CN102291389A - Cross-layer congestion control method in satellite network - Google Patents
Cross-layer congestion control method in satellite network Download PDFInfo
- Publication number
- CN102291389A CN102291389A CN2011101970302A CN201110197030A CN102291389A CN 102291389 A CN102291389 A CN 102291389A CN 2011101970302 A CN2011101970302 A CN 2011101970302A CN 201110197030 A CN201110197030 A CN 201110197030A CN 102291389 A CN102291389 A CN 102291389A
- Authority
- CN
- China
- Prior art keywords
- congestion
- layer
- ack
- tcp
- congested
- 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
Links
Images
Abstract
The invention relates to a cross-layer congestion control designing method in a satellite network, which is a solution for enhancing TCP (Transmission Control Protocol) performance by using a cross-layer design in the satellite network, is mainly used for solving the influence on TCP due to the high error code environment of the satellite network, ensures that the TCP can predict congestion in time and rightly judge the data packet loss caused by the network congestion and error codes, and belongs to the field of the congestion control field of the satellite network. The cross-layer congestion control method is characterized in that the cross-layer interaction between a TCP layer and a link layer is realized in the invention, and the queue state of the link layer can be transmitted to the TCP layer; the TCP layer can judge the congestion condition of a link according to the state of the queue, and thus adopt a corresponding strategy; and finally the effect of TCP performance enhancement is achieved.
Description
Technical field
The present invention be a kind of in the satellite network performance solution of transport layer protocol.Be mainly used in and improve the performance of Transmission Control Protocol in satellite network, belong to the technical field of the congested control of satellite network transport layer.
Background technology
Along with popularizing of Internet, the quantity of last user of Internet and application is all increasing rapidly, when a plurality of users during greater than the actual transmittability of network, will inevitably cause the generation of network congestion to the total demand of network.Though the congested shortage of resources that comes from increases resource and can not avoid congested generation, sometimes even can the aggravate congestion degree.For example, increase the gateway buffer memory and can prevent or alleviate outwardly because the congested grouping that causes abandons, but along with the increase of buffer memory, the also corresponding increase of time delay end to end.Because the duration of grouping is limited, overtime grouping needs to retransmit equally.Therefore, thus excessive spatial cache might make total delay surpass the value that end system retransmits clock causes packet retransmissions present.These groupings have wasted the available bandwidth of network, have increased the weight of congested on the contrary.
The congestion control mechanism that uses on the Internet all is the TCP congestion control mechanism basically at present, and it is a kind of controlling mechanism based on end system.In the Internet early stage of development, mainly be to finish by of the flow control of Transmission Control Protocol middle-end to end group in sliding window, also progressively increased startup slowly, Congestion Avoidance in the flow algorithm of TCP, retransmitted and fast quick-recovery scheduling algorithm fast, in the hope of network traffics are controlled.Along with the development of the abundant and technology of application demand, the researcher begins to recognize that thinking to rely on fully the strategy and the algorithm that are implemented on the terminal system is difficult to satisfy increasing complicated applications demand.So people turn to intermediary device such as router in the network to attentiveness, expectation realizes the host terminal target that is beyond one's reach by the function that strengthens them.With regard to congested control, the network intermediate node might be more timely, even shift to an earlier date the congestion state of accurate awareness network, and implement effective policy in resource management according to this, makes network avoid congested effectively, or recover from serious congestion state as early as possible.
The congestion control mechanism that uses in network node at present mainly contains two classes: queue management and queue scheduling, wherein queue management can be divided into passive queue management (Passive Queue Management again, PQM) and active queue management (Active Queue Management, AQM).Traditional router adopts the most typical " (Drop-Tail) truncates " management strategy in the passive formation algorithm, and when grouping arrived, if buffer queue is own full, router then abandoned this grouping." truncate " though strategy is simple, be easy to produce lasting full queue state, even cause the global synchronization of Business Stream the deadlock and the Business Stream of buffer memory.Floyd has proposed famous earlier detection at random, and (Random Early Detection, RED) algorithm improved " truncating " algorithm on the router effectively, also become unique candidate algorithm of router design simultaneously in 1993.Though the validity of RED algorithm has been passed through the checking of some practices, still there are some defectives, as fairness and stability problem, and the performance sensitive of its algorithm is in selection of parameter etc.These problems at the existence of RED algorithm, numerous scholar both domestic and external has carried out a large amount of research to it, and many more outstanding derivation algorithms have appearred, BLUE, REM, Stabilized-RED, Active RED and GREEN etc. are wherein more typically arranged, the RED algorithm has had some improvement although these algorithms are compared, but they directly are applied in carry out congested control in the satellite network, still have a lot of deficiencies.
