CN104954206A - Disorder analyzing method for multipath parallel transmission system - Google Patents

Disorder analyzing method for multipath parallel transmission system Download PDF

Info

Publication number
CN104954206A
CN104954206A CN201510464585.7A CN201510464585A CN104954206A CN 104954206 A CN104954206 A CN 104954206A CN 201510464585 A CN201510464585 A CN 201510464585A CN 104954206 A CN104954206 A CN 104954206A
Authority
CN
China
Prior art keywords
packet
transmission system
parallel transmission
transmitting terminal
order
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.)
Granted
Application number
CN201510464585.7A
Other languages
Chinese (zh)
Other versions
CN104954206B (en
Inventor
陶洋
赫前进
邓行
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.)
Chongqing University of Post and Telecommunications
Original Assignee
Chongqing University of Post and Telecommunications
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 Chongqing University of Post and Telecommunications filed Critical Chongqing University of Post and Telecommunications
Priority to CN201510464585.7A priority Critical patent/CN104954206B/en
Publication of CN104954206A publication Critical patent/CN104954206A/en
Application granted granted Critical
Publication of CN104954206B publication Critical patent/CN104954206B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Abstract

The invention relates to a disorder analyzing method for a multipath parallel transmission system and belongs to the technical filed of communication. The method comprises the following steps: step 1), in the multipath parallel transmission system, after a data packet is transmitted to a sending end by an application layer, the sending end is used for sending out the data packet immediately, and the data packet reaches a receiving end after path propagation time delay d; step 2), analyzing is performed on the disorder of the multipath parallel transmission system, and the probability of the disorder caused during a package transmission procedure is calculated. Further, impacts caused by transmission time delay of the sending end and the queue time delay on the phenomenon of disorder of the data packet can be further considered, and the probability of disorder caused during the data packet transmission procedure is calculated. The disorder analyzing method for the multipath parallel transmission system, provided by the invention, can be used for estimating the probability of the disorder of the data packet during a multipath parallel transmission process under the current network environment, and by reasonably controlling and reducing the phenomenon of data disorder, the performance of multipath parallel transmission in a heterogeneous network can be improved.

Description

A kind of out of order analytical method of multipath parallel transmission system
Technical field
The invention belongs to communication technical field, relate to a kind of out of order analytical method of multipath parallel transmission system.
Background technology
Along with the development of mobile communication technology, Radio Access Network is day by day diversified, and the terminal equipment being equipped with multiple network interface also gets more and more, for the investigation and application of multipath parallel transmission technology provides the foundation.Multipath parallel transmission technology, by parallel transmission in allocation of packets to multilink, is gathered by the bandwidth under parallel transmission and is increased throughput.But the asymmetry in heterogeneous network between different link in bandwidth, time delay etc., the packet that can result through different link transmission cannot arrive receiving terminal according to the order sent, and occurs the phenomenon of data packet disorder.
Data packet disorder (packet reordering) problem is due to the propagation delay time difference on different path, the order that packet arrives receiving terminal is not identical with sending order, and the packet of rear transmission earlier may arrive receiving terminal than the packet sent before.But, specify to only have the packet arrived according to the order of sequence could up submit application layer and process in Stream Control Transmission Protocol.When data packet disorder phenomenon is more serious, this just makes out of order packet can be trapped in receiving terminal buffer memory, cannot submit upper layer application in time and process, increase the propagation delay time of grouping, reduces effective throughput performance of multipath parallel transmission.
Under heterogeneous network environment, the path used in multipath parallel transmission may belong to different access networks, such as WLAN, UMTS, LTE, satellite network etc., and the difference of various network characteristic causes different path to there are differences in propagation delay time and bandwidth; Even and if in same network, different routed paths also can cause the propagation delay time on each paths different.Because the propagation delay time of each paths there are differences, the grouping on the path that time delay is less earlier will arrive receiving terminal than the grouping on other paths.If the transmitted in packets sequence number first arriving receiving terminal is larger, and the grouping that now transmission sequence number is less is still transmitted in a network and is not yet arrived, the grouping then arrived in advance cannot be submitted to upper layer application according to the order of sequence and process, this part grouping will be stored in the buffer area of receiving terminal, submits according to the order of sequence together after waiting for the grouping arrival that transmission sequence number is less to upper layer application.
Existing multipath parallel transmission technology does not take into full account the isomerism of network, is difficult to get a desired effect when using under heterogeneous network environment.
It is out of order that data disorder phenomenon can divide for the out of order and packet loss caused by the path difference opposite sex to cause.
Current, the out of order performance analysis models causing out of order multipath parallel transmission system because path difference is different in nature mainly utilizes the cumulative distribution function of path end-to-end time delay, gives the computing formula of data packet disorder probability.But, this analytical model tentation data bag when modeling is just sent out away after arriving transmitting terminal by application layer immediately, only considered the impact of propagated time delay for data disorder phenomenon, have ignored the influencing factors such as the propagation delay time of packet, queue time delay, processing delay, thus there is certain error in the numerical result causing model to export and simulation result.
Summary of the invention
In view of this, the object of the present invention is to provide a kind of out of order analytical method of multipath parallel transmission system, can there is out of order probability in calculated data bag transmitting procedure in the method, and reduce data disorder phenomenon by conservative control, improves the performance of multipath parallel transmission in heterogeneous network.
For achieving the above object, the invention provides following technical scheme:
An out of order analytical method for multipath parallel transmission system, there is the separate transmission path of N bar in multipath parallel transmission system, be respectively L between transmitting terminal and receiving terminal 1, L 2..., L k... L n, N>=2; On each paths, propagated time delay meets d end to end 1≤ d 2≤ ... d k...≤d n; The method comprises the following steps:
Step 1) in multipath parallel transmission system, after packet is passed to transmitting terminal by application layer, Packet Generation is gone out by transmitting terminal immediately, packet experience propagated time delay d krear arrival receiving terminal;
Step 2) the out of order of multipath parallel transmission system is analyzed, there is out of order probability in calculated data bag transmitting procedure.
Further, packet i-1 and packet i leaves the time interval of transmitting terminal is △ t(i), △ ti ()=T (i)-T (i-1), T (i) leave the moment of transmitting terminal for packet i-1, T (i-1) leaves the moment of transmitting terminal for packet i-1.
Further, out of order probability occurring in described transmission of data packets is wherein, p kfor packet selects transmission path L kcarry out the probability transmitted,
The propagation delay time cumulative distribution function that D (d) is packet,
D ( d ) = &Sigma; d k < d p k = 0 d &le; d 1 p 1 d 1 &le; d &le; d 2 p 1 + p 2 d 2 &le; d &le; d 3 ... ... 1 d > d N .
Further, described step 1) in multipath parallel transmission system, after packet is passed to transmitting terminal by application layer, the transmit queue being assigned to a certain transmission path according to Diffluence Algorithm by service distributing module is medium to be sent, be sent in the middle of network by transmitting terminal after transmission delay S, experience propagated time delay d krear arrival receiving terminal.
Further, △ ti () leaves the time interval of transmitting terminal, △ for packet i-1 and packet i t(i)=T (i)-T (i-1), T (i) leaves the moment of transmitting terminal for packet i, T (i-1) leaves the moment of transmitting terminal for packet i-1, T (i)=R (i)+S (i), R (i) is the moment of packet i arrival transmitting terminal, the transmission delay that S (i) experiences for packet i.
Further, out of order probability occurring in described transmission of data packets is
P ( &Delta; T ( i ) ) = &Sigma; k = 1 n P k &CenterDot; D ( d ( i - 1 ) &Delta; T ( i ) ) &Delta; T ( i ) = &Delta; R ( i ) + S ( i ) - S ( i - 1 )
P kfor packet selects transmission path L kthe probability carrying out transmitting is,
