CN105915466B - A kind of method and device of MPTCP Path selection - Google Patents
A kind of method and device of MPTCP Path selection Download PDFInfo
- Publication number
- CN105915466B CN105915466B CN201610235594.3A CN201610235594A CN105915466B CN 105915466 B CN105915466 B CN 105915466B CN 201610235594 A CN201610235594 A CN 201610235594A CN 105915466 B CN105915466 B CN 105915466B
- Authority
- CN
- China
- Prior art keywords
- path
- congestion window
- link
- current
- window value
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 84
- 238000005265 energy consumption Methods 0.000 claims abstract description 70
- 230000008569 process Effects 0.000 claims abstract description 53
- 230000005540 biological transmission Effects 0.000 claims abstract description 32
- 230000007704 transition Effects 0.000 claims description 14
- 230000007246 mechanism Effects 0.000 description 5
- 230000009471 action Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000010187 selection method Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/10—Flow control; Congestion control
- H04L47/12—Avoiding congestion; Recovering from congestion
- H04L47/125—Avoiding congestion; Recovering from congestion by balancing the load, e.g. traffic engineering
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L45/00—Routing or path finding of packets in data switching networks
- H04L45/12—Shortest path evaluation
- H04L45/124—Shortest path evaluation using a combination of metrics
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
- Mobile Radio Communication Systems (AREA)
Abstract
The embodiment of the invention discloses a kind of method and devices of Multipath Transmission control protocol MPTCP Path selection, are applied to mobile terminal, wherein method includes: to identify currently available link when there is currently data to be transmitted；When the quantity of currently available link is not less than two, the congestion window value of every available link is obtained；According to the corresponding relationship of the current link scheme to be selected and destination path that locally save, determine the destination path for transmitting the data to be transmitted, wherein currently record needs that link is selected to carry out under any one combination under every kind of congestion window value in the corresponding relationship of link scheme and destination path to be selected, corresponding destination path, what the Markovian decision process model constructed according to the destination path based on handling capacity and energy consumption was determined.Using the embodiment of the present invention, it can switch in real time different paths according to the truth effectively balancing flow in path and carry out data transmission, improve mobile terminal handling capacity, reduce its energy consumption.
Description
Technical Field
The present invention relates to the field of communications technologies, and in particular, to a method and an apparatus for selecting a multipath transmission Control Protocol (MPTCP) path.
Background
Nowadays, mobile terminals are more and more popular, and a plurality of mobile terminals are provided with a plurality of network interfaces. Many studies have been made on how to perform parallel Multipath Transfer (CMT) of data using a plurality of interfaces. Multipath Transmission control protocol (MPTCP) is a parallel data Transmission protocol of a Transmission layer. MPTCP is an extension of the standard Transmission Control Protocol (TCP), supports simultaneous Transmission of multiple paths, but faces a more complex test than in TCP. On one hand, MPTCP provides higher network throughput, and on the other hand, when multiple paths simultaneously pass through a node in coexistence with TCP, the use of MPTCP may preempt the bandwidth of the standard TCP, which affects the transmission performance of the standard TCP.
MPTCP solves the fairness problem of TCP by a joint congestion control mechanism. The joint congestion control mechanism couples the changes of congestion windows of all sub-flows, so that the throughput of the MPTCP does not exceed the maximum throughput of all sub-flows during data transmission independently, thereby ensuring the fairness of the TCP. However, the joint congestion control mechanism does not consider the influence of packet loss rate when balancing the traffic, and only takes Round Trip Time (RTT) as a criterion for judging the quality of the path when balancing the traffic. Because the RTT is only used as a standard for determining the quality of the path under the mechanism, the accuracy of data transmission is reduced, the throughput is reduced, and the energy consumption of the mobile terminal is increased in a finally determined path in an environment with a high packet loss rate.
Disclosure of Invention
An object of the embodiments of the present invention is to provide a method and an apparatus for selecting a Multipath transmission control Protocol (MPTCP) path, so as to solve the problem that the existing mechanism cannot effectively balance the flow according to the actual condition of the path, and switch different paths in real time for data transmission, so as to improve the throughput of a mobile terminal and reduce the energy consumption of the mobile terminal. The specific technical scheme is as follows:
a method for MPTCP path selection is applied to a mobile terminal, and comprises the following steps:
when data to be transmitted currently exist, identifying a currently available link;
when the number of the current available links is not less than two, acquiring a congestion window value of each available link;
and determining a target path for transmitting the data to be transmitted according to a corresponding relation between a locally stored current link scheme to be selected and the target path, wherein the corresponding relation between the current link scheme to be selected and the target path records the corresponding target path of the link to be selected under any combination mode under each congestion window value, and the target path is determined according to a Markov decision process model established based on throughput and energy consumption.
Preferably, the constructing a markov decision process model based on throughput and energy consumption comprises:
fo＝wtt+we/e
wherein wtWeight for set throughput, weFor a given weight of energy consumption, t is throughput, e is energy consumption, foIs a feedback value.