Therefore, at characteristics such as the dynamic of satellite network and high bit-error, large propagation time delay, bandwidth be asymmetric, need in conjunction with the thought of striding layer resource optimization, design a kind of congestion control mechanism flexibly and make that the TCP layer can be according to network real time status predict congestion and can distinguish error code and the congested packet loss that causes exactly, thereby take corresponding congested adjustment strategy, guarantee utilization rate of network resource, improve the transmission performance of TCP in the satellite network.
Summary of the invention
Technical problem:The purpose of this invention is to provide in a kind of satellite network in conjunction with the jamming control method of striding layer design, solve in the satellite network TCP predict congestion and judge the data-bag lost that network congestion and error code cause exactly, thereby improve the performance of Transmission Control Protocol in the satellite network.
Technical scheme:This method is a kind of method of tactic, based on the environment of satellite network, by striding the method and the strategy of layer design, junction link layer protocol, improve the congestion avoidance algorithm of TCP, it is reduced because of the link transmission packet loss of makeing mistakes cause the reduction of TCP send window, throughput to reduce.
Present protocol stack mechanism is the service that a certain layer protocol is provided by following surface layer, then provides service to more high-rise, but there is a lot of shortcomings in this mode.At the TCP layer, the mistake that congestion information produced in the wireless channel will cause losing of IP bag, cause bit rate to reduce.After the incident of losing takes place, will spend long time and go to repair, may cause the interruption of TCP after especially bulk information is lost.System will expend several two-way times (RTT) and just can return to the level before interrupt event takes place this moment.In order to address these problems, the method that a kind of TCP layer is striden layer design has been proposed, make between itself and the link layer and transmit information specific, thereby adapt, thereby make system can satisfy the demand of miscellaneous service with satellite communication environment.
What this method realized is layer design of striding of link layer and TCP layer, and the queue management method that uses message rate and RED to combine in the link layer passes to the TCP layer with real-time quene state, the information adjustment congestion control policy of TCP layer by receiving.
That strides in the satellite network that the congested controlling Design method of layer realize transport layer TCP and link layer strides layer design, makes packet loss that TCP can predict congestion also can correctly judge error code and congested generation to adjust send window in time, and wherein TCP layer design procedure is as follows:
When step 1. is received a new bag when transport layer, confirmation message ack type is judged, is then withdrawed from if not the ack bag,
Step 2. judges whether the new bag receive is the corpse bag that connected last time, if the corpse grouping, discharges this packetized memory and withdraws from,
That step 3. is received if ack bag wraps number with ack and adds 1, and the time of record opposite end when sending out this ack bag,
If the sequence number of this ack of step 4. bag is greater than the ack sequence number of last registration, expression is new ack, and congestion window is added 1, begins then to transmit next time, and congestion window is along with the number of transmissions is pressed the index law growth; When congestion window rises to when beginning threshold value slowly, just change the execution congestion avoidance algorithm into, congestion window is pressed linear rule and is increased,
If step 5. is received the ack of one or two repetition, only adopt re-transmission policy, promptly then also do not receive affirmation when the re-transmission time of timer setting, just retransmit this message segment,
Step 6. is when receiving the ack of three or above repetition, if the congestion flag that link layer transmits is shown as link congestion, be that the congestion flag position is 1 o'clock, call quick re-transmission policy, promptly the source end needn't wait for that RTO is overtime, immediately the packet of retransmission of lost, slowly start half that threshold value is set to the current congestion window size simultaneously, and congestion window is kept to original half, otherwise only adopts re-transmission policy
It is as follows to realize striding layer step of design in the link layer:
Step 7. is when receiving packet, judge the value of average queue length Qavg, when less than the predefined thresholding Qmin of earlier detection algorithm RED at random the time, continue to judge total message arrival rate A, if total message rate A is less than the bandwidth C of output link, formation is put in grouping, and congestion flag is set to non-congested; Otherwise, be congested with probability P=1-C/A packet discard with congestion flag is set,
Step 8. when the value of Qavg more than or equal to Qmin, and during less than Qmax, if total message rate A is less than the bandwidth C of output link, call function calculating probability P is with probability
Abandon grouping, and congestion flag is set to non-congested; Otherwise P gets 1 and 1-
C/A+
In smaller value, then with the probability P packet discard with the congestion flag position is set,
Step 9. is when the value of Qavg more than or equal to the predefined thresholding Qmax of RED the time, if total message rate A is less than the bandwidth C of output link, call function calculating probability P is with probability
Abandon grouping and the congestion flag position is set; Otherwise, directly abandon the packet of arrival, and congestion flag is set to congested.