The propagation delay time cumulative distribution function that D (d) is packet,
D ( d ) = &Sigma; d k < d p k = 0 d &le; d 1 p 1 d 1 &le; d &le; d 2 p 1 + p 2 d 2 &le; d &le; d 3 ... ... 1 d > d N .
Further, the arrival process of described transmitting terminal packet meets Poisson distribution, and packet arrival rate is λ, and the size of packet obeys quantum condition entropy, and the transmission rate on kth bar concurrent path is μ k, the substream of data arrival rate be assigned on kth paths is λ k, λ k=p kλ,
Further, the transmission delay of described packet i experience
The transmission delay that packet i-1 experiences S ( i - 1 ) = Q ( i - 1 ) + M T U B ( i - 1 ) ,
Wherein, the queue length on the transmit path of Q (i-1) representative data bag i-1, the bandwidth of transmit path selected by B (i-1) representative data bag i-1, MTU is the length of a MTU.
Further, there is out of order probability and be in packet i in transmitting procedure
P ( &Delta; T ( i ) ) = &Sigma; k = 1 n P k &CenterDot; D ( d ( i - 1 ) &Delta; T ( i ) ) &Delta; T ( i ) = &Delta; R ( i ) + &Sigma; k = 1 n p k 1 &mu; k - &lambda; k - Q ( i - 1 ) + M T U B ( i - 1 ) .
Further, data disorder phenomenon is reduced, demand fulfillment constraints d (i)>=d (i-1)-△ t(i).
Beneficial effect of the present invention is: the out of order analytical method of a kind of multipath parallel transmission system provided by the invention, out of order probability can be there is in the method in calculated data bag transmitting procedure, and reduce data disorder phenomenon by conservative control, improve the performance of multipath parallel transmission in heterogeneous network.
Accompanying drawing explanation
In order to make the object, technical solutions and advantages of the present invention clearly, below in conjunction with accompanying drawing, the present invention is described in further detail, wherein:
Fig. 1 is the out of order producing cause schematic diagram of multipath transmitting data in parallel;
Fig. 2 be ignore propagation delay time and queue time delay transmission delay on multipath transmitting data in parallel out of order affect schematic diagram;
Fig. 3 be consider propagation delay time and queue time delay transmission delay on multipath transmitting data in parallel out of order affect schematic diagram;
Fig. 4 is the out of order performance analysis models schematic diagram of multipath parallel transmission;
Fig. 5 is transmitting terminal packet shunting model schematic diagram.
Embodiment
Below in conjunction with accompanying drawing, the preferred embodiments of the present invention are described in detail.
The invention provides a kind of out of order analytical method of multipath parallel transmission system, out of order probability can be there is in the method in calculated data bag transmitting procedure, and reduce data disorder phenomenon by conservative control, improve the performance of multipath parallel transmission in heterogeneous network.
An effective multipath parallel transmission schemes under heterogeneous network environment, the related mechanism that can reduce data disorder phenomenon adverse effect as far as possible must be comprised, and carry out comprehensively the out of order situation of multi-path transmission system and calculate accurately being the prerequisite designing related mechanism.
Fig. 1 is the out of order producing cause schematic diagram of multipath transmitting data in parallel.
Suppose to establish three concurrent transmission paths between transmitting terminal and receiving terminal, packet adopts repeating query mode to distribute between three paths, and the propagation delay time on three paths meets d 1< < d 2< < d 3, then the transmission sequence number transmitted by path 1 be 1,4 and 7 packet preferentially will arrive receiving terminal.
Not yet arrive because now packet 2 still transmits in a network, 4 and 7 can only be temporary in receiving terminal buffer memory, wait for that the packet that transmission sequence number is less arrives laggard rearrangement sequence.
Following arrival be on path 2 transmission transmission sequence number be 2,5 and 8 packet, due to 3 fail arrive, 5 and 8 also can only wait in the buffer.
Along with the increase of the packet left in receiving terminal buffer memory, prescribe a time limit when the packet accumulation in buffer memory reaches on spatial cache, receiving terminal buffer memory will be caused to block, receiving terminal can abandon all packets next arrived, and notifying that transmitting terminal performs Fast retransmission operation, through receiving terminal buffer memory blocks end.
End-to-end time delay in transmission of data packets is made up of four parts, respectively: propagation delay (propagation delay), propagation delay time (transmission delay), processing delay (processing delay) and queuing delay (queuing delay).By from propagation delay time and queue time delay, model of the present invention is described for the impact that data are out of order below.
Suppose between transmitting terminal and receiving terminal, to use two paths to carry out multipath parallel transmission.
Fig. 2 be ignore propagation delay time and queue time delay transmission delay on multipath transmitting data in parallel out of order affect schematic diagram.
The data distribution algorithm of transmitting terminal adopts the weight polling algorithm based on the path delay of time, because the propagation delay on path 1 and path 2 is than being 1:2, therefore transmitting terminal according to the shunt ratio of 2:1 to path 1 and path 2 distribute data bag, if packet arrival interval is 20ms on path 1, on path 2, packet arrival interval is 40ms.
Propagation delay on path 1 and path 2 is 50ms and 100ms respectively, ignores transmitting terminal propagation delay time and queue time delay to the impact of data packet disorder.Get moment T as the observation moment, till cut-off time T, all packets sent all can arrive receiving terminal according to the order of sequence, and the phenomenon that data are out of order does not appear in receiving terminal.
Fig. 3 be consider propagation delay time and queue time delay transmission delay on multipath transmitting data in parallel out of order affect schematic diagram.
If increase the impact considering transmitting terminal propagation delay time and queue time delay, the propagation delay on path 1 and path 2 is still 50ms and 100ms, the propagation delay time in path 1 and queue time delay 10ms altogether, the propagation delay time in path 2 and queue time delay 50ms altogether.Increase the consideration to transmitting terminal propagation delay time and queue time delay, be equivalent to extend packet to be sent from entering transmit queue etc., successfully receive experienced overall transmission time Ttotal to receiving end, the Ttotal now on path 1 and path 2 is respectively 60ms and 150ms.
If now still adopt the weight polling algorithm based on the path delay of time to carry out transmitting terminal data dispatch, then, till cut-off time T, two secondary data disorder phenomenon occurring in transmitting procedure altogether, is caused by the packet delay being 3 and 6 by transmission sequence number arrives respectively.In the T moment, the packet being 6 due to transmission sequence number not yet arrives receiving terminal, therefore the packet that transmission sequence number is 7 will be trapped in receiving terminal buffer memory, cannot submit to upper layer application and process.
