FDA00001685799700013
to the packet size that the difference and the receiving terminal of timestamp receives respectively in this time period
Step 4 based on step 1-3, can obtain the current sampling bandwidth value in the network:
b i = d i ‾ Δt i ‾ - - - ( 2 )
Step 5, the available bandwidth value that sampling bandwidth estimated value that utilization obtains and the estimation of available bandwidth value of last time calculate current network carries out level and smooth value with the exponentially weighted moving average (EWMA) filter, and is specific as follows:
B i = β B i - 1 + ( 1 - β ) ( b i + b i - 1 2 ) - - - ( 3 )
Wherein, β representes the mean value weights.
2. improved TCPW jamming control method in the wireless network as claimed in claim 1; It is characterized in that: in step 1; After data sending terminal receives response packet; Be not the time of advent of the current response packet of record, but extract the timestamp on its response packet, the difference Δ t of the timestamp in the response packet of adjacent arrival iIn fact be exactly that last data wrap in the transmitting time on the transmit path;
Δt i=t i-t i-1 (4)
In (4) formula, t iFor after receiving terminal receives i packet, the timestamp of in the echo reply bag, stamping, Δ t iBeing the difference between two timestamps of adjacent arrival, also is that i data wrap in the transmitting time on the transmit path in the network.
3. improved TCPW jamming control method in according to claim 1 or claim 2 the wireless network; It is characterized in that: in step 3; During the slip
Figure FDA00001685799700017
of difference sliding average that stabs computing time and the packet that receives in this time period size, the employing computing formula:
Δt i ‾ = α t i - 1 ‾ + ( 1 - α ) · t i - - - ( 5 )
d i ‾ = α d i - 1 ‾ + ( 1 - α ) · d i - - - ( 6 )
Wherein, α representes the mean value weights, for the transient change of reaction network to sampling bandwidth estimation effect degree, the b of this moment iBe the bandwidth sampling estimated value of being tried to achieve under the condition of the present invention.
CN2012101717629A 2012-05-25 2012-05-25 Modified TCPW congestion control method in wireless network Pending CN102711172A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN2012101717629A CN102711172A (en) 2012-05-25 2012-05-25 Modified TCPW congestion control method in wireless network

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN2012101717629A CN102711172A (en) 2012-05-25 2012-05-25 Modified TCPW congestion control method in wireless network

Publications (1)

Publication Number Publication Date
CN102711172A true CN102711172A (en) 2012-10-03

Family

ID=46903691

Family Applications (1)

Application Number Title Priority Date Filing Date
CN2012101717629A Pending CN102711172A (en) 2012-05-25 2012-05-25 Modified TCPW congestion control method in wireless network

Country Status (1)

Country Link
CN (1) CN102711172A (en)

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105049369A (en) * 2015-08-14 2015-11-11 浙江大学 Video transmission congestion control method based on MPTCP in heterogeneous wireless network
CN108848415A (en) * 2018-05-04 2018-11-20 北京大学 A kind of the network real-time video transmission method and device of adaptive learning
CN109361566A (en) * 2018-11-13 2019-02-19 天津津航计算技术研究所 Utilize the method for test serial number and message time stamp prediction UDP Congestion Level SPCC
WO2019128430A1 (en) * 2017-12-29 2019-07-04 杭州海康威视***技术有限公司 Method, apparatus and device for determining bandwidth, and storage medium
CN112689268A (en) * 2020-12-21 2021-04-20 一飞(海南)科技有限公司 Formation unmanned aerial vehicle, and method and system for improving multi-route data transmission efficiency through networking
CN113824646A (en) * 2021-08-27 2021-12-21 北京达佳互联信息技术有限公司 Slow-start control method and device, electronic equipment and storage medium
CN114145008A (en) * 2019-05-20 2022-03-04 谷歌有限责任公司 Time of flight estimation for transmission control protocol
CN114244778A (en) * 2022-01-07 2022-03-25 平行云科技(北京)有限公司 QoE-aware WebRTC congestion control method

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101369980A (en) * 2008-09-26 2009-02-18 北京航空航天大学 Band width estimation method based on self-adapting sampling
US20090059788A1 (en) * 2007-08-29 2009-03-05 Motorola, Inc. Method and Apparatus for Dynamic Adaptation of Network Transport

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20090059788A1 (en) * 2007-08-29 2009-03-05 Motorola, Inc. Method and Apparatus for Dynamic Adaptation of Network Transport
CN101369980A (en) * 2008-09-26 2009-02-18 北京航空航天大学 Band width estimation method based on self-adapting sampling

