CN113242182B - QoS-sensitive route distribution method in mobile self-organizing network based on SDN - Google Patents
QoS-sensitive route distribution method in mobile self-organizing network based on SDN Download PDFInfo
- Publication number
- CN113242182B CN113242182B CN202110499275.4A CN202110499275A CN113242182B CN 113242182 B CN113242182 B CN 113242182B CN 202110499275 A CN202110499275 A CN 202110499275A CN 113242182 B CN113242182 B CN 113242182B
- Authority
- CN
- China
- Prior art keywords
- route
- link
- link quality
- flow
- network
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L45/00—Routing or path finding of packets in data switching networks
- H04L45/302—Route determination based on requested QoS
-
- 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
- H04L45/70—Routing based on monitoring results
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W40/00—Communication routing or communication path finding
- H04W40/02—Communication route or path selection, e.g. power-based or shortest path routing
- H04W40/12—Communication route or path selection, e.g. power-based or shortest path routing based on transmission quality or channel quality
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks
Abstract
The invention relates to a QoS-sensitive route allocation method in a mobile self-organizing network based on an SDN, and provides a route allocation problem by taking the QoS performance of a maximized user as an objective function aiming at the characteristic of centralized control route of the SDN. In order to effectively solve this problem, the following procedure is followed. Firstly, link quality prediction is carried out, and the link quality of each link at the next moment is predicted by using a time series analysis method. And then forming a routing problem, and forming a routing optimization problem which is easy to solve by the SDN controller according to the obtained link quality of each link at the next moment. And finally, solving the optimal QoS route, and solving the optimal QoS route for each flow in the network in real time by using a differential search algorithm, thereby greatly improving the network throughput and reducing the system delay and the packet loss rate.
Description
Technical Field
The invention relates to a QoS-sensitive route allocation method in a Mobile Ad Hoc Network (MANET) based on a Software Defined Network (SDN), belonging to the technical field of SDN-based novel heterogeneous self-organizing communication Network routes.
Background
A mobile ad hoc network is a collection of wireless nodes that communicate using wireless devices without any pre-existing infrastructure. Due to the mobility of the nodes, the network topology changes dramatically, which results in the degradation of network performance, such as the degradation of network throughput, and the increase of network delay and packet loss rate. Furthermore, its distributed architecture prevents efficient cooperation between the mobile ad hoc network nodes, which makes it difficult for each flow in the network to find an optimal routing path. A software defined network is a centrally controlled network structure that can provide control over an underlying network. Therefore, the SDN controller can reasonably plan the path of each flow in the network according to the collected link quality information, and the overall performance of the network is improved. In addition, the SDN may also provide flexibility and scalability to the network, making the network programmable and easy to manage.
In recent years, the application of the SDN in MANET is increasing, and due to the mobility characteristics of MANET nodes, the network topology structure changes frequently, so that link quality information acquired by the SDN is inaccurate, and the final routing performance is affected. If the link quality information can be predicted in advance, it is an urgent problem to allocate an optimal QoS route to a flow in a network according to the link quality information obtained at the next moment.
Disclosure of Invention
The invention aims to solve the technical problem of providing a QoS-sensitive route allocation method in a mobile self-organizing network based on an SDN, and provides a route allocation problem by taking the QoS performance of a maximized user as an objective function according to the characteristic of centralized control routing of the SDN.
The invention adopts the following technical scheme for solving the technical problems: the invention designs a QoS-sensitive route distribution method in a mobile self-organizing network based on an SDN, which is used for solving an optimal QoS route for each flow in the network in real time, thereby greatly improving the network throughput and reducing the system delay and the packet loss rate; the method comprises the following steps:
the SDN controller collects link quality information through a southbound interface, periodically samples the quality information of each link and records the sampled link quality information w t ，r t ，d t Set of samples W forming a link quality prediction t ，R t ，D t . Wherein the content of the first and second substances,w t representing the throughput at time t, r t Representing the packet loss rate at time t, d t Representing the time delay at time t, W t ＝[w t ,w t-1 ,…,w t-Φ ]，R t ＝[r t ,r t-1 ,…,r t-Φ ]，D t ＝[d t ,d t-1 ,…,d t-Φ ]Then entering step B;
and B, the SDN controller adopts a normalized least square method in time sequence analysis to the link quality w at the t +1 moment according to the input link quality sample set t+1 ，r t+1 ，d t+1 The prediction is carried out by the following specific method:
the input to the system is a set of link quality information W t ，R t ，D t In the sum filter coefficient vector a t ，b t ，c t After multiplication, link quality information predicted at time t +1 is obtained, namely:
w t+1 ＝a t W t
r t+1 ＝b t R t
d t+1 ＝c t D t
in order to predict link quality information more accurately, the design criteria of the filter should behave as the following optimization problem.