Fig. 4 is the out of order performance analysis models schematic diagram of multipath parallel transmission, and in multipath parallel transmission system, what packet occurred in multipath parallel transmission process out of orderly causes mainly due to the delay difference between many concurrent links.
Between transmitting terminal and receiving terminal, there is the transmission path that N (N>=2) bar is separate, be designated as L respectively 1, L 2..., L k... L n, the end-to-end propagation delay on each paths meets d 1≤ d 2≤ ... d k...≤d n.The Business Stream got off by application layer transmission is divided into packet not of uniform size after the service distributing module arriving transmitting terminal, each packet is separate in transmitting procedure, and selects a certain bar in N paths to send according to certain data scheduling algorithm.Remember that a certain packet selects transmission path L kthe probability carrying out transmitting is p k, that is this packet can with p kprobability to be assigned to end-to-end time delay be p klink on transmit.Data distribution scheduling strategy is described as:
P = { p 1 , p 2 , ... , p N | &Sigma; i = 1 N p i = 1 }
Wherein: p irepresent selecting paths L ias the probability in Packet Generation path.
Receiving terminal needs the packet to arriving via each paths transmission to process, and the packet arrived in order is upwards submitted to application layer according to the order of sequence and processes; The packet of out of order arrival is then temporary in buffer area, waits for that packet that all transmission sequence number are less than it is submitted to application layer after all arriving again together.
Suppose to there is not link failure packet loss, the packet that same path is transmitted can arrive receiving terminal in order according to sending order, and the packet that different path is transmitted disorder phenomenon may occur.Therefore, once data disorder phenomenon appears in receiving terminal, illustrate that out of order transmission sequence number continuous print packet occurs must come from different transmission paths.Data packet disorder in described out of order performance analysis models be by different path between asymmetry cause.
Need the variable-definition used as follows in analytic process, in order to the constraints of packet order transfer and the out of order new probability formula derivation of multipath transmitting data in parallel bag describe convenient, the propagation delay time that packet experiences at transmitting terminal by the present invention and queue time delay collectively transmission delay.
R (i)---packet i arrives the moment of transmitting terminal
ri ()---packet i-1 and packet i arrives the time interval of transmitting terminal
The transmission delay that S (i)---packet i experiences
T (i)---packet i leaves the moment of transmitting terminal
ti ()---packet i-1 and packet i leaves the time interval of transmitting terminal
D (i)---the propagation delay that packet i experiences in network transmission process
A (i)---packet i arrives the moment of receiving terminal, A (i)=T (i)+d (i)
ai ()---packet i-1 and packet i arrives the time interval of receiving terminal
Tentation data bag is sent on transmission link after arriving transmitting terminal immediately, namely packet arrives transmitting terminal in T (i) moment and is sent out immediately, arrive receiving terminal after experience propagated time delay d (i), ignore propagation delay time and queue time delay transmission delay to the out of order impact of multipath transmitting data in parallel.
Because packet leaves transmitting terminal in order, therefore the Packet Generation moment is directly proportional to the transmission sequence number of this packet, and the moment of the Packet Generation that transmission sequence number is larger is more late.Therefore, ensure that packet arrives receiving terminal according to the order of sequence, only need the moment allowing adjacent two data bag arrive receiving terminal to meet:
&ForAll; i &GreaterEqual; 0 , &Delta; A ( i ) = A ( i ) - A ( i - 1 ) &GreaterEqual; 0
I.e. T (i)+d (i) >=T (i-1)+d (i-1)
Therefore, d (i)>=d (i-1)-(T (i)-T (i-1))=d (i-1)-△ is had t(i)
Can obtain from above derivation, as long as the end-to-end time delay that the end-to-end time delay that packet i experiences is more than or equal to packet i-1 deducts the time interval that these two packets leave transmitting terminal, i.e. d (i)>=d (i-1)-△ ti () (constraints), so they just can arrive receiving terminal according to the order of sequence.
As can be seen from above formula, the generation of data packet disorder is relevant with the transmission interval of propagated time delay and transmitting terminal.
The out of order probability of multipath transmitting data in parallel bag is the constraints utilizing packet order transfer demand fulfillment, leaves the time interval △ of transmitting terminal according to packet ti () is different, be divided into following two kinds of situations to discuss:
1.△ T(i)≥d N-d 1
Now, packet leaves the time interval △ of transmitting terminal ti () is greater than propagation delay maximum in all paths poor.Therefore, d (i-1)-△ is had t(i)≤d (i-1)-(d n-d 1)≤d 1≤ d (i), namely no matter packet i selects which bar in N transmission paths to transmit, and packet all must arrive according to the order of sequence, and it is p (△ that out of order probability occurs in transmission of data packets t(i))=0.
2.△ T(i)<d N-d 1
From longer a period of time, in mass data bag process of transmitting, the link selection of adjacent two data bag i-1 and i should meet independent same distribution.Tentation data bag i-1 is assigned to link L lon transmit, then d (i-1)=d l.Because packet leaves the time interval △ of transmitting terminal ti () is always greater than zero, therefore have d (i-1)-△ t(i)=d l-△ t(i) <d l, want to meet constraints, make packet can order transfer, only need to select end-to-end time delay to be more than or equal to d lpath transmit packet i and make d (i)>=d l, namely use link L l, L l+1..., L ntransmit the generation that packet i can avoid data disorder phenomenon.Therefore, out of order probability occurs relevant with the transmit path of packet i-1 and the transmit path of itself in packet i transmitting procedure, computing formula is as follows:
p = p 1 &CenterDot; 0 + ... + p k &CenterDot; ( &Sigma; j = 1 k - 1 p j ) + ... + p N &CenterDot; ( 1 - p N ) = &Sigma; i = 1 N ( R &CenterDot; &Sigma; j = 1 i - 1 p j )
In order to data of description wraps in the distribution characteristics of end-to-end propagation delay time in multipath parallel transmission process, the propagation delay time cumulative distribution function of each packet is defined as D (d):
D ( d ) = &Sigma; d k < d p k = 0 d &le; d 1 p 1 d 1 &le; d &le; d 2 p 1 + p 2 d 2 &le; d &le; d 3 ... ... 1 d > d N
Wherein: the physical significance of D (d) is the probability that the selected end-to-end propagation delay time of transmission path is less than d.