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
付强等: "TCPW-REAL:基于无线网络的TCPW改进协议", 《计算机工程与设计》 *

Cited By (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105049369A (en) * 2015-08-14 2015-11-11 浙江大学 Video transmission congestion control method based on MPTCP in heterogeneous wireless network
WO2019128430A1 (en) * 2017-12-29 2019-07-04 杭州海康威视***技术有限公司 Method, apparatus and device for determining bandwidth, and storage medium
CN109995609A (en) * 2017-12-29 2019-07-09 杭州海康威视***技术有限公司 Bandwidth determining method, device and storage medium
CN108848415A (en) * 2018-05-04 2018-11-20 北京大学 A kind of the network real-time video transmission method and device of adaptive learning
CN108848415B (en) * 2018-05-04 2021-01-15 北京大学 Network real-time video transmission method and device for adaptive learning
CN109361566A (en) * 2018-11-13 2019-02-19 天津津航计算技术研究所 Utilize the method for test serial number and message time stamp prediction UDP Congestion Level SPCC
CN114145008A (en) * 2019-05-20 2022-03-04 谷歌有限责任公司 Time of flight estimation for transmission control protocol
CN112689268A (en) * 2020-12-21 2021-04-20 一飞(海南)科技有限公司 Formation unmanned aerial vehicle, and method and system for improving multi-route data transmission efficiency through networking
CN112689268B (en) * 2020-12-21 2022-09-06 一飞(海南)科技有限公司 Method and system for improving multi-route data transmission efficiency by formation unmanned aerial vehicle networking
CN113824646A (en) * 2021-08-27 2021-12-21 北京达佳互联信息技术有限公司 Slow-start control method and device, electronic equipment and storage medium
CN114244778A (en) * 2022-01-07 2022-03-25 平行云科技(北京)有限公司 QoE-aware WebRTC congestion control method
CN114244778B (en) * 2022-01-07 2023-11-21 平行云科技(北京)有限公司 QoE-aware WebRTC congestion control method

Similar Documents

Publication Publication Date Title
CN102711172A (en) Modified TCPW congestion control method in wireless network
CN101854738B (en) Transmission control protocol method for satellite network
KR100717219B1 (en) Method and device for determining a time-parameter
US7965698B2 (en) Method for preventing unnecessary retransmission due to delayed transmission in wireless network and communication device using the same
JP4632874B2 (en) Communication terminal
JP4708978B2 (en) Communication system, communication terminal, session relay device, and communication protocol realizing high throughput
JP4778453B2 (en) Communication terminal, congestion control method, and congestion control program
Polese et al. milliProxy: A TCP proxy architecture for 5G mmWave cellular systems
WO2002019654A2 (en) Method for improving tcp performance over wireless links
KR20060100512A (en) Method and system for estimating average bandwidth in a communication network based on transmission control protocol
KR20030058946A (en) Method and device for robust real-time estimation of the bottleneck bandwidth in the internet
CN101557607A (en) Method for controlling transmission of sink nodes in wireless sensor network
CN104113884B (en) Real-time multimedia transfer rate control mechanism in a kind of wireless network
US9948565B2 (en) Method and implementation of zero overhead rate controlled (ZORC) information transmission via digital communication link
JP2008104018A (en) Communication system, communication equipment, and transmission control method
JP2006173961A (en) Tcp convergence control system in wide band, high delay radio network
JP5387058B2 (en) Transmission device, transmission rate calculation method, and transmission rate calculation program
Kato et al. Comparing TCP Congestion Control Algorithms Based on Passively Collected Packet Traces
Lee et al. TCP-friendly congestion control for streaming real-time applications over wireless networks
KR100608581B1 (en) Method and apparatus for transmitting data in ad hoc network
Sharma et al. Performance evaluation of TCP variants under different node speeds using OPNET simulator
Meng et al. Research on TCPW improvement in hybrid network
KR100915996B1 (en) Method and Terminal for Controlling Transmission according Bandwidth Change
Chang et al. Cross-layer-based adaptive TCP algorithm in 4G packet service LTE-advanced relaying communications
KR20090050125A (en) Apparatus and method for tcp performance enhancement in wireless network

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C02 Deemed withdrawal of patent application after publication (patent law 2001)
WD01 Invention patent application deemed withdrawn after publication

Application publication date: 20121003