Beneficial effect:This method is based on striding layer design, the queue state information of link layer is passed to the TCP layer, make the TCP layer when receiving repetition ack, can judge the congestion condition of link according to the queue state information of link layer, make and TCP predict congestion and can correctly judge the data-bag lost that network congestion and error code cause in time improved the throughput and the bandwidth availability ratio of communication link.
Description of drawings
Fig. 1 is the schematic flow sheet after link layer is received packet.
Fig. 2 is the schematic flow sheet after the TCP layer is received packet.
Embodiment
Method flow:
1. link layer strides a layer design
The tail drop algorithm is adopted in traditional queue management, causes problems such as global synchronization and buffering area easily are filled easily.The basic thought of RED congestion control mechanism is to survey congested by the average length of monitoring router output port queue, in case find congested approaching, it is congested just to select connection to notify randomly, they are overflowed in formation reduce congestion window before causing packet loss, reduce sending data speed, thus alleviating network congestion.Because RED is based on the fifo queue scheduling strategy, and just abandons the packet that is just entering router, so it implements also comparatively simple.The RED algorithm mainly comprises two parts: the probability that calculates average queue length and calculating discarded packets.
(1) calculating of average queue length
Wherein
Expression
Average queue length constantly,
Expression current queue length,
The weighted value of expression average queue length,
Size determined the sensitivity that RED changes input traffic,
(2) calculating of drop probabilities
When grouping arrives formation, if average queue length
Less than minimum threshold
, grouping safety enters formation; When
Greater than maximum threshold
, abandon all and arrive message; If average queue length is positioned at
With
Between, then be calculated as follows packet dropping probability:
Wherein,
Be maximum drop probability,
It is the packet count that success is continuously transmitted.In the actual realization of algorithm, scatter more evenly in order to make the grouping that is labeled, need marking probability
Do to revise to obtain
As the real marking probability.
Supposing has
nIndividual data stream wherein flows through same router
i tThe speed that constantly arrives router is
, the bandwidth of router output link is
C, the message arrival rate is
If
, then the message of all arrival need not to wait for, is sent out away immediately; Otherwise, if
, what part arrived is just needed process to wait in line by message, just can send.And if this state continuance goes down, the message of waiting in line will increase gradually, and buffer area will be filled very soon, finally overflows and causes packet loss.In order to protect instantaneous burst flow, the message arrival rate is calculated as follows:
Wherein,
TBe the sampling interval,
lBe data flow
i TThe number of the packet that arrives in the time,
With
Be respectively stream
iThe packet arrival rate of current time and previous moment,
KBe Control Parameter.
This method is the congestion condition that comes detecting link by average queue length that monitors the router input port and message arrival rate, congestion condition with link passes to the TCP layer then, thereby make the TCP layer can judge when packet loss produces that by error code still be congested causing, in time send window being made adjustment, and then improve the throughput of TCP.May further comprise the steps based on the link layer formation control of striding layer design:
Step 2. when the value of Qavg more than or equal to Qmin, and during less than Qmax, if total message rate A is less than the bandwidth C of output link, call function calculating probability P is with probability
Abandon grouping, and congestion flag is set to non-congested; Otherwise P gets 1 and (1-
C/A+
) in smaller value, then with the probability P packet discard with the congestion flag position is set.
Step 3. is when the value of Qavg more than or equal to the predefined thresholding Qmax of RED the time, if total message rate A is less than the bandwidth C of output link, call function calculating probability P is with probability
Abandon grouping and the congestion flag position is set; Otherwise, directly abandon the packet of arrival, and congestion flag is set to congested.
Layer stride a layer design
The congested control of TCP is a kind of controlling mechanism end to end, and this machine-processed target is to make network be in the state of high-throughput, low delay, avoids congested generation as far as possible.In case it is congested that network takes place, and network can be recovered from congested by this mechanism, enters normal operating state.The congested control of TCP can be divided into startup slowly, Congestion Avoidance usually, retransmit and fast quick-recovery four-stage fast.Congestion avoidance algorithm need be kept two variablees to each connection: congestion window cwnd and one start threshold values ssthresh slowly.The congestion avoidance algorithm step is as follows:
(1) to a given connection, initialization cwnd is 1 message segment, and ssthresh is 65535 bytes.
(2) output of TCP can not surpass the size of cwnd and reception advertised window.The flow control that transmit leg uses during Congestion Avoidance.
(3) when congested generation, ssthresh is set to half of current window size.In addition, if overtime re-transmission, then cwnd is set to 1 message segment.
(4) when new data are confirmed by the other side, just increase cwnd, but the method that increases depends on and whether starts slowly or Congestion Avoidance.If cwnd is less than or equal to ssthresh, then starts slowly, otherwise carry out Congestion Avoidance.
If receive 3 or above repetition ack continuously,, overflow and need not the wait timeout timer with regard to the packet literary composition section of retransmission of lost.Step is as follows:
(1) when receiving the ack of the 3rd repetition, ssthresh is set to half of current congestion window cwnd, the message segment of retransmission of lost.
When (2) receiving the ack of another repetition, cwnd increases by 1 message segment size and sends 1 grouping at every turn.
(3) when the next one confirmed that the ack of new data arrives, it was ssthresh that cwnd is set, and this ack should be the affirmation to retransmitting in the step 1 in two-way time after retransmitting.In addition, this ack also should be the affirmation to all the middle message segments between the ack of the grouping of losing and first repetition of receiving.
As can be seen from the above, traditional TCP design is when receiving three repetition ack, can not correctly judge packet loss still is congested causing by error code, therefore directly reduce congestion window, make whole link utilization reduce, be by striding the TCP jamming control method of layer design below, may further comprise the steps:
When step 1. is received a new bag when transport layer, ack bag type is judged, then withdrawed from if not the ack bag.
Step 2. judges whether the new bag receive is the corpse bag that connected last time, if the corpse grouping, discharges this packetized memory and withdraws from.
That step 3. is received if ack bag wraps number with ack and adds 1, and the time of record opposite end when sending out this ack bag.
If the sequence number of this ack of step 4. bag is greater than the ack sequence number of last registration, expression is new ack, adjusts congestion window.
If step 5. is received the ack of one or two repetition, only adopt re-transmission policy.
Step 6. is when receiving the ack of three or above repetition, if the congestion flag that link layer transmits is shown as link congestion, promptly the congestion flag position is 1 o'clock, calls quick re-transmission policy, otherwise only adopts re-transmission policy.
Claims (1)
1. stride the congested controlling Design method of layer in a satellite network, it is characterized in that realizing layer design of striding of transport layer TCP and link layer, make TCP can predict congestion and the packet loss that can correctly judge error code and congested generation to adjust send window in time, wherein TCP layer design procedure is as follows:
When step 1. is received a new bag when transport layer, confirmation message ack type is judged, is then withdrawed from if not the ack bag,
Step 2. judges whether the new bag receive is the corpse bag that connected last time, if the corpse grouping, discharges this packetized memory and withdraws from,
That step 3. is received if ack bag wraps number with ack and adds 1, and the time of record opposite end when sending out this ack bag,