Therefore, under the prerequisite of path transmission packet i-1 determining the d (i-1) that selection propagation delay time is, the probability that the transmission path selected by packet i can meet constraints is: D (d (i-1)-△ t(i)), namely there is out of order probability and can be expressed as further in packet i in transmitting procedure:
P ( &Delta; T ( i ) ) = &Sigma; k = 1 N p k &CenterDot; D ( d ( i - 1 ) &Delta; T ( i ) )
Can learn from above-mentioned analytic process, situation about discussing in 1 is a special case in 2, and therefore in multipath parallel transmission system, data packet disorder probability can be unified to use above-mentioned formula to calculate.
And then increase transmitting terminal propagation delay time and queue time delay to the impact of packet disorder phenomenon, a kind of out of order analytical model of multipath transmitting data in parallel bag based on accumulated probability distribution is proposed.On the basis of above-mentioned analysis, the impact of transmission delay on data packet disorder is incorporated the consideration category of model.Simultaneously, the constraints of packet order transfer and the computing formula of the out of order probability of multipath transmitting data in parallel bag is gone out according to this model inference, under can estimating current network conditions more exactly, in multipath parallel transmission process, there is the probability of data packet disorder.
Transmission delay comprises the propagation delay time and queue time delay that packet experiences at transmitting terminal.In multipath parallel transmission system, after packet is passed to transmitting terminal by application layer, first the transmit queue being assigned to a certain transmission path by service distributing module according to certain Diffluence Algorithm is medium to be sent, then be sent in the middle of network by transmitting terminal after transmission delay S (propagation delay time including the wait time delay in transmit queue and be sent to from transmitting terminal link), receiving terminal is arrived after eventually passing link propagation time delay d, T (i)=R (i)+S (i), packet leaves transmitting terminal time interval △ t(i) be:
T(i)=T(i)-T(i-1)=(R(i)+S(i))-(R(i-1)+S(i-1))
=△ R(i)+S(i)-S(i-1)
Wherein: R (i) is that packet i is passed to the moment of transmitting terminal by application layer, △ ri () is the time interval that packet i-1 and packet i arrive transmitting terminal, S (i) is the transmission delay experienced in packet i transmitting procedure.Below by under the prerequisite of the transmit path selection result of given data bag i-1, respectively the transmission delay of packet i and i-1 is derived.
By the queue of transmitting terminal, transmission delay waits for that time delay and propagation delay time two parts form, about in the packet queuing problem in network, be adopt queueing theory to carry out modeling analysis to the impact of transmission delay.
Suppose that the arrival process of transmitting terminal packet meets Poisson distribution, packet arrival rate is λ, and the size of packet obeys quantum condition entropy, and the transmission rate on kth bar concurrent path is μ k, the capacity of transmitting terminal buffer memory is limited.The allocation of packets of arrival is carried out concurrent transmission according to certain Diffluence Algorithm by the data distribution module of transmitting terminal in the transmit queue of N paths, and the substream of data arrival rate be fitted on kth paths of scoring is λ k, have λ k=p kλ, according to the relative theory of queueing theory, the arrival of a Poisson process is distributed to multiple subprocess independently with certain probability, these subprocess are also Poisson processes, therefore on the concurrent link of this N bar, the arrival process of packet meets Poisson distribution equally.Therefore, the queuing process of packet on the concurrent link of every bar can regard the queuing problem of the M/M/1 queuing system of a limited capacity as.
According to the correlative study of M/M/1 queuing system, the average waiting time of packet in this queuing system is:
t &OverBar; = 1 &mu; - &lambda;
Therefore, the transmission delay that packet i experiences in process of transmitting can be estimated by the average Sojourn times progress row of M/M/1 queuing system, and it is relevant with length service time in link selection result and this link, is designated as:
S ( i ) = &Sigma; k = 1 N p k 1 &mu; k - &lambda; k
Wherein: k represents kth bar link.
And the transmit path selection result of packet i-1 is known, queue length when its transmission delay experienced in process of transmitting and this packet enter transmit queue and the bandwidth of selected transmission link relevant, be designated as:
S ( i - 1 ) = Q ( i - 1 ) + M T U B ( i - 1 )
Queue length on the transmit path of wherein Q (i-1) representative data bag i-1, the bandwidth of transmit path selected by B (i-1) representative data bag i-1, MTU is the length of a MTU.
Therefore, △ ti () can be expressed as further:
&Delta; T ( i ) = &Delta; R ( i ) + S ( i ) - S ( i - 1 ) = &Delta; R ( i ) + &Sigma; k = 1 n p k 1 &mu; k - &lambda; k - Q ( i - 1 ) + M T U B ( i - 1 )
After increasing the consideration for transmission delay impact, under given data bag i-1 sends the prerequisite of situation, there is out of order probability and can be expressed as further in packet i in transmitting procedure:
P ( &Delta; T ( i ) ) = &Sigma; k = 1 n P k &CenterDot; D ( d ( i - 1 ) &Delta; T ( i ) ) &Delta; T ( i ) = &Delta; R ( i ) + &Sigma; k = 1 n p k 1 &mu; k - &lambda; k - Q ( i - 1 ) + M T U B ( i - 1 )
Can be learnt by above formula and analytic process thereof, the time interval that routing strategy and packet leave transmitting terminal is comparatively large on the impact of data disorder phenomenon, and ideally multipath parallel transmission system can completely by out of order impact.Therefore, in order to reduce data disorder phenomenon, the time interval can leaving transmitting terminal from control data bag starts with.Due to traffic data bag, to be delivered to transmitting terminal speed from application layer uncontrollable, can increase certain stand-by period, with control △ before Packet Generation ti () makes it meet the constraints of packet order transfer, avoid out of order by the mode of sacrificing efficiency of transmission.In addition, also can transmit by selecting when packet is shunted suitable transmission path, or indirectly carry out the means such as regulating and controlling to the maximum transmitted delay inequality on each transmission paths at transmitting terminal, make multipath parallel transmission system to meet constraints.
What finally illustrate is, above preferred embodiment is only in order to illustrate technical scheme of the present invention and unrestricted, although by above preferred embodiment to invention has been detailed description, but those skilled in the art are to be understood that, various change can be made to it in the form and details, and not depart from claims of the present invention limited range.

Claims (10)

1. an out of order analytical method for multipath parallel transmission system, is characterized in that: multipath parallel transmission system exists the separate transmission path of N bar between transmitting terminal and receiving terminal, is respectively L 1, L 2..., L k... L n, N>=2; On each paths, propagated time delay meets d end to end 1≤ d 2≤ ... d k...≤d n; The method comprises the following steps:
Step 1) in multipath parallel transmission system, after packet is passed to transmitting terminal by application layer, Packet Generation is gone out by transmitting terminal immediately, packet experience propagated time delay d krear arrival receiving terminal;
Step 2) the out of order of multipath parallel transmission system is analyzed, there is out of order probability in calculated data bag transmitting procedure.
2. the out of order analytical method of a kind of multipath parallel transmission system according to claim 1, is characterized in that: the time interval that packet i-1 and packet i leaves transmitting terminal is △ t(i), △ ti ()=T (i)-T (i-1), T (i) leave the moment of transmitting terminal for packet i-1, T (i-1) leaves the moment of transmitting terminal for packet i-1.
3. the out of order analytical method of a kind of multipath parallel transmission system according to claim 2, is characterized in that: out of order probability occurs in described transmission of data packets is wherein, p kfor packet selects transmission path L kcarry out the probability transmitted, the propagation delay time cumulative distribution function that D (d) is packet,
D ( d ) = &Sigma; d k < d p k = 0 d &le; d 1 p 1 d 1 &le; d &le; d 2 p 1 + p 2 d 2 &le; d &le; d 3 ... ... 1 d > d N .
4. the out of order analytical method of a kind of multipath parallel transmission system according to claim 1, it is characterized in that: described step 1) in multipath parallel transmission system, after packet is passed to transmitting terminal by application layer, the transmit queue being assigned to a certain transmission path according to Diffluence Algorithm by service distributing module is medium to be sent, be sent in the middle of network by transmitting terminal after transmission delay S, experience propagated time delay d krear arrival receiving terminal.
5. the out of order analytical method of a kind of multipath parallel transmission system according to claim 4, is characterized in that: △ ti () leaves the time interval of transmitting terminal, △ for packet i-1 and packet i t(i)=T (i)-T (i-1), T (i) leaves the moment of transmitting terminal for packet i, T (i-1) leaves the moment of transmitting terminal for packet i-1, T (i)=R (i)+S (i), R (i) is the moment of packet i arrival transmitting terminal, the transmission delay that S (i) experiences for packet i.
6. the out of order analytical method of a kind of multipath parallel transmission system according to claim 5, is characterized in that: out of order probability occurs in described transmission of data packets is:
P ( &Delta; T ( i ) ) = &Sigma; k = 1 n P k &CenterDot; D ( d ( i - 1 ) &Delta; T ( i ) ) &Delta; T ( i ) = &Delta; R ( i ) + S ( i ) - S ( i - 1 ) ,
P kfor packet selects transmission path L kthe probability carrying out transmitting is,
The propagation delay time cumulative distribution function that D (d) is packet,
D ( d ) = &Sigma; d k < d p k = 0 d &le; d 1 p 1 d 1 &le; d &le; d 2 p 1 + p 2 d 2 &le; d &le; d 3 ... ... 1 d > d N .
7. the out of order analytical method of a kind of multipath parallel transmission system according to claim 5, it is characterized in that: the arrival process of described transmitting terminal packet meets Poisson distribution, packet arrival rate is λ, the size of packet obeys quantum condition entropy, and the transmission rate on kth bar concurrent path is μ k, the substream of data arrival rate be assigned on kth paths is λ k, &lambda; k = p k &CenterDot; &lambda; , &Sigma; k = 1 N &lambda; k = &lambda; .
8. the out of order analytical method of a kind of multipath parallel transmission system according to claim 7, is characterized in that: the transmission delay that described packet i experiences
The transmission delay that packet i-1 experiences
Wherein, the queue length on the transmit path that Q (i-1) is packet i-1, the bandwidth of B (i-1) transmit path selected by packet i-1, MTU is the length of a MTU.
9. the out of order analytical method of a kind of multipath parallel transmission system according to claim 8, is characterized in that: packet i out of order probability occurs in transmitting procedure and is
P ( &Delta; T ( i ) ) = &Sigma; k = 1 n P k &CenterDot; D ( d ( i - 1 ) &Delta; T ( i ) ) &Delta; T ( i ) = &Delta; R ( i ) + &Sigma; k = 1 n p k 1 &mu; k - &lambda; k - Q ( i - 1 ) + M T U B ( i - 1 ) .
10. the out of order analytical method of a kind of multipath parallel transmission system according to claim 3 or 6, is characterized in that: reduce data disorder phenomenon, demand fulfillment constraints d (i)>=d (i-1)-△ t(i).
CN201510464585.7A 2015-07-31 2015-07-31 A kind of out of order analysis method of multipath parallel transmission system Active CN104954206B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201510464585.7A CN104954206B (en) 2015-07-31 2015-07-31 A kind of out of order analysis method of multipath parallel transmission system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201510464585.7A CN104954206B (en) 2015-07-31 2015-07-31 A kind of out of order analysis method of multipath parallel transmission system

Publications (2)

Publication Number Publication Date
CN104954206A true CN104954206A (en) 2015-09-30
CN104954206B CN104954206B (en) 2018-03-30

Family

ID=54168580

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201510464585.7A Active CN104954206B (en) 2015-07-31 2015-07-31 A kind of out of order analysis method of multipath parallel transmission system

Country Status (1)

Country Link
CN (1) CN104954206B (en)

Cited By (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105682146A (en) * 2016-01-08 2016-06-15 努比亚技术有限公司 Data transmission method and device and user equipment
CN105873096A (en) * 2016-03-24 2016-08-17 重庆邮电大学 Optimization method of efficient throughput capacity of multipath parallel transmission system
CN106211343A (en) * 2016-06-17 2016-12-07 重庆邮电大学 A kind of maximization concurrent transmission power-economizing method based on heterogeneous network
CN107018399A (en) * 2017-03-28 2017-08-04 北京犀牛数字互动科技有限公司 Image data transfer method and device
CN108337160A (en) * 2017-01-20 2018-07-27 华为技术有限公司 Data transmission method, destination node and source node
WO2018157819A1 (en) * 2017-03-02 2018-09-07 华为技术有限公司 Method and apparatus for multiple sub-current network transmission
CN108924063A (en) * 2018-06-29 2018-11-30 浙江大学 A kind of flow allocation method of the minimum end-to-end time delay of delay constraint perception
CN109120545A (en) * 2017-06-23 2019-01-01 华为技术有限公司 Data pack transmission method and device
CN109561468A (en) * 2017-09-26 2019-04-02 ***通信有限公司研究院 A kind of data delivery method, data transmission method for uplink and equipment
CN110113142A (en) * 2019-04-28 2019-08-09 天通畅达(深圳)科技有限公司 The transmission method and system of carrying large-data operation are concurrently bundled based on multiple data channel
CN110650089A (en) * 2019-10-24 2020-01-03 北京大学 Intermediate device supporting multi-path aggregation communication
CN110662256A (en) * 2019-09-18 2020-01-07 北京交通大学 Multi-path cross-protocol transmission data packet scheduling method and system
CN110769465A (en) * 2019-10-24 2020-02-07 Oppo(重庆)智能科技有限公司 Data transmission method and device, electronic equipment and readable storage medium
CN112422421A (en) * 2020-11-23 2021-02-26 北京交通大学 Multi-path data packet transmission method of heterogeneous network
CN113347096A (en) * 2021-08-06 2021-09-03 湖南三湘银行股份有限公司 Transmission processing method of third party payment data
US11171862B2 (en) 2017-03-02 2021-11-09 Huawei Technolgoies Co., Ltd. Multi-subflow network transmission method and apparatus
CN114980199A (en) * 2022-07-13 2022-08-30 鹏城实验室 Data transmission control method for multi-communication link fusion and satellite communication system

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101895466A (en) * 2010-07-02 2010-11-24 北京交通大学 Method for reducing influence of data packet disorder on SCTP multipath transmission
US8369311B1 (en) * 2005-07-01 2013-02-05 Callwave Communications, Llc Methods and systems for providing telephony services to fixed and mobile telephonic devices
CN103401778A (en) * 2013-08-26 2013-11-20 西安电子科技大学 Receiving-end buffer overflow probability guarantee based multi-path transmission packet scheduling method
CN103763217A (en) * 2014-02-07 2014-04-30 清华大学 Packet scheduling method and device for multi-path TCP

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8369311B1 (en) * 2005-07-01 2013-02-05 Callwave Communications, Llc Methods and systems for providing telephony services to fixed and mobile telephonic devices
CN101895466A (en) * 2010-07-02 2010-11-24 北京交通大学 Method for reducing influence of data packet disorder on SCTP multipath transmission
CN103401778A (en) * 2013-08-26 2013-11-20 西安电子科技大学 Receiving-end buffer overflow probability guarantee based multi-path transmission packet scheduling method
CN103763217A (en) * 2014-02-07 2014-04-30 清华大学 Packet scheduling method and device for multi-path TCP

Cited By (23)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105682146A (en) * 2016-01-08 2016-06-15 努比亚技术有限公司 Data transmission method and device and user equipment
CN105873096B (en) * 2016-03-24 2019-05-10 重庆邮电大学 A kind of optimization method of multipath parallel transmission system effective throughput
CN105873096A (en) * 2016-03-24 2016-08-17 重庆邮电大学 Optimization method of efficient throughput capacity of multipath parallel transmission system
CN106211343A (en) * 2016-06-17 2016-12-07 重庆邮电大学 A kind of maximization concurrent transmission power-economizing method based on heterogeneous network
CN108337160A (en) * 2017-01-20 2018-07-27 华为技术有限公司 Data transmission method, destination node and source node
US11171862B2 (en) 2017-03-02 2021-11-09 Huawei Technolgoies Co., Ltd. Multi-subflow network transmission method and apparatus
WO2018157819A1 (en) * 2017-03-02 2018-09-07 华为技术有限公司 Method and apparatus for multiple sub-current network transmission
CN107018399A (en) * 2017-03-28 2017-08-04 北京犀牛数字互动科技有限公司 Image data transfer method and device
CN109120545A (en) * 2017-06-23 2019-01-01 华为技术有限公司 Data pack transmission method and device
CN109120545B (en) * 2017-06-23 2021-05-04 华为技术有限公司 Data packet transmission method and device
CN109561468A (en) * 2017-09-26 2019-04-02 ***通信有限公司研究院 A kind of data delivery method, data transmission method for uplink and equipment
CN108924063A (en) * 2018-06-29 2018-11-30 浙江大学 A kind of flow allocation method of the minimum end-to-end time delay of delay constraint perception
CN108924063B (en) * 2018-06-29 2020-05-15 浙江大学 Flow distribution method for minimizing end-to-end time delay based on time delay constraint perception
CN110113142A (en) * 2019-04-28 2019-08-09 天通畅达(深圳)科技有限公司 The transmission method and system of carrying large-data operation are concurrently bundled based on multiple data channel
CN110662256B (en) * 2019-09-18 2021-03-23 北京交通大学 Multi-path cross-protocol transmission data packet scheduling method and system
CN110662256A (en) * 2019-09-18 2020-01-07 北京交通大学 Multi-path cross-protocol transmission data packet scheduling method and system
CN110769465A (en) * 2019-10-24 2020-02-07 Oppo(重庆)智能科技有限公司 Data transmission method and device, electronic equipment and readable storage medium
CN110650089A (en) * 2019-10-24 2020-01-03 北京大学 Intermediate device supporting multi-path aggregation communication
CN110769465B (en) * 2019-10-24 2023-03-24 Oppo(重庆)智能科技有限公司 Data transmission method and device, electronic equipment and readable storage medium
CN112422421A (en) * 2020-11-23 2021-02-26 北京交通大学 Multi-path data packet transmission method of heterogeneous network
CN113347096A (en) * 2021-08-06 2021-09-03 湖南三湘银行股份有限公司 Transmission processing method of third party payment data
CN114980199A (en) * 2022-07-13 2022-08-30 鹏城实验室 Data transmission control method for multi-communication link fusion and satellite communication system
CN114980199B (en) * 2022-07-13 2022-11-01 鹏城实验室 Data transmission control method for multi-communication link fusion and satellite communication system

Also Published As

Publication number Publication date
CN104954206B (en) 2018-03-30

Similar Documents

Publication Publication Date Title
CN104954206A (en) Disorder analyzing method for multipath parallel transmission system
CN101404622B (en) Wireless internet congestion control method based on multi-path load balancing and controller thereof
CN103841041B (en) A kind of multi-stream service concurrent transmission control method and device
CN105933232A (en) Multi-path transmission control terminal and method for supporting multi-service data transmission
US20180013683A1 (en) Kind of transmission method based on the network learnable power model
CN102149140B (en) End-to-end multi-link parallel transmission network control method
CN100576822C (en) Transfer of data in the cordless communication network and analytic method and device thereof
CN102726009B (en) Data flow exchange control method and relative device
CN101938770A (en) Wireless network maximum retransmission times optimization method based on network channel state
CN105656799A (en) Scheduling method based on business features in SDN network
CN106102093A (en) A kind of multi-path data bag allocation schedule method in wireless self-organization network
CN103748845B (en) Packet sending and receiving method, device and system
WO2020181597A1 (en) Grouping-based random access and data transmission method in large-scale mtc network
CN112350949B (en) Rerouting congestion control method and system based on flow scheduling in software defined network
CN105025524A (en) A multi-path parallel-transmitted data scheduling method and a transmission control protocol
CN109039935B (en) Multipath data scheduling method based on time delay
CN102025628A (en) Distribution method of network flow
CN113438707A (en) Apparatus, method and computer program for routing data in a dual-connection or multi-connection configuration
CN102186211B (en) Load distribution method for end-to-end multilink parallel transmission
CN104935524B (en) The soft load-balancing method that a kind of multipath network is controlled based on chain-circuit time delay
CN102271092B (en) Congestion control system and method based on node load quantization level
CN104219170A (en) Packet scheduling method based on probability retransmission in wireless network
CN106027208A (en) Feedback-based network code TCP (Transmission Control Protocol) decoding method and device
CN105227457B (en) Towards the protocol Data Unit cross-layer optimizing method under DTN dynamic links
CN106105282A (en) Link buffer zone state is utilized to carry out the system and method for traffic engineering

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant