Mobile ad hoc network multipath routing method and device, storage medium and communication method

Abstract

The invention discloses a mobile ad hoc network multipath routing method and device, a storage medium and a communication method, relates to the field of mobile communication, and is used for improving at least one of calculation efficiency, accuracy, objectivity and optimal solution diversity of multipath routing. The method comprises the following steps: constructing a multi-objective optimization model for path routing, wherein the multi-objective optimization model takes an index representing path service quality as an optimization objective; solving the multi-objective optimization model to obtain a non-dominated solution set; and determining a multi-path route according to the non-dominated solution set. According to the method, the objectivity, the accuracy and the calculation efficiency of multi-path routing and the diversity of optimal solution sets are improved, the path routing service quality of converged communication of the satellite internet and the mobile ad hoc network is improved, and the overall link transmission efficiency and the resource utilization rate are improved.

Description

Technical Field

[0001] The present invention relates to the field of mobile communications, and in particular to a mobile ad-hoc network multi-path routing method, device, storage medium and communication method. Background Art

[0002] In scenarios where a large number of ad-hoc network devices cooperate, it is necessary to construct multi-path routing to improve the QoS (Quality of Service) communication quality of the overall Internet transmission link, and to improve the link transmission efficiency and resource utilization rate. Compared with the single-path routing method, multi-path routing not only increases the transmission bandwidth, but also can provide multiple paths for the same type of service, and can achieve higher service quality.

[0003] The multi-path routing problem is mostly solved by transforming it into a multi-objective optimization problem and then using particle swarm algorithm, ant colony algorithm, genetic algorithm, etc. These methods have problems such as long iteration time, low algorithm efficiency, non-convergence of the algorithm, lack of diversity in the optimal solution set, etc. Moreover, the process of solving mainly depends on the prior knowledge and experience of the researcher in the selection of model parameters, and is seriously affected by subjectivity. Summary of the Invention

[0004] The object of the present invention is to provide a mobile ad-hoc network multi-path routing method, device, storage medium and communication method to improve at least one of the calculation efficiency, accuracy, objectivity and diversity of the optimal solution of multi-path routing for all or part of the above problems.

[0005] The technical solution adopted by the present invention is as follows: A mobile ad-hoc network multi-path routing method, which includes: Construct a multi-objective optimization model for path routing, and the multi-objective optimization model uses the index characterizing the path service quality as the optimization target; the multi-objective optimization model is characterized in a stage description form; Divide the multi-objective optimization model into multi-stage routing problems for solution to obtain a non-dominated solution set; Determine the multi-path routing according to the non-dominated solution set.

[0006] The present invention also provides a communication method, wherein a first terminal in a first subnet communicates with a second terminal in a second subnet, and the first subnet and the second subnet communicate through a satellite Internet; the above-mentioned mobile ad-hoc network multi-path routing method is used to determine the multi-path routing between the first terminal and the satellite Internet terminal in the first subnet.

[0007] The present invention also provides a storage medium, which stores a computer program, and running the computer program can execute the above-mentioned mobile ad-hoc network multi-path routing method.

[0008] The present invention also provides a mobile ad-hoc network multi-path routing device, including a processor and a storage medium. A computer program is stored in the storage medium. When the processor runs the computer program in the storage medium, the above-mentioned mobile ad-hoc network multi-path routing method is executed.

[0009] In summary, due to the adoption of the above technical solutions, the beneficial effects of the present invention are as follows: The present invention constructs a multi-objective optimization model for path routing with the index characterizing the path service quality as the optimization goal. This multi-objective optimization model does not require model parameters or coefficients, and will not cause different multi-objective optimization results due to different designers, avoiding the error terms introduced by the subjective thinking of designers and improving the objectivity of multi-path routing. The present invention iteratively solves the non-dominated solution set of the multi-objective optimization model in stages, and derives the solution of the original problem using the solution of the stage routing problem, improving the accuracy of multi-path routing. In the process of solving the multi-objective optimization model in stages, the present invention uses a buffer mechanism to temporarily store the non-dominated solution set of the previous stage, reducing the repeated calculation process, improving the calculation efficiency of the method, and saving storage space. The present invention determines the multi-path routing based on the non-dominated solution set, making the optimal solution set of the multi-path routing diverse. The present invention improves the path routing service quality of the satellite Internet and mobile ad-hoc network integrated communication, and improves the overall link transmission efficiency and resource utilization rate. Description of the Drawings

[0010] The present invention will be described by way of examples with reference to the drawings, where: Figure 1 is a possible path routing diagram between the source node and the destination node provided by the embodiment of the present application.