Preferably, the process of determining the target path according to the markov decision model comprises:
according to the Markov decision process model, when links to be selected are randomly combined under each congestion window value, determining a candidate scheme corresponding to a target path;
determining the throughput and energy consumption of the path according to the current congestion window value of the link to be selected aiming at the path contained in each candidate scheme; determining a corresponding feedback value under the current congestion window value of the path according to the throughput and the energy consumption of the path;
and when the links to be selected are randomly combined under each congestion window value, determining a path scheme set which enables average feedback to be maximum according to the feedback value of the path contained in each candidate scheme and an average feedback formula which is pre-constructed on the basis of the average criterion of the Markov decision process, and determining the path contained in the path scheme set as a target path.
Preferably, the determining the throughput and the energy consumption of the path according to the current congestion window value of the link to be selected includes:
according toDetermining the throughput of said path based onDetermining the energy consumption of said path, wherein t (i, a)m) Selecting a Path scenario a for a Current Congestion Window State imThe latter throughput, e (i, a)m) Selecting a Path scenario a for a Current Congestion Window State imPost-power consumption, λk、βkAre constants determined by link types, S is a set of congestion window values of each link, and S { (cw)1,cw2,...,cwn)|cwk∈[1,Wk],k＝1,2,...,n}，amFor the routing scheme, m is 1,2n-1，RmFor the path scheme amAll paths, R, contained inmSatisfy the requirement of P(j|i,am) To use path scheme amThen, the transition probability of changing from the current congestion window state i to the next congestion window state j, wherein the current congestion window state i is the current congestion window value of the path kThe current congestion window value of the path k is increased by 1 or halved, and when the current congestion window value of the next congestion window state j is increased by 1,when the next congestion window state j is the current congestion window value halved,wherein, the number of successfully transmitted packets for path k, where,wherein,the congestion window value for the current congestion window state for path k,state congestion window value, W, for the next congestion window for path kkFor the maximum window value of path k, RTTkFor the round trip delay of path k, R is all n links, pkIndicating the packet loss rate of path k.
Preferably, the method further comprises:
and when only one available link exists currently, the available link is adopted for data transmission.
The embodiment of the invention also relates to a device for selecting the MPTCP path, which is applied to a mobile terminal and comprises the following components:
the identification module is used for identifying a current available link when the data to be transmitted currently exists;
the acquisition module is used for acquiring the congestion window value of each available link when the number of the current available links is not less than two;
and the determining module is used for determining a target path for transmitting the data to be transmitted according to a locally stored corresponding relation between a current link scheme to be selected and the target path, wherein the corresponding relation between the current link scheme to be selected and the target path records the corresponding target path of the link to be selected in any combination mode under each congestion window value, and the target path is determined according to a Markov decision process model established based on throughput and energy consumption.
Preferably, the markov decision process model comprises:
fo＝wtt+we/e
wherein wtWeight for set throughput, weFor a given weight of energy consumption, t is throughput, e is energy consumption, foIs a feedback value.
Preferably, the apparatus further comprises:
a target path determining module, configured to determine, according to the markov decision process model, a candidate scheme corresponding to a target path when links to be selected are arbitrarily combined at each congestion window value; determining the throughput and energy consumption of the path according to the current congestion window value of the link to be selected aiming at the path contained in each candidate scheme; determining a corresponding feedback value of the path under the current congestion window value according to the throughput and the energy consumption of the path; and when the links to be selected are randomly combined under each congestion window value, determining a path scheme set which enables average feedback to be maximum according to the feedback value of the path contained in each candidate scheme and an average feedback formula which is pre-constructed on the basis of the average criterion of the Markov decision process, and determining the path contained in the path scheme set as a target path.
Preferably, the target path determination module is specifically configured to determine the target path according toDetermining the throughput of said path based onDetermining the energy consumption of said path, wherein t (i, a)m) Selecting a path at a current congestion window state iScheme amThe latter throughput, e (i, a)m) Selecting a Path scenario a for a Current Congestion Window State imPost-power consumption, λk、βkAre constants determined by link types, S is a set of congestion window values of each link, and S { (cw)1,cw2,...,cwn)|cwk∈[1,Wk],k＝1,2,...,n}，amFor the routing scheme, m is 1,2n-1，RmFor the path scheme amAll paths, R, contained inmSatisfy the requirement of P(j|i,am) To use path scheme amThen, the transition probability of changing from the current congestion window state i to the next congestion window state j, wherein the current congestion window state i is the current congestion window value of the path kThe current congestion window value of the path k is increased by 1 or halved, and when the current congestion window value of the next congestion window state j is increased by 1,when the next congestion window state j is the current congestion window value halved,wherein, the number of successfully transmitted packets for path k, where,wherein,the congestion window value for the current congestion window state for path k,state congestion window value, W, for the next congestion window for path kkFor the maximum window value of path k, RTTkFor the round trip delay of path k, R is all n links, pkIndicating the packet loss rate of path k.
Preferably, the apparatus further comprises:
and the transmission module is used for carrying out data transmission by adopting one available link when only one available link exists currently.