If the sequence number of this ack of step 4. bag is greater than the ack sequence number of last registration, expression is new ack, and congestion window is added 1, begins then to transmit next time, and congestion window is along with the number of transmissions is pressed the index law growth; When congestion window rises to when beginning threshold value slowly, just change the execution congestion avoidance algorithm into, congestion window is pressed linear rule and is increased,
If step 5. is received the ack of one or two repetition, only adopt re-transmission policy, promptly then also do not receive affirmation when the re-transmission time of timer setting, just retransmit this message segment,
Step 6. is when receiving the ack of three or above repetition, if the congestion flag that link layer transmits is shown as link congestion, be that the congestion flag position is 1 o'clock, call quick re-transmission policy, promptly the source end needn't wait for that RTO is overtime, immediately the packet of retransmission of lost, slowly start half that threshold value is set to the current congestion window size simultaneously, and congestion window is kept to original half, otherwise only adopts re-transmission policy
It is as follows to realize striding layer step of design in the link layer:
Step 7. is when receiving packet, judge the value of average queue length Qavg, when less than the predefined thresholding Qmin of earlier detection algorithm RED at random the time, continue to judge total message arrival rate A, if total message rate A is less than the bandwidth C of output link, formation is put in grouping, and congestion flag is set to non-congested; Otherwise, be congested with probability P=1-C/A packet discard with congestion flag is set,
Step 8. when the value of Qavg more than or equal to Qmin, and during less than Qmax, if total message rate A is less than the bandwidth C of output link, call function calculating probability P is with probability
Abandon grouping, and congestion flag is set to non-congested; Otherwise P gets 1 and 1-
C/A+
In smaller value, then with the probability P packet discard with the congestion flag position is set,
Step 9. is when the value of Qavg more than or equal to the predefined thresholding Qmax of RED the time, if total message rate A is less than the bandwidth C of output link, call function calculating probability P is with probability
Abandon grouping and the congestion flag position is set; Otherwise, directly abandon the packet of arrival, and congestion flag is set to congested.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2011101970302A CN102291389A (en) | 2011-07-14 | 2011-07-14 | Cross-layer congestion control method in satellite network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2011101970302A CN102291389A (en) | 2011-07-14 | 2011-07-14 | Cross-layer congestion control method in satellite network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102291389A true CN102291389A (en) | 2011-12-21 |
Family
ID=45337498
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2011101970302A Pending CN102291389A (en) | 2011-07-14 | 2011-07-14 | Cross-layer congestion control method in satellite network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102291389A (en) |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102664867A (en) * | 2012-03-15 | 2012-09-12 | 南京邮电大学 | Method for enhancing transmission protocols in satellite communication system |
| CN104038438A (en) * | 2013-12-03 | 2014-09-10 | 江苏达科信息科技有限公司 | Explicit congestion control method improved on the basis of conventional transmission control protocol |
| WO2014141006A1 (en) * | 2013-03-15 | 2014-09-18 | International Business Machines Corporation | Scalable flow and congestion control in a network |
| CN104093169A (en) * | 2014-06-10 | 2014-10-08 | 北京邮电大学 | MAC-TCP cross-layer optimization congestion control method and system |
| CN104092625A (en) * | 2014-05-28 | 2014-10-08 | 中南大学 | Adaptive request batch scheduling method used in DCN |
| US9104643B2 (en) | 2013-03-15 | 2015-08-11 | International Business Machines Corporation | OpenFlow controller master-slave initialization protocol |
| US9118984B2 (en) | 2013-03-15 | 2015-08-25 | International Business Machines Corporation | Control plane for integrated switch wavelength division multiplexing |
| WO2016065676A1 (en) * | 2014-10-27 | 2016-05-06 | 北京华夏创新科技有限公司 | Optimization method and system based on throughput rate and packet loss control cwnd |
| US9407560B2 (en) | 2013-03-15 | 2016-08-02 | International Business Machines Corporation | Software defined network-based load balancing for physical and virtual networks |
| US9590923B2 (en) | 2013-03-15 | 2017-03-07 | International Business Machines Corporation | Reliable link layer for control links between network controllers and switches |
| US9609086B2 (en) | 2013-03-15 | 2017-03-28 | International Business Machines Corporation | Virtual machine mobility using OpenFlow |
| US9769074B2 (en) | 2013-03-15 | 2017-09-19 | International Business Machines Corporation | Network per-flow rate limiting |
| CN107395522A (en) * | 2017-08-25 | 2017-11-24 | 深圳市华讯方舟空间信息产业科技有限公司 | Loss treating method and computer-readable recording medium based on Transmission Control Protocol |
| CN107659386A (en) * | 2017-09-19 | 2018-02-02 | 深圳乐腾无线科技有限公司 | A kind of method for solving to apply packet loss under WiFi in real time based on router |
| CN109194450A (en) * | 2018-08-15 | 2019-01-11 | 西安电子科技大学 | SNACK-P answering system and method, the wireless communication system of Incorporate network communication |
| CN113543208A (en) * | 2021-05-26 | 2021-10-22 | 浙江大华技术股份有限公司 | Wireless transmission control method, wireless communication system, and computer storage medium |
| CN117544567A (en) * | 2024-01-09 | 2024-02-09 | 南京邮电大学 | Memory transfer integrated RDMA data center congestion control method |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101854297A (en) * | 2010-05-21 | 2010-10-06 | 南京邮电大学 | Method for designing transmission control protocol (tcp) cross-layer in satellite network |
-
2011
- 2011-07-14 CN CN2011101970302A patent/CN102291389A/en active Pending
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101854297A (en) * | 2010-05-21 | 2010-10-06 | 南京邮电大学 | Method for designing transmission control protocol (tcp) cross-layer in satellite network |
Non-Patent Citations (2)
| Title |
|---|
| 谢立峰等: "采用基于速率的缓存控制的CSFQ算法", 《浙江大学学报（工学版）》, vol. 38, no. 7, 31 July 2004 (2004-07-31), pages 851 - 855 * |
| 高文宇等: "基于RED的带宽分配算法", 《计算机工程》, vol. 31, no. 19, 31 October 2005 (2005-10-31) * |
Cited By (30)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102664867B (en) * | 2012-03-15 | 2014-11-19 | 南京邮电大学 | Method for enhancing transmission protocols in satellite communication system |
| CN102664867A (en) * | 2012-03-15 | 2012-09-12 | 南京邮电大学 | Method for enhancing transmission protocols in satellite communication system |
| GB2525832B (en) * | 2013-03-15 | 2017-03-01 | Ibm | Scalable flow and congestion control in a network |
| US9407560B2 (en) | 2013-03-15 | 2016-08-02 | International Business Machines Corporation | Software defined network-based load balancing for physical and virtual networks |
| US9596192B2 (en) | 2013-03-15 | 2017-03-14 | International Business Machines Corporation | Reliable link layer for control links between network controllers and switches |
| WO2014141006A1 (en) * | 2013-03-15 | 2014-09-18 | International Business Machines Corporation | Scalable flow and congestion control in a network |
| US9104643B2 (en) | 2013-03-15 | 2015-08-11 | International Business Machines Corporation | OpenFlow controller master-slave initialization protocol |
| US9110866B2 (en) | 2013-03-15 | 2015-08-18 | International Business Machines Corporation | OpenFlow controller master-slave initialization protocol |
| US9118984B2 (en) | 2013-03-15 | 2015-08-25 | International Business Machines Corporation | Control plane for integrated switch wavelength division multiplexing |
| GB2525832A (en) * | 2013-03-15 | 2015-11-04 | Ibm | Scalable flow and congestion control in a network |
| US9614930B2 (en) | 2013-03-15 | 2017-04-04 | International Business Machines Corporation | Virtual machine mobility using OpenFlow |
| US9609086B2 (en) | 2013-03-15 | 2017-03-28 | International Business Machines Corporation | Virtual machine mobility using OpenFlow |
| US9444748B2 (en) | 2013-03-15 | 2016-09-13 | International Business Machines Corporation | Scalable flow and congestion control with OpenFlow |
| US9503382B2 (en) | 2013-03-15 | 2016-11-22 | International Business Machines Corporation | Scalable flow and cogestion control with openflow |
| US9769074B2 (en) | 2013-03-15 | 2017-09-19 | International Business Machines Corporation | Network per-flow rate limiting |
| US9590923B2 (en) | 2013-03-15 | 2017-03-07 | International Business Machines Corporation | Reliable link layer for control links between network controllers and switches |
| CN104038438A (en) * | 2013-12-03 | 2014-09-10 | 江苏达科信息科技有限公司 | Explicit congestion control method improved on the basis of conventional transmission control protocol |
| CN104092625B (en) * | 2014-05-28 | 2017-07-18 | 中南大学 | A kind of self adaptation being used in DCN asks dispatching method in batches |
| CN104092625A (en) * | 2014-05-28 | 2014-10-08 | 中南大学 | Adaptive request batch scheduling method used in DCN |
| CN104093169A (en) * | 2014-06-10 | 2014-10-08 | 北京邮电大学 | MAC-TCP cross-layer optimization congestion control method and system |
| WO2016065676A1 (en) * | 2014-10-27 | 2016-05-06 | 北京华夏创新科技有限公司 | Optimization method and system based on throughput rate and packet loss control cwnd |
| CN107395522A (en) * | 2017-08-25 | 2017-11-24 | 深圳市华讯方舟空间信息产业科技有限公司 | Loss treating method and computer-readable recording medium based on Transmission Control Protocol |
| CN107395522B (en) * | 2017-08-25 | 2018-10-12 | 深圳市华讯方舟空间信息产业科技有限公司 | Loss treating method based on Transmission Control Protocol and computer readable storage medium |
| CN107659386A (en) * | 2017-09-19 | 2018-02-02 | 深圳乐腾无线科技有限公司 | A kind of method for solving to apply packet loss under WiFi in real time based on router |
| CN109194450A (en) * | 2018-08-15 | 2019-01-11 | 西安电子科技大学 | SNACK-P answering system and method, the wireless communication system of Incorporate network communication |
| CN109194450B (en) * | 2018-08-15 | 2021-05-04 | 西安电子科技大学 | SNACK-P response system and method for heaven-earth integrated network communication and wireless communication system |
| CN113543208A (en) * | 2021-05-26 | 2021-10-22 | 浙江大华技术股份有限公司 | Wireless transmission control method, wireless communication system, and computer storage medium |
| CN113543208B (en) * | 2021-05-26 | 2024-04-19 | 浙江大华技术股份有限公司 | Wireless transmission control method, wireless communication system, and computer storage medium |
| CN117544567A (en) * | 2024-01-09 | 2024-02-09 | 南京邮电大学 | Memory transfer integrated RDMA data center congestion control method |
| CN117544567B (en) * | 2024-01-09 | 2024-03-19 | 南京邮电大学 | Memory transfer integrated RDMA data center congestion control method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102291389A (en) | Cross-layer congestion control method in satellite network | |
| CN109120544B (en) | Transmission control method based on host end flow scheduling in data center network | |
| Floyd | A report on recent developments in TCP congestion control | |
| US7859996B2 (en) | Intelligent congestion feedback apparatus and method | |
| CN101052043B (en) | TCP sending algorithm based on sending window and reciprocating time | |
| US20140043994A1 (en) | Providing Feedback To Media Senders Over Real Time Transport Protocol (RTP) | |
| CN101854297B (en) | Method for designing transmission control protocol (tcp) cross-layer in satellite network | |
| Alipio et al. | RT-CaCC: A reliable transport with cache-aware congestion control protocol in wireless sensor networks | |
| Ni et al. | OCPS: Offset compensation based packet scheduling mechanism for multipath TCP | |
| CN103957121A (en) | Cross-layer TCP optimization method oriented to aerospace information network | |
| Oyeyinka et al. | TCP window based congestion control-slow-start approach | |
| CN101969432A (en) | Random backoff based control method of TCP (Transmission Control Protocol) congestion window | |
| CN102801502B (en) | Packet loss method in LTE and LTE-A system based on RED algorithm | |
| CN102088414B (en) | Method for managing passive queue by virtue of combining random drop and drop front | |
| CN112689268B (en) | Method and system for improving multi-route data transmission efficiency by formation unmanned aerial vehicle networking | |
| US20130188511A1 (en) | Method and system for controlling transmission rate of datagrams in a packet-switched network | |
| Kumar et al. | A multipath packet scheduling approach based on buffer acknowledgement for congestion control | |
| Mohanty et al. | Transport protocols in wireless sensor networks | |
| Dayma et al. | Improved TCP Hybla: A TCP enhancement for link with high RTT and error rate | |
| CN102082735B (en) | Method for managing passive queue by abandoning head for N times | |
| Shi | Token-based congestion control: Achieving fair resource allocations in P2P networks | |
| Gharegozi | The study of ECN application effect at the performance improvement of RED | |
| Fukuhara et al. | SRF TCP: A TCP-friendly and fair congestion control method for high-speed networks | |
| Cheng et al. | Improving the ramping up behavior of TCP slow start | |
| CN103188166A (en) | Flow control method utilizing TCP (Transmission Control Protocol)-self-consciousness Random Early Detection (RED) |
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: 20111221 |