[0011] Figure 2 is a flowchart of the mobile ad-hoc network multi-path routing method provided by the embodiment of the present application.

[0012] Figure 3 is a flowchart of the single-stage routing problem solving method provided by the embodiment of the present application.

[0013] Figure 4 is a flowchart of the multi-objective optimization model solving method provided by the embodiment of the present application.

[0014] Figure 5 is a network diagram of the communication method provided by the embodiment of the present application. Detailed Embodiments

[0015] All features disclosed in this specification, or all steps in any method or process disclosed, can be combined in any way, except for mutually exclusive features and / or steps.

[0016] Any feature disclosed in this specification (including any additional claims, abstract) can be replaced by other equivalent or alternative features with similar purposes, unless specifically stated. That is, unless specifically stated, each feature is only an example of a series of equivalent or similar features.

[0017] As Figure 1 shown, for the multi-path routing requirements between a source node and a destination node, traditional methods either convert the multi-objective optimization problem into a single-objective optimization problem for separate solutions, or use particle swarm optimization, ant colony optimization, genetic algorithm, etc. to solve the multi-objective optimization problem. The former obtains the multi-objective optimization result by weighted summation of the single-objective optimization results, which faces problems such as difficult determination of weighting values, inconsistent dimensions of each objective, and low robustness; the methods adopted by the latter face problems such as long iteration time, low algorithm efficiency, strong influence of subjective factors, and lack of diversity in the optimal solution set. In view of these deficiencies, the embodiments of the present application provide a multi-path routing method, device, storage medium, and communication method for mobile ad hoc networks (MANETs), which comprehensively consider the indicators reflecting path quality for path planning, aiming to improve one or more of the objectivity, accuracy, and computational efficiency of multi-path routing.

[0018] In some embodiments, as Figure 2 shown, the mobile ad hoc network multi-path routing method includes the following steps: S1. Construct a multi-objective optimization model for path routing, where the multi-objective optimization model takes the indicators characterizing path service quality as optimization objectives.

[0019] In a possible implementation manner, in this step S1, the indicators to be optimized in the constructed multi-objective optimization model are quantifiable indicators, which are used to measure the degree of satisfaction of service requirements.

[0020] In view of the service quality QoS access design characteristics of MANETs, in some embodiments, a multi-objective optimization model is constructed by comprehensively considering two or more of the indicators reflecting bandwidth situation, packet loss situation, and delay situation to characterize path service quality.

[0021] Specifically, as a feasible implementation manner, two or more of the bandwidth availability rate, delivery rate, and load availability rate are selected as the optimization objectives of the multi-objective optimization model. Among them, the bandwidth availability rate reflects the interference and other situations during the data transmission process; the delivery rate reflects the packet loss rate and other situations during the data transmission process; the load availability rate reflects the delay and jitter and other situations during the data transmission process. Through the multi-objective optimization of higher transmission bandwidth, lower data packet loss rate, and lower transmission delay, the service quality of the overall network transmission service can be improved.

[0022] In some embodiments, the index characterizing the service quality of the path is constructed based on the index characterizing the service quality of the corresponding nodes in the path, that is, the service quality of the path is calculated based on the service quality of the corresponding nodes on the path.

[0023] For the service quality index of the node, taking the bandwidth availability, delivery rate and load availability in the previous embodiment as an example, the characterization method is as follows: 1) Node bandwidth availability : Formula (1): ; In the formula, Indicates the measurement period; For Node i Neighboring nodes j exist The NAV (Network Allocation Vector) value in is the DCF interframe space, For Node i The value of the backoff counter. Obviously, the objective function of this indicator does not require model parameters or coefficients and will not introduce the designer's subjective experience bias.

[0024] 2) Node delivery rate : Formula (2): ; Representation Node i To Node i+1 delivery rate.

[0025] 3) Node load availability : Formula (3): ; In the formula, Representation Node i The load queue length, For Node i The cache size.

[0026] Taking the above three indicators as optimization objectives, the objective function of the constructed multi-objective optimization model is: Formula (4): ; In the formula, N Indicates the number of stages the path is divided into, k Indicates k stage. represents the path bandwidth availability, represents the path delivery rate, Indicates the path load availability rate. This objective function represents the multi-objective optimization model in a phased form.

[0027] For an objective function, there are usually corresponding constraint conditions. For a multi-objective optimization model with the index representing the path service quality as the optimization objective, in some embodiments, its constraint conditions include the constraint on the degree of deviation of the corresponding nodes on the path from the direction of the destination node.

[0028] For the degree of deviation of a node from the direction of the destination node, in some embodiments, it is characterized from two dimensions of the offset amount and the offset angle, that is, for the constraint conditions of the multi-objective optimization model, there are constraints in two dimensions of the offset amount and the offset angle.

[0029] Taking the example in the previous embodiment where the multi-objective optimization model is represented in a phased description form, in some embodiments, the above-mentioned offset amount is the lateral offset amount of the current-stage node deviating from the first baseline and the above-mentioned offset angle is the lateral offset angle of the next-stage node deviating from the second baseline . The first baseline is the connection line between the source node and the destination node, and the second baseline is the connection line between the current-stage node and the destination node.

[0030] On the basis that the constraint objects have been determined, the constraint objects in two dimensions are constrained. In some embodiments, the constraint conditions for the lateral offset amount and the lateral offset angle are designed as: Formula (5): ; In the formula, is the lateral offset amount of the k -stage node deviating from the first baseline ; is the minimum lateral offset amount, is the maximum lateral offset amount. is the lateral offset angle of the k -stage, that is, the included angle k between the connection line k+1 from the -stage node to the k -stage node and the second baseline of the -stage node, and respectively represent the minimum value and the maximum value of this included angle ; as a feasible way, set , .

[0031] S2. Solve the multi-objective optimization model to obtain the non-dominated solution set.

[0032] In some embodiments, step S2 is solved using the non-dominated sorting dynamic programming method. Specifically, the multi-objective optimization model is divided into a multi-stage routing problem for solution. By decomposing the problem of solving the multi-objective optimization model into several overlapping sub-problems, iteratively solving and caching the solutions of each sub-problem, and using the solutions of the sub-problems to derive the solution of the multi-objective optimization model. This method is essentially different from decomposing the original problem into single-objective optimization problems. It is a synchronous solution of multiple objectives rather than an independent solution, does not involve the weighting problem, and does not depend on the configuration of model parameters, ensuring the accuracy and objectivity of the solution results. Moreover, the solution results of each stage are cached for use in the next stage, eliminating the need for repeated calculations, improving the computational efficiency of the algorithm, and enabling the reusable caching of resources, saving storage space. The number of divided stages, in some embodiments, is the same as the number of stages divided in the embodiment of the objective function for establishing the multi-objective optimization model in formula (4).

[0033] In some embodiments, the calculation of the non-dominated solution set of the multi-objective optimization model is performed by taking the parameters constrained in the constraint conditions in the previous embodiments as variables for solution.

[0034] Taking the constraint conditions proposed in the previous embodiment of formula (5) as an example, the constrained parameters are the lateral offset and the lateral offset angle. Then, these two parameters can be used as variables to solve the multi-objective optimization model.

[0035] As a feasible approach, in the solution of each stage, the lateral offset of the current stage node from the first baseline is used as the state variable, that is, is used as the state variable x; the lateral offset angle of the next stage node from the second baseline is used as the control variable u, that is, is used as the control variable. The definitions of the first baseline and the second baseline are the same as those in the corresponding previous embodiments. It should be noted that each stage has several state variables x, and each state variable x has several control variables.

[0036] For the multi-objective optimization model after stage division, in some embodiments, the non-dominated solution set of the multi-objective optimization model can be iteratively solved from the last stage to the initial stage according to the divided multiple stages; in other embodiments, the non-dominated solution set of the multi-objective optimization model can be iteratively solved from the initial stage to the last stage according to the divided multiple stages. Regardless of the iteration order adopted, when solving each stage, it is based on the non-dominated solution set obtained from the solution of the previous stage to iteratively solve the non-dominated solution set of the current stage. In this way, by cycling, the non-dominated solution set obtained by solving the last stage is the non-dominated solution set of the multi-objective optimization model.

[0037] In some embodiments, the method for solving the non-dominated solution set of the current stage based on the non-dominated solution set of the previous stage is to calculate the solution of the multi-objective optimization model of the current stage based on the non-dominated solution set solved in the previous stage, and select the non-dominated solution set of the current stage through non-dominated sorting of the solution of the multi-objective optimization model of the current stage.

[0038] It should be noted that since each stage has several state variables, and there are several control variables under each state variable, therefore, the calculation of the non-dominated solution set of the current stage here is for each control variable respectively. That is, for each control variable (representing the k -th stage, the i -th state variable, and the j -th control variable), the corresponding non-dominated solution needs to be solved and then put into the non-dominated solution set. And each stage also needs to traverse each control variable. As Figure 3 shown is the flowchart for solving the non-dominated solution set of a single stage in this embodiment.

[0039] As a feasible implementation, the solution of the multi-objective optimization model of the current stage is obtained by adding the instantaneous objective solution of the current stage to the non-dominated solution set solved in the previous stage. The so-called instantaneous objective solution refers to the values of each optimization objective obtained under the action of the variables in the current stage.

[0040] Taking the calculation starting from the last stage as an example, let ) represent the instantaneous objective solution under the action of the k -th stage, the i -th state variable and the j -th control variable . Let represent the non-dominated solution set of the previous stage (i.e., the k + 1-th stage), then the solution of the multi-objective optimization model of the current stage corresponding to is expressed as . By traversing each state variable k of the current (the -th) stage, as well as the control variables under each state variable , all the instantaneous objective solutions of the current stage can be calculated.

[0041] In some embodiments, when selecting the non-dominated solution set of the current stage through non-dominated sorting of the solution of the multi-objective optimization model of the current stage, for each state variable , at most a set maximum number of solutions are selected.

[0042] In some specific embodiments, the method for selecting the non-dominated solution set of the current stage based on the non-dominated sorting of the solutions of the multi-objective optimization model in the current stage includes: Perform non-dominated sorting on the solutions of the multi-objective optimization model under each state variable of the current stage, and select the non-dominated solution set under each state variable; If the number of solutions in the selected non-dominated solution set exceeds the set maximum number, then screen out the set maximum number of solutions from the selected non-dominated solution set.

[0043] For example, assume that the set maximum number is pareto-num, that is, the number of non-dominated solutions limited for each state variable. For the state variable , by traversing each of its control variables , obtain the set of solutions of the multi-objective optimization model under the state variable . Here, use ) to represent this set. Perform non-dominated sorting on ), and select the non-dominated solution set ). If ) the number of solutions in is greater than pareto-num, then select pareto-num non-dominated solutions from it as ), and this ) is the non-dominated solution set retained when the k stage iterates to the state variable .

[0044] As a feasible implementation for screening out pareto-num non-dominated solutions from ) obtained after non-dominated sorting, when performing non-dominated solution screening, screen according to the order of the crowding degree of each non-dominated solution from large to small. That is, if ) the number of non-dominated solutions in is greater than pareto-num, then calculate the crowding degree of each non-dominated solution among them respectively, and select pareto-num non-dominated solutions according to the order of the crowding degree from large to small as the retained ).

[0045] Traverse each state variable k of the stage by the same method, and respectively obtain the corresponding retained non-dominated solution sets ), that is, complete the solution calculation of the multi-objective optimization model of the k stage. Repeat this solution calculation process for each stage, and the non-dominated solution set of the entire multi-objective optimization model can be obtained.

[0046] As an example executable by a computer, referring to Figure 4 , step S2 includes the following processes: S2.1. Set the lateral offset angle As a control variable , set the lateral offset L As a state variable .

[0047] S2.2. Set the maximum number pareto - num of non - dominated solutions in the retained non - dominated solution set; divide the solution - solving problem of the multi - objective optimization model into N stages, that is , each stage has m state variables, that is, for each stage k, there is , each state variable corresponds to n control variables, that is, for each , there is . Let k = N.

[0048] S2.3. Calculate the instantaneous objective solution generated under ), that is, the values of each optimization objective under the control variable k at the stage and state .

[0049] S2.4. Calculate the cumulative multi - objective optimization model solution under the control variable .

[0050] S2.5. For the state , traverse each , that is , execute steps S2.3 - S2.4, and respectively obtain the under each control variable , and put them into the multi - objective optimization model solution set ).

[0051] S2.6. Perform non - dominated sorting on ), and select the non - dominated solution set ); if the number of non - dominated solutions in is greater than pareto - num, then calculate the crowding degree of each non - dominated solution, and select pareto - num non - dominated solutions in descending order of crowding degree as the retained ).

[0052] S2.7. For the k stage, traverse each , that is , and repeat steps S2.3 - S2.6 for each , that is, obtain the non - dominated solution set k of the stage. Thus, the solution of the non - dominated solution set for a single stage is completed. Let k = k - 1.

[0053] S2.8. Repeat steps S2.3 to S2.7 until k = 1. After execution, the non-dominated solution set of the entire multi-objective optimization model is obtained.

[0054] If iterating from the initial stage to the final stage, then similarly, adaptively modify the parameters related to the stage and the relationship between the previous and subsequent stages in steps S2.2, S2.4, S2.7, and S2.8.

[0055] S3. Determine the multi-path routing according to the non-dominated solution set.

[0056] Determine the multi-path routing according to the non-dominated solution set, so that the optimal solution set of the multi-path routing has the characteristic of diversity.

[0057] The number of path routings corresponding to the non-dominated solution set calculated in step S2 may not be exactly the same as the required number. If the number of path routings exceeds the requirement, then the path routings need to be screened. In some embodiments, when the number of path routings indicated by the non-dominated solution set calculated in step S2 exceeds a predetermined value, select the path routings with the number of the predetermined value from them.

[0058] As a feasible implementation manner, when the number of path routings indicated by the non-dominated solution set exceeds a predetermined value, sort all the path routings according to the set rules, or sort each path routing by calculating the crowding degree of each path routing, and select the top predetermined number of path routings from them.

[0059] The embodiment of the present application also provides a storage medium, which stores a computer program. Running this computer program can execute the mobile ad-hoc network multi-path routing method of the above embodiment.

[0060] In addition, the embodiment of the present application also provides a mobile ad-hoc network multi-path routing device. The device includes a processor and a storage medium. A computer program is stored in the storage medium. When the processor runs the computer program in the storage medium, it executes the mobile ad-hoc network multi-path routing method of the above embodiment.

[0061] The solution provided by the present application can be applied to the mobile Internet communication scenario, or can be applied to the communication scenario of the integration of satellite Internet and mobile Internet.

[0062] For the communication scenario of the integration of satellite Internet and mobile Internet, the present application also provides a communication method. In this method, as Figure 5As shown, the first terminal in the first subnet communicates with the second terminal in the second subnet, and the first subnet and the second subnet communicate via satellite Internet. The multi-path routing method of the mobile ad-hoc network in the above embodiment is used to determine the multi-path routing between the first terminal and the first satellite Internet terminal in the first subnet. This embodiment improves the path routing service quality of the integrated communication between satellite Internet and mobile ad-hoc network, and improves the overall link transmission efficiency and resource utilization rate.

[0063] As a possible implementation, data is transmitted from the first terminal to the second terminal. At this time, in the first subnet, the first terminal is the source node, and the first satellite Internet terminal is the destination node. The multi-path routing method of the mobile ad-hoc network in the above embodiment is used to determine the multi-path routing between the first terminal and the first satellite Internet terminal. The data of the first terminal is transmitted to the satellite Internet through the first satellite Internet terminal, and then transmitted to the second subnet, and received by the second satellite Internet terminal in the second subnet. In the second subnet, the second terminal is the destination node, and this second satellite Internet terminal is the source node. The multi-path routing method of the mobile ad-hoc network in the above embodiment can be used again to determine the multi-path routing between the second satellite Internet terminal and the second terminal.

[0064] The present invention is not limited to the foregoing specific embodiments. The present invention extends to any new feature or any new combination disclosed in this specification, as well as any new method or process step or any new combination disclosed.

Claims

1. A multi-path routing method for a mobile ad hoc network, characterized in that: include: Constructing a multi-objective optimization model for path routing, wherein the multi-objective optimization model takes an indicator representing the path service quality as an optimization target; The multi-objective optimization model is characterized in a staged description form; Dividing the multi-objective optimization model into a multi-stage routing problem for solving to obtain a non-dominated solution set; A multipath route is determined according to the non-dominated solution set.

2. The multi-path routing method for a mobile ad hoc network as claimed in claim 1, characterized in that: The constraint conditions of the multi-objective optimization model include constraints on the degree to which corresponding nodes on the path deviate from the direction of the destination node; The multi-objective optimization model is divided into multi-stage routing problems for solving, including: solving the multi-objective optimization model by taking the constrained parameters in the constraint conditions as variables.

3. The multi-path routing method for a mobile ad hoc network as claimed in claim 2, characterized in that: The degree to which each node on the path deviates from the direction of the destination node is characterized from two dimensions: offset and offset angle; the offset is the lateral offset of the node from the first baseline in the current stage, and the offset angle is the lateral offset angle of the node from the second baseline in the next stage; the first baseline is the line between the source node and the destination node, and the second baseline is the line between the node in the current stage and the destination node.

4. The multi-path routing method for a mobile ad hoc network as claimed in claim 3, characterized in that: In solving each stage, the lateral offset of the node in the current stage from the first baseline is used as the state variable, and the lateral offset angle of the node in the next stage from the second baseline is used as the control variable.

5. The multi-path routing method for a mobile ad hoc network as claimed in claim 4, characterized in that: According to the divided multiple stages, the non-dominated solution set of the multi-objective optimization model is iteratively solved from the last stage to the initial stage; or, the non-dominated solution set of the multi-objective optimization model is iteratively solved from the initial stage to the last stage.

6. The multi-path routing method for a mobile ad hoc network as claimed in claim 5, characterized in that: When solving each stage, the non-dominated solution set of the previous stage is iteratively solved to the non-dominated solution set of the current stage.

7. The multi-path routing method for a mobile ad hoc network as claimed in claim 6, characterized in that: The multi-objective optimization model solution of the current stage is calculated based on the non-dominated solution set solved in the previous stage, and the non-dominated solution set of the current stage is selected based on the non-dominated sorting of the multi-objective optimization model solutions of the current stage.

8. The multi-path routing method for a mobile ad hoc network as claimed in claim 7, characterized in that: By accumulating the instantaneous target solution of the current stage on the basis of the non-dominated solution set solved in the previous stage, the multi-objective optimization model solution of the current stage is obtained.

9. The multi-path routing method for a mobile ad hoc network as claimed in claim 8, characterized in that: By traversing each state variable of the current stage and the control variables under each state variable, all instantaneous target solutions of the current stage are calculated.

10. The multi-path routing method for a mobile ad hoc network according to any one of claims 7 to 9, characterized in that: When the non-dominated solution set of the current stage is selected based on the non-dominated sorting of the solutions of the multi-objective optimization model of the current stage, at most a set maximum number of solutions are selected for each state variable.

11. The multi-path routing method for a mobile ad hoc network according to claim 10, characterized in that: Methods for selecting the non-dominated solution set of the current stage based on the non-dominated sorting of the solutions of the multi-objective optimization model of the current stage include: The non-dominated sorting of the solutions of the multi-objective optimization model under each state variable in the current stage is performed respectively, and the non-dominated solution set under each state variable is selected; If the number of solutions in the selected non-dominated solution set exceeds the set maximum number, the set maximum number of solutions is screened out from the selected non-dominated solution set.

12. The multi-path routing method for a mobile ad hoc network according to claim 11, characterized in that: When selecting solutions from the selected non-dominated solution set, the solutions are selected in descending order according to their congestion degree.

13. A communication method, characterized in that: A first terminal in a first subnet performs data communication with a second terminal in a second subnet, wherein the first subnet and the second subnet communicate via satellite Internet; The multipath routing of the first terminal and the satellite Internet terminal in the first subnet is determined by using the mobile ad hoc network multipath routing method as described in any one of claims 1-12.

14. A storage medium storing a computer program, characterized in that: Running the computer program can execute the multi-path routing method for a mobile ad hoc network as described in any one of claims 1-12.

15. A multi-path routing device for a mobile ad hoc network, comprising a processor and a storage medium, wherein the storage medium stores a computer program, characterized in that: When the processor runs the computer program in the storage medium, it executes the multi-path routing method for a mobile ad hoc network as described in any one of claims 1-12.

Images