The embodiment of the invention provides a method and a device for selecting an MPTCP path, wherein when data to be transmitted currently exist in the method, a currently available link is identified; when the number of the current available links is not less than two, acquiring a congestion window value of each available link; and determining a target path for transmitting the data to be transmitted according to a corresponding relation between a locally stored current link scheme to be selected and the target path, wherein the corresponding relation between the current link scheme to be selected and the target path records the corresponding target path of the link to be selected under any combination mode under each congestion window value, and the target path is determined according to a Markov decision process model established based on throughput and energy consumption. The target path is determined according to the Markov decision process model constructed based on the throughput and the energy consumption, so that the throughput can be effectively improved and the energy consumption can be effectively reduced when the mobile terminal transmits data under the target path. Of course, it is not necessary for any product or method of practicing the invention to achieve all of the above-described advantages at the same time.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a flow chart of a method for MPTCP path selection based on a Markov decision process according to an embodiment of the present invention;
fig. 2 is a flowchart of a method for determining a target path according to a markov decision process model constructed based on throughput and energy consumption in a method for MPTCP path selection based on a markov decision process according to an embodiment of the present invention;
fig. 3 is a flowchart of a method for MPTCP path selection when the currently available links are Wifi and 4G links according to an embodiment of the present invention;
FIG. 4 is a block diagram of an apparatus for MPTCP path selection based on Markov decision process according to an embodiment of the present invention;
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The embodiment of the invention discloses a method for selecting an MPTCP path based on a Markov decision process, which is applied to a mobile terminal and is explained in detail below.
Fig. 1 is a flowchart of a method for MPTCP path selection based on a markov decision process according to an embodiment of the present invention, including the following steps:
step S101: when data to be transmitted currently exists, a currently available link is identified.
The embodiment of the invention is applied to the mobile terminal, and when the mobile terminal determines that the data to be transmitted currently exists, each current available link is identified.
Identifying each currently available link belongs to the prior art, and is not described in detail in the present invention.
Step S102: and when the number of the current available links is not less than two, acquiring the congestion window value of each available link.
And when the current available links are identified, judging the number of the current available links, and when the number of the current available links is not less than two, acquiring the congestion window value of each available link. Obtaining the congestion window value of each available link belongs to the prior art, and is not described in detail in the present invention.
If only one link is identified in the number of available links, the link can be directly used for data transmission.
Step S103: and determining a target path for transmitting the data to be transmitted according to the corresponding relation between the locally stored current link scheme to be selected and the target path.
The corresponding relation between the current link scheme to be selected and the target path records the corresponding target path of the link to be selected under any combination mode under each congestion window value, wherein the target path is determined according to a Markov decision process model constructed based on throughput and energy consumption.
For example, in the embodiment of the present invention, the mobile terminal locally stores the correspondence between the current link scheme to be selected and the target path, specifically, for example, links owned by the mobile terminal include an a link, a b link, and a c link, and the ranges of the congestion window values are cw ∈ [1,2], and then the correspondence between the locally stored current link scheme to be selected and the target path is shown in table 1.
TABLE 1
Table 1 records the corresponding target paths of the links to be selected in any combination manner under each congestion window value. As shown in table 1, when the current link to be selected is a link a and a link b, the table records a corresponding target path of the two links in any combination mode of each congestion window value; when the current links to be selected are the link b and the link c, corresponding target paths are determined in any combination mode of each congestion window value; when the current link to be selected is the a link and the c link, the corresponding target path is in any combination mode of each congestion window value, and when the current link to be selected is the a link, the b link and the c link, the corresponding target path is in any combination mode of each congestion window value.
When the number of the current available links is identified to be not less than two and the congestion window value of each link is obtained, the corresponding target path can be searched according to the table 1. For example, the currently available links are identified as a link and b link, and the obtained congestion window values are cwa1 and cwbAs can be seen from table 1, the corresponding target path is the a link, that is, the mobile terminal may select the a link for data transmission, and the data transmission using the a link may ensure the throughput of the link and reduce the energy consumption of the mobile terminal.
In this step, the target path is determined according to a markov decision process model constructed based on throughput and energy consumption, which is a path included in a path scheme set that maximizes an average feedback determined according to a feedback value of the markov decision process model corresponding to the path at its current congestion window value and an average feedback formula pre-constructed according to an average criterion of the markov decision process.
Specifically, the Markov decision process model constructed based on throughput and energy consumption comprises:
fo＝wtt+we/e
wherein wtWeight of throughput set according to user demand, weWeight of energy consumption set according to user demand, t is throughput, e is energy consumption, foIs a feedback value.
Specifically, a process of determining a target path according to a markov decision process model constructed based on throughput and energy consumption is shown in fig. 2, and fig. 2 is a flowchart of a method for determining a target path according to a markov decision process model constructed based on throughput and energy consumption in a method for MPTCP path selection based on a markov decision process according to an embodiment of the present invention, and includes the following steps:
step S201: and according to the Markov decision process model, determining a candidate scheme corresponding to the target path when the links to be selected are randomly combined under each congestion window value.
Step S202: and aiming at the paths contained in each candidate scheme, determining the throughput and the energy consumption of the paths according to the current congestion window value of the link to be selected.
According toDetermining the throughput of said path based onDetermining the energy consumption of said path, wherein t (i, a)m) For selecting path scheme a at current congestion window value imThe latter throughput, e (i, a)m) For selecting path scheme a at current congestion window value imPost-power consumption, λk、βkAre constants determined by link types, S is a set of congestion window values of each link, and S { (cw)1,cw2,...,cwn)|cwk∈[1,Wk],k＝1,2,...,n}，amFor the path candidate, m is 1,2n-1，RmFor the path scheme amAll paths, R, contained inmSatisfy the requirement ofP(j|i,am) To use path scheme amThen, the transition probability of changing from the current congestion window state i to the next congestion window state j, wherein the current congestion window state i is the current congestion window value of the path kThe current congestion window value of the path k is increased by 1 or halved, and when the current congestion window value of the next congestion window state j is increased by 1,when the next congestion window state j is the current congestion window value halved,wherein, is the number of data packets successfully transmitted using path k, where,wherein,the congestion window value for the current congestion window state for path k,congestion window value, W, for the next congestion window state for path kkFor the maximum window value of path k, RTTkFor the round trip delay of path k, R is all n links, pkIndicating the packet loss rate of path k.
Step S203: and determining a corresponding feedback value of the path under the current congestion window value according to the throughput and the energy consumption of the path.
Specifically, the feedback value is an instantaneous value corresponding to the path scheme corresponding to the path.
Step S204: and when the links to be selected are randomly combined under each congestion window value, determining a path scheme set which enables average feedback to be maximum according to the feedback value of the path contained in each candidate scheme and an average feedback formula which is pre-constructed on the basis of the average criterion of the Markov decision process, and determining the path contained in the path scheme set as a target path.
Specifically, first, an average feedback V (i, pi) using a path scheme set pi when a state i is used as an initial state is established according to an average criterion of a markov decision process, that is, the path scheme set pi (path scheme a) is used when a congestion window value i of a current link to be selected is used as an initial state in the embodimentmSet) of the average feedback V (i, pi).
Wherein s (h) represents the system state at the decision time h, i.e. the current congestion window value of the link to be selected in this embodiment, R (s (h), aπ(s (h)) taking action a when state (s (h) is indicatedπ(s (h)) the obtained immediate feedback, that is, the path scheme a is adopted when the current congestion window value s (h) of the link to be selected in the embodiment is obtainedπ(s (h)) immediate feedback obtained. Pi is a waySet of path schemes pi ═ pi1,π2,...,πlL represents the number of congestion window values in a set S of congestion window values of each link, and l is W1×W2...×Wk，WkIs the maximum window value for link k.
Calculating average feedback V (i, pi)*) Maximum pi*Calculated pi*And giving a corresponding target path of the current link to be selected in any combination mode of each congestion window value. In particular, the average feedback V (i, π) can be calculated using a strategic iteration method*) Maximum pi*The calculation process belongs to the prior art, and the present invention is not described in detail.
Specifically, in practical application, a target path is determined according to a Markov decision model constructed based on throughput and energy consumption, for example, links to be selected are a Wifi link and a 4G link, and the ranges of congestion window values are cw e [1,2]]Then S { (cw)wifi,cw4g)|cwwifi∈[1,2],cw4g∈[1,2]The arbitrary combination method under each congestion window value is as follows: (cw)wifi,cw4g)＝(1,1)、(cwwifi,cw4g)＝(1,2)、(cwwifi,cw4g)＝(2,1)、(cwwifi,cw4g) When the target path is selected, the candidate schemes corresponding to the target path include: path scheme a1For selecting Wifi link and 4G link, path scheme a2Path scheme a for Wifi Link selection only3To select only the 4G link. In the embodiment of the present invention, for each path selection scheme, a plurality of links included in one path selection scheme are referred to as one path, and therefore, for the three candidate schemes described above, three paths are included in total.
And determining the throughput and the energy consumption of each path according to the current congestion window values of the Wifi link and the 4G link.
When the congestion window value is in any combination, e.g. (cw)wifi,cw4g) Take path scheme a ═ 1,11In time, namely, when the Wifi link and the 4G link are used simultaneously for data transmission, the method is based onDetermining the throughput of said path based onDetermining the energy consumption of said path, wherein t (i, a)m)＝t(i,a1) Indicating that the path scheme a is selected when the congestion window state i has a current congestion window value of (1,1)1The latter throughput, e (i, a)m)＝e(i,a1) Indicating that the path scheme a is selected when the congestion window state i has a current congestion window value of (1,1)1The energy consumption of the back is reduced,the sum of the congestion window value of the next congestion window state of the Wifi link and the congestion window value of the next congestion window state of the 4G link is zero value due to the fact that the congestion window value is rounded down, and for convenience of representation, a function h (cw) for reducing the congestion window value by half is defined, the maximum value of the Wifi link round-trip delay and the 4G link round-trip delay is obtained,P(j|i,am) To use path scheme a1Then, the transition probability of the congestion window state with the current congestion window value of (1,1) changing to the next congestion window state j, namely the transition probability of the congestion window state i with the current congestion window value of (1,1) changing to the state that the congestion window values of both the Wifi link and the 4G link are reduced by halfThe current congestion window value is (1,1) and the congestion window state i is changed into the Wifi linkAnd the transition probability of the state of the 4G link in which the congestion window value is halved and the congestion window value is increased by 1The transition probability of the state i of the congestion window with the current congestion window value of (1,1) changing to the state of the congestion window value of the Wifi link plus 1 and the congestion window value of the 4G link minus halfTransition probability of congestion window state i with current congestion window value of (1,1) changing to state of congestion window value of Wifi link and 4G link plus 1The specific expression is as follows:
wherein,pwifipacket loss ratio, p, for Wifi link4gFor the packet loss rate of the 4G link,
then path scenario a is selected at the current congestion window state i1The latter throughput is:
selecting Path scenario a at Current Congestion Window State i1The energy consumption after the process is as follows:
in the same way, when taking the path scheme a2When in use, namely only using the Wifi link for data transmission, according toDetermining the throughput of said path based onDetermining the energy consumption of said path, wherein t (i, a)m)＝t(i,a2) Indicating that the path scheme a is selected when the congestion window state i has a current congestion window value of (1,1)2The latter throughput, e (i, a)m)＝e(i,a2) Indicating that the path scheme a is selected when the congestion window state i has a current congestion window value of (1,1)2The energy consumption of the back is reduced,the congestion window value of the next congestion window state of the Wifi link is zero value due to the rounding-down of the congestion window value, and for the convenience of representation, a function h (cw) for halving the congestion window value is defined, for the round-trip delay of the Wifi link,P(j|i,am) To use path scheme a2Then, the transition probability of the congestion window state i with the current congestion window value of (1,1) changing to the next congestion window state j, namely the transition probability of the congestion window state i with the current congestion window value of (1,1) changing to the state of the congestion window value halving of the Wifi link and the congestion window value of the 4G link not changingTransition probability of state i of congestion window with current congestion window value of (1,1) changing to state of congestion window value of Wifi link plus congestion window values of 1 and 4G links unchangedThe specific expression is as follows:
wherein,pwifithe packet loss rate of the Wifi link is obtained.
Then path scenario a is selected at the current congestion window state i2The latter throughput is:
selecting Path scenario a at Current Congestion Window State i2The energy consumption after the process is as follows:
the same scheme, when taking path scheme a3When only 4G link is used for data transmission, according toDetermining the throughput of said path based onDetermining the energy consumption of said path, wherein t (i, a)m)＝t(i,a3) Indicating that the path scheme a is selected when the congestion window state i has a current congestion window value of (1,1)3The latter throughput, e (i, a)m)＝e(i,a3) Indicating that the path scheme a is selected when the congestion window state i has a current congestion window value of (1,1)3The energy consumption of the back is reduced,the congestion window value of the next congestion window state of the 4G link can be zero value due to the rounding-down of the congestion window value, and for the convenience of representation, a function h (cw) for halving the congestion window value is defined, for the round-trip delay of the 4G link,P(j|i,am) To use path scheme a3Then, the transition probability of changing from the congestion window state i with the current congestion window value of (1,1) to the next congestion window state j, namely changing the congestion window state i with the current congestion window value of (1,1) to the congestion window value of the Wifi link and reducing the congestion window value of the 4G link by halfState transition probability ofTransition probability of state i of congestion window with current congestion window value of (1,1) changing to state of congestion window value of Wifi link unchanged and congestion window value of 4G link added with 1The specific expression is as follows:
in the formulap4gThe packet loss rate of the 4G link is obtained.
Then path scenario a is selected at the current congestion window state i3The latter throughput is:
selecting Path scenario a at Current Congestion Window State i3The energy consumption after the process is as follows:
using fo＝wtt+weAnd e, calculating feedback values of the three paths according to the calculated throughput and energy consumption.
Selecting a route scheme a1In time, even when the Wifi link and the 4G link are used for data transmission, the feedback value is:
in the formula, t (i, a)1) And e (i, a)1) Selecting path schemes a for the calculated current congestion window states i1Later throughput and energy consumption.
Selecting a route scheme a2In time, namely when only using the Wifi link to transmit data, the feedback value is:
in the formula, t (i, a)2) And e (i, a)2) Selecting path schemes a for the calculated current congestion window states i2Later throughput and energy consumption.
Selecting a route scheme a3When the data is transmitted only by using the 4G link, the feedback value is:
in the formula, t (i, a)3) And e (i, a)3) Selecting path schemes a for the calculated current congestion window states i3Later throughput and energy consumption.
Solving the average feedback V (i, pi) by using a strategy iteration method according to an average feedback formula*) Maximum pi*That is, the calculation results of the target paths corresponding to Wifi and 4G links under the four congestion window values can be shown in table 2:
TABLE 2
Specifically, the detailed description will be given with an embodiment in which the mobile terminal has a Wifi link and a 4G link.
Fig. 3 is a flowchart of a method for MPTCP path selection when the currently available links are Wifi and 4G links according to an embodiment of the present invention, including the following steps:
step S301: when data to be transmitted currently exists, a currently available link is identified.
Step S302: and when the current available link is identified, judging the number of the current available links, and when both the links can be used, acquiring the congestion window value of each available link.
And if the identification result is only the Wifi link available, the Wifi link is used for transmitting the data packet, and if the identification result is only the 4G link available, the 4G link is used for transmitting the data packet.
Step S303: determining a target path for transmitting the data to be transmitted according to a corresponding relation between a locally stored current link scheme to be selected and the target path;
for example, the obtained current congestion window values of the Wifi link and the 4G link are (cw)wifi,cw4g) If the target path is a Wifi link, the table look-up 2 determines that the target path is a Wifi link, and then the Wifi link is selected for data transmission.
Therefore, by using the embodiment of the invention, the determined target path is determined according to the Markov decision process model constructed based on the throughput and the energy consumption, so that the throughput can be effectively improved and the energy consumption can be effectively reduced when the mobile terminal transmits data under the target path.
Fig. 4 is a block diagram of an apparatus for MPTCP path selection based on a markov decision process according to an embodiment of the present invention, where the apparatus is applied to a mobile terminal, and the apparatus includes:
the recognition module 401: for identifying a currently available link when there is currently data to be transmitted.
The embodiment of the invention is applied to the mobile terminal, and when the mobile terminal determines that the data to be transmitted currently exists, each current available link is identified.
Identifying each currently available link belongs to the prior art, and is not described in detail in the present invention.
The acquisition module 402: and the congestion window value of each available link is obtained when the number of the current available links is not less than two.
And when the current available links are identified, judging the number of the current available links, and when the number of the current available links is not less than two, acquiring the congestion window value of each available link. Obtaining the congestion window value of each available link belongs to the prior art, and is not described in detail in the present invention.
If only one link is identified in the number of available links, the link can be directly used for data transmission.
The determination module 403: and the target path used for transmitting the data to be transmitted is determined according to the corresponding relation between the locally stored current link scheme to be selected and the target path.
The corresponding relation between the current link scheme to be selected and the target path records the corresponding target path of the link to be selected under any combination mode under each congestion window value, wherein the target path is determined according to a Markov decision process model constructed based on throughput and energy consumption.
Wherein the Markov decision process model comprises:
fo＝wtt+we/e
wherein wtWeight for set throughput, weFor a given weight of energy consumption, t is throughput, e is energy consumption, foIs a feedback value.
Specifically, the apparatus further comprises:
a target path determining module (not shown in fig. 4) configured to determine, according to the markov decision process model, a candidate scheme corresponding to a target path when a link to be selected is arbitrarily combined at each congestion window value; determining the throughput and energy consumption of the path according to the current congestion window value of the link to be selected aiming at the path contained in each candidate scheme; determining a corresponding feedback value of the path under the current congestion window value according to the throughput and the energy consumption of the path; and when the links to be selected are randomly combined under each congestion window value, determining a path scheme set which enables average feedback to be maximum according to the feedback value of the path contained in each candidate scheme and an average feedback formula which is pre-constructed on the basis of the average criterion of the Markov decision process, and determining the path contained in the path scheme set as a target path.
In particular, in a practical use, the target path determination module is specifically configured to determine the target path based onDetermining the throughput of said path based onDetermining the energy consumption of said path, wherein t (i, a)m) Selecting a Path scenario a for a Current Congestion Window State imThe latter throughput, e (i, a)m) Selecting a Path scenario a for a Current Congestion Window State imPost-power consumption, λk、βkAre constants determined by link types, S is a set of congestion window values of each link, and S { (cw)1,cw2,...,cwn)|cwk∈[1,Wk],k＝1,2,...,n}，amFor the routing scheme, m is 1,2n-1，RmFor the path scheme amZhongzhu bagAll paths covered, RmSatisfy the requirement of P(j|i,am) To use path scheme amThen, the transition probability of changing from the current congestion window state i to the next congestion window state j, wherein the current congestion window state i is the current congestion window value of the path kThe current congestion window value of the path k is increased by 1 or halved, and when the current congestion window value of the next congestion window state j is increased by 1,when the next congestion window state j is the current congestion window value halved,wherein, the number of successfully transmitted packets for path k, where,wherein,the congestion window value for the current congestion window state for path k,state congestion window value, W, for the next congestion window for path kkIs the maximum window value of path k, RTTkFor the round trip delay of path k, R is all n links, pkIndicating the packet loss rate of path k.
Specifically, the apparatus further comprises:
and a transmission module (not shown in fig. 4) configured to use one available link for data transmission when only one available link currently exists.
For the device embodiment, since it is basically similar to the method embodiment, the description is simple, and for the relevant points, refer to the partial description of the method embodiment.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
Those skilled in the art will appreciate that all or part of the steps in the above method embodiments may be implemented by a program to instruct relevant hardware to perform the steps, and the program may be stored in a computer-readable storage medium, which is referred to herein as a storage medium, such as: ROM/RAM, magnetic disk, optical disk, etc.
The above description is only for the preferred embodiment of the present invention, and is not intended to limit the scope of the present invention. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention shall fall within the protection scope of the present invention.
Claims (4)
1. A method for selecting a multipath transmission control protocol (MPTCP) path is applied to a mobile terminal and is characterized by comprising the following steps:
when data to be transmitted currently exist, identifying a currently available link;
when the number of the current available links is not less than two, acquiring a congestion window value of each available link;
determining a target path for transmitting the data to be transmitted according to a corresponding relation between a locally stored current link scheme to be selected and the target path, wherein the corresponding relation between the current link scheme to be selected and the target path records the corresponding target path of the link to be selected under any combination mode under each congestion window value, and the target path is determined according to a Markov decision process model established based on throughput and energy consumption;
the constructing a Markov decision process model based on throughput and energy consumption comprises:
fo＝wtt+we/e
wherein wtWeight for set throughput, weFor a given weight of energy consumption, t is throughput, e is energy consumption, foIs a feedback value;
the process of determining the target path from the Markov decision process model comprises:
according to the Markov decision process model, when links to be selected are randomly combined under each congestion window value, determining a candidate scheme corresponding to a target path;
determining the throughput and energy consumption of the path according to the current congestion window value of the link to be selected aiming at the path contained in each candidate scheme; determining a corresponding feedback value of the path under the current congestion window value of the path according to the throughput and the energy consumption of the path;
when the links to be selected are randomly combined under each congestion window value, determining a path scheme set which enables average feedback to be maximum according to the feedback value of the path contained in each candidate scheme and an average feedback formula which is pre-constructed on the basis of the average criterion of the Markov decision process, and determining the path contained in the path scheme set as a target path;
determining the throughput and the energy consumption of the path according to the current congestion window value of the link to be selected comprises:
according toDetermining the throughput of said path based onDetermining the energy consumption of said path, wherein t (i, a)m) Selecting a Path scenario a for a Current Congestion Window State imThe latter throughput, e (i, a)m) Selecting a Path scenario a for a Current Congestion Window State imPost-power consumption, λk、βkAre constants determined by link types, S is a set of congestion window values of each link, and S { (cw)1，cw2，...，cwn)|cwk∈[1，Wk]，k＝1，2，...，n}，amFor the routing scheme, m is 1,2n-1，RmFor the path scheme amAll paths, R, contained inmSatisfy the requirement of P(j|i,am) To use path scheme amThen, the transition probability of changing from the current congestion window state i to the next congestion window state j, wherein the current congestion window state i is the current congestion window value of the path kThe current congestion window value of the path k is increased by 1 or halved, and when the current congestion window value of the next congestion window state j is increased by 1,when the next congestion window state j is the current congestion window value halved,wherein, the number of successfully transmitted packets for path k, where,wherein,the congestion window value for the current congestion window state for path k,state congestion window value, W, for the next congestion window for path kkFor the maximum window value of path k, RTTkFor the round trip delay of path k, R is all n links, pkIndicating the packet loss rate of path k.
2. The method of claim 1, further comprising:
and when only one available link exists currently, the available link is adopted for data transmission.
3. A device for selecting a multipath transmission control protocol (MPTCP) path is applied to a mobile terminal, and is characterized by comprising:
the identification module is used for identifying a current available link when the data to be transmitted currently exists;
the acquisition module is used for acquiring the congestion window value of each available link when the number of the current available links is not less than two;
the determining module is used for determining a target path for transmitting the data to be transmitted according to a locally stored corresponding relationship between a current link scheme to be selected and the target path, wherein the corresponding relationship between the current link scheme to be selected and the target path records the corresponding target path of the link to be selected in any combination mode under each congestion window value, and the target path is determined according to a Markov decision process model established based on throughput and energy consumption;
the Markov decision process model includes:
fo＝wtt+we/e
wherein wtWeight for set throughput, weFor a given weight of energy consumption, t is throughput, e is energy consumption, foIs a feedback value;
the device further comprises:
a target path determining module, configured to determine, according to the markov decision process model, a candidate scheme corresponding to a target path when links to be selected are arbitrarily combined at each congestion window value; determining the throughput and energy consumption of the path according to the current congestion window value of the link to be selected aiming at the path contained in each candidate scheme; determining a corresponding feedback value of the path under the current congestion window value of the path according to the throughput and the energy consumption of the path; when the links to be selected are randomly combined under each congestion window value, determining a path scheme set which enables average feedback to be maximum according to the feedback value of the path contained in each candidate scheme and an average feedback formula which is pre-constructed on the basis of the average criterion of the Markov decision process, and determining the path contained in the path scheme set as a target path;
the target path determination module is specifically configured to determine the target path based onDetermining the throughput of said path based onDetermining the energy consumption of said path, wherein t (i, a)m) Selecting a Path scenario a for a Current Congestion Window State imThe latter throughput, e (i, a)m) Selecting a Path scenario a for a Current Congestion Window State imPost-power consumption, λk、βkAre constants determined by link types, S is a set of congestion window values of each link, and S { (cw)1，cw2，...，cwn)|cwk∈[1，Wk]，k＝1，2，...，n}，amFor the routing scheme, m is 1,2n-1，RmFor the path scheme amAll paths, R, contained inmSatisfy the requirement of P(j|i,am) To use path scheme amThen, the transition probability of changing from the current congestion window state i to the next congestion window state j, wherein the current congestion window state i is the current congestion window value of the path kThe current congestion window value of the path k is increased by 1 or halved, and when the current congestion window value of the next congestion window state j is increased by 1,when the next congestion window state j is the current congestion window value halved,wherein, the number of successfully transmitted packets for path k, where,wherein,the congestion window value for the current congestion window state for path k,state congestion window value, W, for the next congestion window for path kkFor the maximum window value of path k, RTTkFor the round trip delay of path k, R is all n links, pkIndicating the packet loss rate of path k.
4. The apparatus of claim 3, further comprising:
and the transmission module is used for carrying out data transmission by adopting one available link when only one available link exists currently.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610235594.3A CN105915466B (en) | 2016-04-15 | 2016-04-15 | A kind of method and device of MPTCP Path selection |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610235594.3A CN105915466B (en) | 2016-04-15 | 2016-04-15 | A kind of method and device of MPTCP Path selection |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105915466A CN105915466A (en) | 2016-08-31 |
| CN105915466B true CN105915466B (en) | 2019-03-19 |
Family
ID=56747103
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610235594.3A Active CN105915466B (en) | 2016-04-15 | 2016-04-15 | A kind of method and device of MPTCP Path selection |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105915466B (en) |
Families Citing this family (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108023817B (en) * | 2016-11-04 | 2021-03-02 | 华为技术有限公司 | Method and device for determining data transmission path |
| CN106878192B (en) * | 2016-12-26 | 2020-02-14 | 北京邮电大学 | Data scheduling method of self-adaptive MPTCP |
| CN106657362B (en) * | 2016-12-30 | 2019-10-29 | 清华大学 | A kind of terminal content dispatching method of New Virtual multipath TCP |
| CN108667880A (en) * | 2017-03-31 | 2018-10-16 | 华为技术有限公司 | A kind of SiteServer LBS, method and device |
| CN108737264B (en) * | 2017-04-19 | 2020-07-24 | 腾讯科技（深圳）有限公司 | Data transmission method and device |
| CN107171842B (en) * | 2017-05-22 | 2020-01-03 | 南京大学 | Multipath transmission protocol congestion control method based on reinforcement learning |
| CN107797764B (en) * | 2017-08-31 | 2021-05-25 | 郑州云海信息技术有限公司 | Method and device for determining path |
| CN107733811B (en) * | 2017-09-18 | 2020-02-21 | 中国电子科技集团公司第五十四研究所 | Method for dynamically adjusting coding strength |
| CN109729007B (en) | 2017-10-28 | 2022-07-22 | 华为技术有限公司 | Method, device and equipment for transmitting data |
| CN115426693A (en) | 2018-08-02 | 2022-12-02 | 华为技术有限公司 | Network switching method, electronic equipment and chip subsystem |
| CN110768900B (en) * | 2019-09-18 | 2021-06-22 | 华为技术有限公司 | Data transmission method and electronic equipment |
| CN111277502B (en) * | 2020-01-20 | 2022-05-17 | 北京红云融通技术有限公司 | Method for transmitting data by multi-link aggregation and transmitting equipment |
| CN111416770B (en) * | 2020-02-21 | 2021-05-14 | 北京交通大学 | Self-adaptive scheduling system and method for cross-protocol convergence transmission |
| CN114071634B (en) * | 2020-07-30 | 2023-11-21 | 中国电信股份有限公司 | Multipath transmission control method and device and mobile terminal |
| CN113037648B (en) * | 2021-03-10 | 2022-07-12 | 首都师范大学 | Data transmission method and device |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103299710A (en) * | 2011-01-14 | 2013-09-11 | 诺基亚公司 | Method for multipath scheduling based on a lookup table |
| CN103929369A (en) * | 2014-05-07 | 2014-07-16 | 北京邮电大学 | Multichannel transmission control mechanism based on TCP friendliness |
-
2016
- 2016-04-15 CN CN201610235594.3A patent/CN105915466B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103299710A (en) * | 2011-01-14 | 2013-09-11 | 诺基亚公司 | Method for multipath scheduling based on a lookup table |
| CN103929369A (en) * | 2014-05-07 | 2014-07-16 | 北京邮电大学 | Multichannel transmission control mechanism based on TCP friendliness |
Non-Patent Citations (1)
| Title |
|---|
| 面向多路径并行数据传输的动态路径选择方法;金小敏 等;《北京邮电大学学报》;20151015;第38卷(第5期);第99-103页 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105915466A (en) | 2016-08-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105915466B (en) | A kind of method and device of MPTCP Path selection | |
| EP2689602B1 (en) | Buffer sizing for multi-hop network | |
| US8279753B2 (en) | Efficient determination of fast routes when voluminous data is to be sent from a single node to many destination nodes via other intermediate nodes | |
| US20150215922A1 (en) | Method and apparatus for traffic distribution control in heterogeneous wireless networks | |
| CN108696449B (en) | Data scheduling method and device | |
| CN111835564B (en) | Self-adaptive recovery method and system for power Internet of things communication link fault | |
| CN101895466A (en) | Method for reducing influence of data packet disorder on SCTP multipath transmission | |
| CN107409099A (en) | The method, apparatus and machine readable media of traffic engineering in communication network with service quality stream and stream of doing one's best | |
| US20100091660A1 (en) | Communication Control Apparatus, Wireless Communication Apparatus, Communication Control Method and Wireless Communication Method | |
| CN103841040A (en) | Network system and load balance method | |
| CN103312607B (en) | A kind of selection of transmission paths method and device | |
| CN108684046A (en) | A kind of access net service function chain dispositions method based on incidental learning | |
| CN107154897B (en) | Isomery stream partition method based on packet scattering in DCN | |
| CN106878192A (en) | A kind of data dispatching method of adaptive M PTCP | |
| CN113329413B (en) | Multipath transmission data pre-scheduling method and system for high-speed rail scene | |
| CN104092625A (en) | Adaptive request batch scheduling method used in DCN | |
| Tardioli et al. | Real time communications over 802.11: RT-WMP | |
| CN105262534B (en) | A kind of method for routing and device suitable for satellite communication network | |
| CN105393597B (en) | Method for controlling network congestion and controller | |
| US20220224604A1 (en) | Mptcp incast performance evaluation model based on a queuing network | |
| Mallapur et al. | Multipath load balancing technique for congestion control in mobile ad hoc networks | |
| CN114448899A (en) | Method for balancing network load of data center | |
| CN107231318B (en) | Method, equipment and device for dynamic switching control based on real-time network performance | |
| CN111901237B (en) | Source routing method and system, related device and computer readable storage medium | |
| US10412000B2 (en) | Wireless mesh communications networks |
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 |