min||a t+1 -a t || 2 +λ(w t+1 -a t+1 W t )
||b t+1 -b t || 2 +λ(r t+1 -b t+1 R t )
||c t+1 -c t || 2 +λ(d t+1 -c t+1 D t )
This filter can be obtained by solving the above optimization problem, with the following results:
where μ is a constant representing the step size, and the value of μ is between 0 and 2. Due to the fact that the time complexity of the algorithm is small, the SDN controller can predict link quality information quickly. In addition, due to the adaptability of the algorithm, the high prediction precision can be always kept. The SDN controller obtains link quality information at the time of t +1 and then enters a step C;
SDN controller calculates each flow f by Dijkstra algorithm i Feasible path set P i Taking the link quality as a weight, and then entering the step D;
the SDN controller formulates a routing optimization problem according to the collected link quality information, and the optimization aim is to enable the flow f in the network i The method maximizes the QoS performance, ensures to find a routing scheme with the best QoS performance for the network, and comprises the following specific steps:
modeling an SDN-based MANET as a connected undirected graph G (V, E), where V is a set of nodes, E is a set of links, and E is a link uv And (u, v) E represents a link between the node u and the node v, and all the nodes are connected through the wireless link and have the same communication priority. The SDN controller collects link quality information through a southbound interface, and the link quality information of the link e considered in the method is throughputw e Packet loss rate r e And link delay d e The total throughput W, the total packet loss rate R, and the total delay D of all flows in the network can be expressed by the following equations:
wherein x is ip Representing flow f i Whether to select feasible path P ∈ P i ，P i Is a flow f i F is the flow F i The set of (a) and (b),
indicating whether link e is present in flow f i In the feasible path p. The route optimization problem may be represented by the problem P1:
P1:max W
max D -1
max R -1
x ip ∈{0,1}
the optimization goal of P1 is to make flow f in the network i Wherein the constraint means that it must be taken from flow f i Feasible path set P i To select a path. The optimization objective may ensure that a routing scheme with the best QoS performance is found for the network. By solving the above problem, an optimal route allocation strategy can be obtained, the problem is a 0-1 integer programming problem, which is an NP-hard problem, and further transformation of P1 is required for simple solution:
P2:maxαW+εD -1 +γR -1
x ip ∈{0,1}
the problem P2 converts the multi-objective optimization into a single-objective optimization, where α, ε, and γ are weighting coefficients, and α + ε + γ is 1. This conversion has two advantages, one is that the SDN controller can determine the optimal priority of the three QoS performances according to the type of the flow. For example, for a video stream with a high delay requirement, α may be set larger. Secondly, the P2 becomes a typical 0-1 knapsack problem through the transformation, the problem can be solved by using a differential search algorithm, and then the step E is carried out;
step e.sdn controller calculates per flow f by solving a route optimization problem using a differential search algorithm within a route calculation module i The method comprises the following specific processes of sending routing information to a mobile ad hoc node in a network through a flow table mode:
SDN controller calculates per-flow f by solving a route optimization problem within a route calculation module i To solve problem P2, P2 is transformed into the form:
λ 1 ＝1+αW+εD -1 +γR -1
wherein λ 1 Is a constraint factor, X m Is a solution of problem 2, X m ＝[x 1 ,x 2 ,…,x ip ,…,x N ]And N represents the dimension of the solution. N ═ P i If the assigned route satisfies the constraint in problem P2, then:
F(X m )＝max{αW+εD -1 +γR -1 }≥0
if the constraint in question P2 is not satisfied:
at this time F (X) m ) ≦ 1, further converting problem P2 to an unconstrained problem, SDN controller need only be based on F (X) m ) The obtained solution is judged to be a feasible solution, then a route calculation module of the SDN controller can obtain the optimal QoS route by using a differential search algorithm, the differential search is a group intelligent optimization algorithm with excellent performance which is recently proposed, and the idea of the algorithm comes from the random walk behavior of organisms in the migration process. As food and other resources in nature vary with seasons, many organisms are gathered together to form a super organism and move toward more resource-rich areas by using collective wisdom. Based on the principle, the positions of organisms in the super organisms can be regarded as a set of solutions, the position of each organism in the organisms is a single solution, and the process that the super organisms continuously migrate to search resources is a process that the solutions are continuously updated until the organisms find the most suitable habitat, namely the optimal solution of the route. The process of route calculation module within SDN controller using differential search is as follows: first, the SDN controller randomly generates an initial solution matrix M × N, where M represents the number of organisms in the super-organism (i.e., the number of initial solutions). The initial solution represents the initial position of the living being in the super organism, and is set as follows:
at this time X m Should satisfy F (X) m ) When the value is more than or equal to 0, in the process of generating the solution, if X is equal to m If the condition is not met, the solution is discarded, and a new solution meeting the condition is randomly generated. In thatDuring the process of biological migration, the super-organic body will check whether some randomly selected locations are suitable temporary locations (i.e., determine whether the obtained temporary solution is closer to the optimal solution). If a site is suitable for temporary residence during the migration process, it is found that the creatures of the site are immediately located at the site and then continue to migrate from the site, which is represented by:
S m ＝X m +μ(X d -X m )
X d ＝argmax{F(X m )}，m＝1,2,…,M
wherein S m For the stopover site (i.e. temporary solution) during the migration of the super organism, S m ＝(s m1 ,…,s mn ,…,s mN )，m＝1,2,…,M，S m F (S) should also be satisfied m ) And the SDN controller regenerates the condition that F (S) is met if the condition is not met m ) Random solution S of more than or equal to 0 m ，X d Is X m Zhongshi F (X) m ) The largest solution, μ, is a scale factor, the value of which is determined by the gamma distribution with parameters:
μ＝1/GamRnd(Φ 1 ,Φ 2 )
despite the fact that the stopover site S is obtained m Due to s, however mn Is a real number and therefore needs to be converted into a binary number:
wherein f(s) mn ) Is a sigmoid function which maps real numbers into the interval 0-1, r, using appropriate values of τ n Is a random number between 0 and 1 which determines s mn If f(s) mn )<r n Then variable s mn Is 0; otherwise it is 1. At each calculation of S m Then, F (X) is added m ) And F (S) m ) A comparison is made. If F (S) m )≥F(X m ) Then solve X m Is updated to S m Otherwise, the solution X is retained m Through g max After the second iteration, the algorithm ends, resulting in F (X) m ) Maximum X m I.e. the optimal route allocation for the network.
Since the constraint objectives are relaxed in the process of converting P1 to P2, some flows may not be assigned routing paths after the end of the differential search algorithm, for which the weights of the links are set according to the types of the remaining flows, such as video flows with high delay requirements, the link delay may be taken as a weight, and the SDN controller calculates each remaining flow f using Dijkstra's algorithm i The best path of (a). After the optimal QoS route of each flow in the network is confirmed, the SDN controller issues the flow to the nodes of the mobile ad hoc network in the form of the flow table.
Has the advantages that: compared with the prior art, the invention has the following remarkable advantages: the invention provides a route distribution problem by taking the maximum QoS performance of a user as an objective function based on the characteristic of the centralized control route of the SDN, and adopts a time sequence analysis method to quickly and accurately predict the link quality of each link at the next moment in real time, thereby being capable of coping with the complex and changeable network topology in the mobile self-organizing network. According to the obtained link quality of each link at the next moment, the SDN controller converts a complex route optimization problem into an easily solved route optimization problem, and solves the optimal QoS route for each flow in the network by using a differential search algorithm, so that the network throughput is greatly improved, and the system delay and the packet loss rate are reduced.
Drawings
Fig. 1 is a schematic view of a scenario of a QoS-sensitive route allocation method of a mobile ad hoc network based on an SDN network architecture according to the present invention;
fig. 2 is a link quality prediction model in a QoS-sensitive route allocation method of a mobile ad hoc network designed based on an SDN network architecture according to the present invention;
fig. 3 is a flow chart of a QoS-sensitive route allocation method of a mobile ad hoc network designed based on an SDN network architecture according to the present invention.
Detailed Description
The following description will explain embodiments of the present invention in further detail with reference to the accompanying drawings.
As shown in fig. 1, the present invention contemplates a new SDN-based heterogeneous ad-hoc communication network scenario comprising an SDN controller, MANET nodes and southbound interfaces. Modeling SDN-based MANET as a connected undirected graph G (V, E), where V is a set of nodes, E is a set of links, and link E is a link set uv And (u, v) ∈ E denotes a link between the node u and the node v. All nodes are connected by wireless links, with the same communication priority. The control function of the SDN controller to the network is mainly realized through a southbound interface protocol, and the control function comprises the functions of link discovery, topology management, strategy formulation, routing table distribution and the like. The controller receives network information from the southbound interface and is responsible for controlling various forwarding rules and route generation. When a link in the network is broken, the controller will re-establish the topology and routes, update the topology and routing tables, or add a new node to replace the failed node in the route. The routing algorithm herein is implemented in a route calculation module of the SDN controller.
As shown in fig. 2, a link quality prediction model in a QoS route allocation method for a mobile ad hoc network based on an SDN network architecture is described, in the design of the present invention, an SDN controller collects link quality information through a southbound interface, the SDN controller periodically samples the quality information of each link, and records sampled link quality information w t ，r t ，d t Set of samples W forming a link quality prediction t ，R t ，D t . Wherein, w t Represents the throughput at time t, r t Representing the packet loss rate at time t, d t Representing the time delay at time t, W t ＝[w t ,w t-1 ,…,w t-Φ ]，R t ＝[r t ,r t-1 ,…,r t-Φ ]，D t ＝[d t ,d t-1 ,…,d t-Φ ]Then, the SDN controller adopts a normalized least square method in time series analysis according to the input link quality sample set to determine the link quality w at the t +1 moment t+1 ，r t+1 ，d t+1 Making predictions, i.e. sets W of link quality information t ，R t ，D t And the filter coefficient vector a t ，b t ，c t Multiplying to obtain the predicted link quality information at time t + 1:
w t+1 ＝a t W t
r t+1 ＝b t R t
d t+1 ＝c t D t
in order to predict link quality information more accurately, the design criteria of the filter should behave as the following optimization problem.
min||a t+1 -a t || 2 +λ(w t+1 -a t+1 W t )
||b t+1 -b t || 2 +λ(r t+1 -b t+1 R t )
||c t+1 -c t || 2 +λ(d t+1 -c t+1 D t )
This filter can be obtained by solving the above optimization problem, with the following results:
where μ is a constant representing the step size, and the value of μ is between 0 and 2. Due to the fact that the time complexity of the algorithm is small, the SDN controller can predict link quality information quickly. In addition, due to the adaptability of the algorithm, the high prediction precision can be always kept.
The invention designs a QoS-sensitive route distribution method in a mobile self-organizing network based on an SDN, which is used for solving an optimal QoS route for each flow in the network in real time, thereby greatly improving the network throughput and reducing the system delay and the packet loss rate; as shown in fig. 3, in practical application, the method specifically includes the following steps:
the SDN controller collects link quality information through a southbound interface, periodically samples the quality information of each link and records the sampled link quality information w t ，r t ，d t Set of samples W forming a link quality prediction t ，R t ，D t Then entering into step B;
and B, the SDN controller adopts a normalized least square method in time sequence analysis to the link quality w at the t +1 moment according to the input link quality sample set t+1 ，r t+1 ，d t+1 Predicting by using the model of fig. 2, and entering step C after the SDN controller obtains link quality information at the time of t + 1;
SDN controller calculates each flow f by Dijkstra algorithm i Feasible path set P i Taking the link quality as a weight, and then entering the step D;
the SDN controller formulates a routing optimization problem according to the collected link quality information, and the optimization aim is to enable the flow f in the network i The method maximizes the QoS performance, ensures to find a routing scheme with the best QoS performance for the network, and comprises the following specific steps:
modeling SDN-based MANETIs a connected undirected graph G (V, E), wherein V is a node set, E is a link set, and a link E uv And (u, v) E represents a link between the node u and the node v, and all the nodes are connected through the wireless link and have the same communication priority. The SDN controller collects link quality information through a southbound interface, and the link quality information of the link e considered herein is the throughput w e Packet loss rate r e And link delay d e The total throughput W, total packet loss rate R and total delay D of all flows in the network can be respectively expressed by the following equations:
wherein x is ip Representing flow f i Whether to select a feasible path P ∈ P i ，P i Is a flow f i F is the flow F i The set of (a) and (b),
indicating whether link e is present in flow f i In the feasible path p. The route optimization problem may be represented by the problem P1:
P1:max W
max D -1
max R -1
x ip ∈{0,1}
the optimization goal of P1 is to make flow f in the network i Wherein the constraint means thatMust follow the flow f i Feasible path set P i To select a path. The optimization objective may ensure that a flow routing scheme with the best QoS performance is found for the network. By solving the above problem, an optimal route allocation strategy can be obtained, the problem is a 0-1 integer programming problem, which is an NP-hard problem, and further transformation of P1 is required for simple solution:
P2:maxαW+εD -1 +γR -1
x ip ∈{0,1}
the problem P2 converts the multi-objective optimization into a single-objective optimization, where α, ε, and γ are weighting coefficients, and α + ε + γ is 1. This conversion has two advantages, one is that the SDN controller can determine the optimal priority of the three QoS performances according to the type of the flow. For example, for a video stream with a high delay requirement, α may be set larger. Secondly, the P2 becomes a typical 0-1 knapsack problem through the transformation, a differential search algorithm can be used for solving the problem, and then the step E is carried out;
step e.sdn controller calculates per flow f by solving a route optimization problem using a differential search algorithm within a route calculation module i The method comprises the following specific processes of sending routing information to a mobile ad hoc node in a network through a flow table mode:
SDN controller calculates per-flow f by solving a route optimization problem within a route calculation module i To solve problem P2, P2 is transformed into the form:
λ 1 ＝1+αW+εD -1 +γR -1
wherein λ 1 Is a constraint factor, X m Is a solution of problem 2, X m ＝[x 1 ,x 2 ,…,x ip ,…,x N ]And N represents the dimension of the solution. N ═ P i If the assigned route satisfies the constraint in problem P2, then:
F(X m )＝max{αW+εD -1 +γR -1 }≥0
if the constraint in question P2 is not satisfied:
at this time F (X) m ) ≦ 1, further converting problem P2 to an unconstrained problem, SDN controller need only be based on F (X) m ) The obtained solution is judged to be a feasible solution, then a route calculation module of the SDN controller can obtain an optimal QoS route by using a differential search algorithm, the differential search is a group intelligent optimization algorithm with excellent performance recently proposed, and the idea of the algorithm comes from random walk behavior of organisms in the migration process. As food and other resources in nature vary with seasons, many organisms are gathered together to form a super organism and move toward more resource-rich areas by using collective wisdom. Based on the principle, the positions of organisms in the super organisms can be regarded as a set of solutions, the position of each organism in the organisms is a single solution, and the process that the super organisms continuously migrate to search resources is a process that the solutions are continuously updated until the organisms find the most suitable habitat, namely the optimal solution of the route. The process of route calculation module within SDN controller using differential search is as follows: first, the SDN controller randomly generates an initial solution matrix M × N, where M represents the number of organisms in the super-organism (i.e., the number of initial solutions). The initial solution represents the initial position of the living being in the super organism, and is set as follows:
at this time X m Should satisfy F (X) m ) When the value is more than or equal to 0, in the process of generating the solution, if X is equal to m If the condition is not met, the solution is discarded, and a new solution meeting the condition is randomly generated. During the process of biological migration, the super-organic body will check whether some randomly selected locations are suitable temporary locations (i.e., determine whether the resulting temporary solution is closer to the optimal solution). If a site is suitable for temporary residence during the migration process, it is found that the creatures of the site are immediately located at the site and then continue to migrate from the site, which is represented by:
S m ＝X m +μ(X d -X m )
X d ＝argmax{F(X m )},m＝1,2,…,M,
wherein S m For the stopover site (i.e. temporary solution) during the migration of the super organism, S m ＝(s m1 ,…,s mn ,…,s mN )，m＝1,2,…,M,S m F (S) should also be satisfied m ) And the SDN controller regenerates the condition that F (S) is met if the condition is not met m ) Random solution S of more than or equal to 0 m ，X d Is X m Zhongshi F (X) m ) The largest solution, μ, is a scale factor, the value of which is determined by the gamma distribution with parameters:
μ＝1/GamRnd(Φ 1 ,Φ 2 )
although the stopover location S is obtained m Due to s, however mn Is a real number and therefore needs to be converted into a binary number:
wherein f(s) mn ) Is a sigmoid function which maps real numbers into the interval 0-1, r, using appropriate values of τ n Is a random number between 0 and 1 which determines s mn If f(s) mn )<r n Then variable s mn Is 0; otherwise it is 1. At each calculation of S m Then, F (X) is added m ) And F (S) m ) A comparison is made. If F (S) m )≥F(X m ) Then solve X m Is updated to S m Otherwise, the solution X is retained m Through g max After the second iteration, the algorithm ends, resulting in F (X) m ) Maximum X m I.e. the optimal route allocation for the flows in the network.
Since the constraint objective is relaxed in the process of converting P1 to P2, some flows may not be assigned routing paths after the differential search algorithm ends, for which the weights of the links are set according to the types of the remaining flows, such as video flows with high delay requirements, the link delay may be taken as a weight, and the SDN controller calculates each remaining flow f using Dijkstra's algorithm i The best path of (a). After the optimal QoS route of each flow in the network is confirmed, the SDN controller issues the flow to the nodes of the mobile ad hoc network in the form of the flow table.
The technical scheme designs a route allocation method sensitive to QoS in the mobile self-organizing network based on the SDN, based on the characteristic of centralized control of the SDN, the QoS performance of a user is maximized as an objective function, a route allocation problem is provided, a time sequence analysis method is adopted, the link quality of each link at the next moment is rapidly and accurately predicted in real time, and therefore complex and variable network topology in the mobile self-organizing network can be coped with. According to the obtained link quality of each link at the next moment, the SDN controller converts a complex route optimization problem into an easily solved route optimization problem, and solves the optimal QoS route for each flow in the network by using a differential search algorithm, so that the network throughput is greatly improved, and the system delay and the packet loss rate are reduced.
The embodiments of the present invention have been described in detail with reference to the drawings, but the present invention is not limited to the above embodiments, and various changes can be made within the knowledge of those skilled in the art without departing from the gist of the present invention.
Claims (2)
1. A QoS-sensitive route allocation method in a mobile ad hoc network based on an SDN is used for realizing optimal QoS route allocation of each flow in the mobile ad hoc network based on an SDN network architecture, and is characterized by comprising the following steps:
the SDN controller periodically samples the quality information of each link through a southbound interface and records the sampled link quality information w t ，r t ，d t Forming a set of link quality samples W t ，R t ，D t Wherein w is t Representing the throughput at time t, r t Representing the packet loss rate at time t, d t Representing the time delay at time t, W t ＝[w t ,w t-1 ,…,w t-Φ ]，R t ＝[r t ,r t-1 ,…,r t-Φ ]，D t ＝[d t ,d t-1 ,…,d t-Φ ]Wherein, each link quality sample set contains phi +1 samples;
the SDN controller carries out link quality w at the t +1 moment according to the input link quality sample set t+1 ，r t+1 ，d t+1 Carrying out prediction;
SDN controller calculates each flow f i Feasible path set P i With link quality as a weight;
d, the SDN controller formulates a routing optimization problem according to the link quality information obtained by predicting in the step B, and the optimization aim is to enable the flow f in the network i Maximize QoS performance;
sdn controller solving route optimization problem to calculate per flow f i The routing information is issued to the mobile self-organizing node in the network through the mode of the flow table;
in the step D, the SDN controller formulates a routing optimization problem according to the collected link quality information, and the specific process is as follows:
modeling an SDN-based MANET as a connected undirected graph G (V, E), where V is a set of nodes, E is a set of links, and link E is a link set uv The E belongs to a link between the node u and the node v, and all the nodes are connected through the wireless link and have the same communication priority; the link quality information of link e is the throughput w e Packet loss rate r e And link delay d e The total throughput W, total packet loss rate R and total delay D of all flows in the network are respectively expressed by the following equations:
wherein x is ip Representing flow f i Whether to select feasible path P ∈ P i ，P i Is a flow f i F is the flow F i The set of (a) and (b),
indicating whether link e is present in flow f i In feasible path p; the route optimization problem may be represented by the problem P1:
P1:max W
max D -1
max R -1
x ip ∈{0,1}
the optimization goal of P1 is to make flow f in the network i QoS performance maximization ofWherein the constraint means that it must be taken from the flow f i Feasible path set P i Selecting a path; by solving the problem P1, an optimal route distribution strategy can be obtained;
the specific process for solving the problem P1 is as follows:
problem P1 was converted into:
P2:maxαW+εD -1 +γR -1
x ip ∈{0,1}
the problem P2 converts multi-objective optimization into single-objective optimization, wherein alpha, epsilon and gamma are weighting coefficients, and alpha + epsilon + gamma is 1;
in step E, the SDN controller solves a route optimization problem in a route calculation module by using a differential search algorithm so as to calculate each flow f i The specific process of the optimal path is as follows:
problem P2 was transformed into the following form:
λ 1 ＝1+αW+εD -1 +γR -1
wherein λ 1 Is a constraint factor, X m Is the solution of problem P2, X m ＝[x 1 ,x 2 ,…,x ip ,…,x N ]N denotes the dimension of the solution, N ═ P i If the assigned route satisfies the constraint in problem P2, then:
F(X m )＝max{αW+εD -1 +γR -1 }≥0
if the constraint in problem P2 is not satisfied:
at this time F (X) m ) ≦ 1, problem P2 translates into an unconstrained problem, SDN controller according to F (X) m ) Judging whether the obtained solution is a feasible solution or not by the numerical value, and then obtaining the optimal QoS route by a route calculation module of the SDN controller by using a differential search algorithm and issuing the optimal QoS route to the nodes of the mobile ad hoc network in a flow table mode.
2. The method of claim 1, wherein the method comprises the following steps: in the step B, a normalized least square method in time sequence analysis is adopted to determine the link quality w at the t +1 moment t+1 ，r t+1 ，d t+1 The prediction is described in detail as follows:
set W of link quality samples t ，R t ，D t And the filter coefficient vector a t ，b t ，c t Multiplying to obtain the link quality information predicted at the time t +1, namely:
w t+1 ＝a t W t
r t+1 ＝b t R t
d t+1 ＝c t D t
the design criteria of the filter are represented by the following optimization problem,
min||a t+1 -a t || 2 +λ(w t+1 -a t+1 W t )
||b t+1 -b t || 2 +λ(r t+1 -b t+1 R t )
||c t+1 -c t || 2 +λ(d t+1 -c t+1 D t )
where λ is the lagrange multiplier, the filter can be obtained by solving the optimization problem described above.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110499275.4A CN113242182B (en) | 2021-05-08 | 2021-05-08 | QoS-sensitive route distribution method in mobile self-organizing network based on SDN |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110499275.4A CN113242182B (en) | 2021-05-08 | 2021-05-08 | QoS-sensitive route distribution method in mobile self-organizing network based on SDN |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113242182A CN113242182A (en) | 2021-08-10 |
| CN113242182B true CN113242182B (en) | 2022-08-16 |
Family
ID=77132542
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110499275.4A Active CN113242182B (en) | 2021-05-08 | 2021-05-08 | QoS-sensitive route distribution method in mobile self-organizing network based on SDN |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113242182B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106341346A (en) * | 2016-09-08 | 2017-01-18 | 重庆邮电大学 | Routing algorithm of guaranteeing QoS in data center network based on SDN |
| CN108512760A (en) * | 2018-03-09 | 2018-09-07 | 西安电子科技大学 | The method for routing of QoS of survice is ensured based on SDN |
-
2021
- 2021-05-08 CN CN202110499275.4A patent/CN113242182B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106341346A (en) * | 2016-09-08 | 2017-01-18 | 重庆邮电大学 | Routing algorithm of guaranteeing QoS in data center network based on SDN |
| CN108512760A (en) * | 2018-03-09 | 2018-09-07 | 西安电子科技大学 | The method for routing of QoS of survice is ensured based on SDN |
Non-Patent Citations (3)
| Title |
|---|
| "QoS Sensitive Routing Algorithm with Link Quality Prediction in SDN-based Ad Hoc Networks";Long Jiang等;《 2020 International Conference on Wireless Communications and Signal Processing (WCSP)》;20201228;全文 * |
| "QoS-aware Routing Optimization Algorithm using Differential Search in SDN-based MANETs";Long Jiang等;《2021 IEEE Global Communications Conference (GLOBECOM)》;20220222;全文 * |
| "基于SDN的自组织网络路由框架及构建方法";董芳 等;《通信学报》;20190930;第40卷(第9期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113242182A (en) | 2021-08-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112491712B (en) | Data packet routing algorithm based on multi-agent deep reinforcement learning | |
| Randhawa et al. | MLBC: Multi-objective load balancing clustering technique in wireless sensor networks | |
| CN109947545A (en) | A kind of decision-making technique of task unloading and migration based on user mobility | |
| CN110198278A (en) | A kind of Lyapunov optimization method in car networking cloud and the scheduling of edge Joint Task | |
| Xu et al. | Joint channel allocation and power control based on PSO for cellular networks with D2D communications | |
| CN102006237A (en) | Routing decision method for delay tolerant network | |
| CN110167204A (en) | A kind of relay transmission policy selection and power distribution method based on MS-BAS algorithm | |
| CN107148064A (en) | A kind of In-vehicle networking routed path optimization method based on population | |
| Cicioğlu | Multi-criteria handover management using entropy‐based SAW method for SDN-based 5G small cells | |
| Sarasvathi et al. | QoS guaranteed intelligent routing using hybrid PSO-GA in wireless mesh networks | |
| CN111538571B (en) | Method and system for scheduling task of edge computing node of artificial intelligence Internet of things | |
| CN103618674B (en) | A united packet scheduling and channel allocation routing method based on an adaptive service model | |
| CN110191480B (en) | Three-dimensional wireless sensor network data collection method with mobile Sink nodes | |
| CN107969008A (en) | A kind of software definition Sensor Network concentrated route computational methods | |
| CN113242182B (en) | QoS-sensitive route distribution method in mobile self-organizing network based on SDN | |
| CN103781140A (en) | Ant colony algorithm-based dynamic spectrum routing management method | |
| CN108834173A (en) | A kind of centralized optimizing distribution method of wireless multi-hop network | |
| Goh et al. | Energy efficient routing for wireless sensor networks with grid topology | |
| CN116634504A (en) | Unmanned aerial vehicle networking topology relation and bandwidth allocation optimization strategy based on improved NSGA-II algorithm | |
| Nabavi et al. | An optimal routing protocol using multi-objective cultural algorithm for wireless sensor networks (ORPMCA) | |
| Chai et al. | A multi-objective Dyna-Q based routing in wireless mesh network | |
| CN115134928A (en) | Frequency band route optimized wireless Mesh network congestion control method | |
| CN111447658A (en) | SDWSN-based clustering routing method | |
| Narendran et al. | Optimized lowest ID in wireless sensor network using Invasive Weed Optimization (IWO)-genetic algorithm (GA) | |
| Sihai et al. | A Weight-based Clustering Routing Algorithm for Ad Hoc Networks |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |