Efficient routing algorithm for offshore buoy networking

Abstract

The invention discloses an efficient routing algorithm for offshore buoy networking. The problem that existing offshore buoy networking information transmission is poor in real-time performance and stability is mainly solved. According to the scheme, the method comprises the following steps: inputting the position of an offshore buoy network, constructing a communication link connectivity model, and abstracting a network link topology by means of a graph theory representation mode; comprehensively considering network efficiencies such as time delay, packet loss rate and network load balancing, and determining link cost between two buoy nodes; aiming at the defect that the ant colony algorithm is easy to fall into local optimum, two-point improvement is carried out on the basic ant colony algorithm. The method comprises the following steps: firstly, during construction of initial pheromones, in a first iteration process of an algorithm, 20% of ants in a population are selected, a path with the minimum cost is selected based on a neighbor algorithm, and the remaining ants adopt random selection paths, so that the convergence of the algorithm is improved, and meanwhile, the diversity of the population is ensured; and a second point, aiming at the problem that a large number of ants are easy to walk to a repeated path when the ant colony is converged to the local optimum in the later stage of iteration, adding repeated check operation in an algorithm updating process. According to the specific operation, when 20% of ants select repeated paths, the ants of the repeated paths select new paths in a random mode, the path search space is expanded, and the algorithm performance is improved. According to the method, the efficient route of buoy networking communication can be dynamically planned, and the real-time performance and the stability of the network are met.

Description

Technical Field

[0001] The present invention belongs to the technical field of network communication, and further relates to an efficient routing problem of communication transmission, and can be used for efficient routing selection of offshore buoy networking. Background Art

[0002] As an economical and applicable ocean monitoring device, the buoy networking system can collect underwater information at sea for a long time and complete the task of remote data transmission in the open sea with the characteristics of flexibility, high efficiency, and low self-interference. It plays an irreplaceable role of other devices in the research of underwater acoustic channels. The offshore buoy communicates underwater through optical fiber and underwater acoustic, and performs remote transmission through satellite communication above the water surface. Considering the characteristics of the offshore buoy networking, such as the uncertainty of the network topology structure, the sparsity of nodes, the limited satellite communication bandwidth resources, and the high network transmission delay, when the buoy network collects a large amount of monitoring data and environmental detection data, the service types and traffic volume of the entire network increase sharply. When using the ant colony optimization algorithm to perform routing planning for the buoy networking, the network is likely to fall into problems such as local optimal solutions, search stagnation, and slow initial convergence speed. The important reason is that the update of pheromone in the network cannot accurately reflect the actual situation of the path. How to dynamically plan an efficient transmission routing strategy for information in real time is an important support for realizing large-capacity data transmission in the offshore buoy network. To solve this problem, the algorithm of the present invention mainly optimizes the strategy of pheromone update from two aspects: pheromone initialization and additional detection of duplicate paths, so as to achieve efficient routing of the offshore buoy networking.

[0003] Regarding the efficient routing algorithm for offshore networking communication, many researchers have proposed technical solutions.

[0004] Yang Yuxuan proposed an underwater acoustic opportunistic routing protocol based on distance vector in "Underwater Acoustic Opportunistic Routing Protocol for Multiple Buoys Based on Distance Vector [D]. Guangdong: South China University of Technology, 2020.". This protocol introduces distance vector instead of geographical location to reduce the end-to-end delay of data transmission. However, when facing the topological environment with dense surface buoy nodes, this method does not show the expected good performance and cannot achieve the maximum performance of buoy networking communication.

[0005] Yang Haitao et al. in "Analysis of Optimal Planning Method for Ship Wireless Network Communication Path [J]. Ship Science and Technology, 2021, 43(08): 136-138. DOI: 10.3404 / j.issn.1672-7649.2021.4A.046", with constraints in multiple aspects such as link bandwidth, deployment cost, and time overhead, used the ant colony algorithm to obtain the optimal planning result of the ship wireless network communication path. This method improves the communication rate of the ship and reduces the bit error rate of data transmission. However, this method also does not consider the defect that the ant colony algorithm is prone to falling into local optimum, and does not consider the impact of the movement of ships at sea on the stability of the communication link.

[0006] Aiming at the deficiencies of the existing ant colony algorithm, the present invention proposes two improvements. First, when the algorithm starts the initial iteration, 20% of the ants use the nearest neighbor algorithm to select the next node, and the remaining ants randomly select the next node, updating the initial pheromone, ensuring a certain diversity of the population while improving the convergence speed. The subsequent iteration process updates and optimizes the population in the traditional pheromone-based manner. Second, add an operation to delete duplicate path points. In the later stage of iteration, when 20% of the ants in the population iterate to the same path, delete the same path points, and at the same time add the same number of random solution sets, increasing the chance of free selection of nodes and expanding the search space, which can effectively solve the defect that the ant colony algorithm is prone to falling into local optimum. Summary of the Invention

[0007] The present invention aims at the efficient routing problem of offshore buoy networking, and proposes an efficient routing algorithm based on an improved ant colony algorithm to improve the network performance of buoy networking communication.

[0008] To achieve the above object, the technical solution of the present invention includes the following:

[0009] (1) Construct a communication network link model for offshore buoys, determine the routing nodes of the network, and determine the source routing node and the destination routing node; construct a network link connectivity model based on node positions and node communication distance limitations, and use relevant knowledge of graph theory to represent the communication network link model, which is represented as G=(V, E), where V={v 1 , v 2 …, v n} represents the set of buoy nodes, and E={e 1 , e 2 …, e n} represents the set of links between nodes.

[0010] (2) Determine the measurement criteria for the link cost between buoy nodes. Considering the user Qos requirements and network load comprehensively, use the expected transmission time ETT and the queuing delay of nodes as the measurement criteria for the link cost between buoy nodes;

[0011] (3) Initialize the buoy communication network, set network parameters such as the network bandwidth, propagation delay, packet loss rate, queuing delay, etc., as well as the upper limit of the number of iterations and the number of ants, and calculate the path cost between nodes. The link cost between nodes is a function of the expected transmission time ETT and the queuing delay of the nodes. If the distance between two nodes exceeds the communication distance, the link cost is infinite.

[0012] (4) In the first process of iteration, place m ants at the starting node. 80% of the ants randomly select the next node, and 20% of the ants select the next node according to the nearest neighbor algorithm, and add this node to the taboo list to prevent the generation of loops until reaching the destination node and stopping. If there is no next selectable node, the ant dies. In the remaining iteration process, all ants select the next node according to the selection formula until the iteration ends.

[0013] (5) When all ants reach the target node, complete one cycle. Place the ants at the starting node and update the pheromone on the path. Calculate the path cost of the nodes passed by m ants during the cycle process, find the path with the minimum cost among all the feasible paths found, and record it as the optimal path for this cycle.

[0014] (6) Repeat the path detection operation: Perform the path repetition detection operation in each iteration process. When 20% of the ants select a repeated path, delete all repeated paths, and let the ants that select the repeated path randomly select the next node. When the selection ends, perform the update of the new pheromone.

[0015] (7) Repeat steps (4) to (6) until the given number of iterations is completed or a satisfactory result is obtained. If there is a satisfactory path, then this path sequence is the optimal communication path.

[0016] The present invention has the following advantages compared with the prior art:

[0017] The present invention fully considers the network characteristics of the offshore buoy networking. By improving the ant colony algorithm, it solves the problem of local optimal solutions and improves the efficiency of the routing selection for offshore buoy networking. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 is the implementation flowchart of the present invention;

[0019] Figure 2 is the input buoy networking topology structure;

[0020] Figure 3 is the schematic diagram of the best path solved;

[0021] Figure 4 is the schematic diagram of the comparison result of the convergence performance between the present invention and the traditional ant colony algorithm;

[0022] Figure 5 Schematic diagram of the comparison of the influence of the number of nodes on the present invention and the traditional ant colony algorithm; Detailed implementation manners

[0023] The embodiments and effects of the present invention will be further described in detail below with reference to the accompanying drawings.

[0024] Refer to Figure 1 , the implementation steps of the present invention are as follows:

[0025] Step 1: Construct a network link model for offshore buoy networking. Randomly generate a certain number of buoy communication nodes within a certain area, and determine the source routing node and the destination routing node; construct a network link connectivity model based on the node positions and node communication distance limitations, and use relevant knowledge of graph theory to represent the communication network link model, which is represented as G=(V, E), where V={v 1 , v 2 …, v n} represents the set of communication nodes, and E={e 1 , e 2 …, e n} represents the set of links between nodes, and determine the starting node s and the destination node t.

[0026] Step 2: Determine the measurement criteria for the link cost between nodes. Considering the user Qos requirements, network load, and communication link stability comprehensively, use the expected transmission time ETT and the queuing delay t qd of the node as the measurement criteria for the link cost between nodes;

[0027] (2.1) Expected transmission time ETT:

[0028] ETT = ETX × τ

[0029] Among them, is the expected number of transmissions between two nodes, which is related to the packet loss rate of the node, and is the transmission delay, which is related to the user's data size and network bandwidth. The expected number of transmissions comprehensively considers the bandwidth, delay, and packet loss rate of the link and can well reflect the user's Qos requirements.

[0030] Step 3: Initialize the network parameters, construct a basic buoy networking topology structure, set parameters such as bandwidth, delay, and packet loss rate for the network, calculate the expected transmission time and node queuing delay, set the upper limit of the iteration number and the number of ants of the ant colony algorithm, and set that the link connection nodes that do not meet the transmission distance are mutually unreachable. According to the network parameters, calculate the link cost between each buoy node and use its reciprocal as the heuristic factor of the ant colony algorithm.

[0031] (3.1) The calculation formula for the link cost between node i and node j is as follows:

[0032]

[0033] Among them, Cost i→j represents the path cost between node i and node j, and ett i→j , respectively represent the expected transmission time, queuing delay, and ETT from node i to node j max , T qdmax respectively represent the maximum values of the expected transmission time and queuing delay in the network, ω 1 , ω 2 is a parameter for adjusting the weight of the influence of the expected transmission time and queuing delay on the link cost. Here, both values are 0.5.

[0034] Step 4: In the first iteration process, 20% of the m ants select nodes using the nearest neighbor algorithm, and 80% of the ants construct paths in a random selection manner. 80% of the ants randomly select paths to construct the initial pheromone. In subsequent iteration processes, place the m ants at the starting node. Each ant selects the next node according to the formula probability selection formula and adds the node to the taboo list to prevent loops until reaching the destination node and stopping.

[0035] (4.1) At the initial stage of the algorithm, probability selection is performed according to the heuristic factor and the pheromone on the path. The probability selection formula is as follows:

[0036]

[0037] is the probability that the k-th ant selects node j from node i, τ ij (t) is the pheromone concentration between node i and node j, η ik (t) is the heuristic factor, which is the reciprocal of the path cost between ij here.

[0038] Step 5: When all ants reach the target node, one cycle is completed. The ants are placed at the starting node to update the pheromone T ij on the path. Denote the paths walked by the m ants as Pm, calculate the path cost Costm of the nodes passed through during the cycling process of the m ants, find the path with the minimum cost among all the feasible paths found, and record it as the optimal path for this cycle.

[0039] Step 5: (5.1) Pheromone update strategy formula:

[0040] τ ij (t + 1) = (1 - ρ) * τ ij (t) + Δτ ij (t)

[0041] Among them, ρ is the pheromone evaporation factor, Δτij (t) is the pheromone increment on path ij. Its calculation method is as follows:

[0042] (5.2) Pheromone increment on the optimal path

[0043]

[0044] where Q is a fixed value, and l min is the cost of the optimal path at the end of this iteration. At the same time, to prevent the pheromone concentration on local paths from being too small or too large, the pheromone concentration on the path is maintained within the range of (τ min , τ max ).

[0045] (5.3) Total path cost on a path

[0046] Cost m = ∑Cost i→j

[0047] where i and j are all the nodes passed by ant m.

[0048] Step 6, repeated path detection operation: Perform the repeated path detection operation in each iteration. When 20% of the ants choose a repeated path, delete all repeated paths, and let the ants that choose the repeated path randomly select the next node. When the selection is completed, update the new pheromone.

[0049] Step 7, repeat steps (4) - (6) until the given number of iterations is completed or a satisfactory result is obtained. If there is a satisfactory path, then this path sequence is the optimal communication path.

[0050] II. Simulation experiment content

[0051] Conduct a simulation experiment on the proposed efficient routing algorithm for buoy networking and compare it with the basic ant colony algorithm.

[0052] Simulation experiment 1: Randomly generate 80 communication nodes with a uniform distribution within the range of 1000*1000, set the communication distance to 250, and the generated network topology is as Figure 2 shown, where node 1 is the source routing node and node 80 is the destination routing node. Set the communication network parameters and initialize the ant colony parameters. The number of ant colonies is set to 50, and the number of iterations is set to 100 times. The obtained simulation results are as Figure 3 . Figure 3 It shows that the optimal communication path planned by the algorithm proposed in the invention is 1 - 78 - 43 - 68 - 30 - 51 - 80.

[0053] Simulation Experiment 2: The improved ant colony algorithm proposed in the present invention is compared with the basic ant colony algorithm. The simulation scenario is the same as that in Experiment 1. The optimal path costs of the two algorithms during the iteration process are compared. The results are as Figure 4 , from Figure 4 It can be seen that the improved ant colony algorithm proposed in the present invention can expand the search range of solutions, and the optimal solution obtained is better than that of the basic ant colony algorithm.

[0054] Simulation Experiment 3: The improved ant colony algorithm proposed in the present invention is compared with the improved ant colony algorithm. The number of buoy nodes is changed. When the number of nodes increases from 40 to 140, the changes in the optimal path costs planned by the two algorithms are as Figure 5 shown. From Figure 5 It can be seen that when the number of buoy nodes is small, the advantages and disadvantages of the paths planned by the two are not much different. When the number of nodes increases, the path cost obtained by the improved ant colony algorithm proposed in the present invention is lower than that of the basic ant colony algorithm, which proves that the improved ant colony algorithm proposed in this paper can expand the search space during operation, well solve the defect that the ant colony algorithm is prone to fall into local optimal solutions, and is more likely to plan an optimal path with a small cost.

[0055] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Obviously, for professionals in the field, after understanding the content and principle of the present invention, various corrections and changes in form and details may be made without departing from the principle and structure of the present invention. However, these corrections and changes based on the idea of the present invention are still within the scope of protection of the claims of the present invention.

Claims

1. An efficient routing algorithm for offshore buoy networking, characterized in that: include: (1) Input the network link model of offshore buoy networking, randomly generate a certain number of buoy communication nodes in a certain area, and determine the source routing node and the destination routing node; construct a network link connectivity model based on the node position and node communication distance limit, and use graph theory to represent the communication network link model, which is expressed as G = (V, E), where V = {v1, v2…, v n } represents the communication node set, E = {e1, e2…, e n } represents the set of links between nodes. (2) Determine the link cost metric between nodes, taking into account user QoS requirements and network load, using the expected transmission time ETT and the node queuing delay t qd As a metric for link costs between nodes; (3) Initialize the network, set the network parameters such as bandwidth, propagation delay, packet loss rate, queuing delay, as well as the upper limit of iterations and the number of ants, and calculate the path cost between nodes. The link cost between nodes is the expected transmission time ETT, the queuing delay t of the node qd If the distance between two nodes exceeds the communication distance, the link cost is infinite. (4) In the first iteration, m ants are placed at the starting node, and 80% of the ants choose the node according to the probability formula Select the next node. 20% of the ants select the next node according to the greedy rule of the nearest neighbor algorithm and add the node to the taboo table to prevent loops until they reach the destination node. If there is no next node to select, the ant dies. In the remaining iterations, all ants follow the selection formula Select the next node until the iteration ends. (5) When all ants reach the target node, a cycle is completed. The pheromone T placed by the ants on the path to the starting node ij Update, record the path taken by m ants as P m , calculate the path cost Cost of the nodes passed by the m ants in the cycle process m , find the path with the lowest cost among all feasible paths and record it as the optimal path for this cycle. (6) Duplicate path detection operation: A duplicate path detection operation is performed during each iteration. When 20% of the ants choose duplicate paths, all duplicate paths are deleted and the ants that choose duplicate paths are allowed to randomly select the next node. When the selection is completed, new pheromones are updated. (7) Repeat steps (4) to (6) until the given number of iterations is completed or a satisfactory result is obtained. If a satisfactory path exists, the path sequence is the optimal communication path.

2. The method according to claim 1, wherein (2) the calculation method of each metric is as follows: 2a) Expected Transmission Time (ETT): ETT=ETX×τ in, ETX is the expected number of transmissions between two nodes, which is related to the packet loss rate of the nodes. It is the transmission delay, which is related to the user's data size and network bandwidth. The expected number of transmissions takes into account the link's bandwidth, delay and packet loss rate, and can well reflect the user's QoS requirements. 2b) Queuing delay T qd : Considering the load of each buoy node in the communication network, the queuing delay of each node is used as the measurement standard of load balancing. When selecting the next hop node, the queuing delay of the candidate node is considered, and the node with a small queuing delay is preferred when selecting the next hop node.

3. The method as claimed in claim 1, wherein in (3), the calculation formula for the link cost between node i and node j is as follows: Cost i→j represents the path cost between node i and node j, ett i→j , They represent the expected transmission time from node i to node j, queuing delay, and ETT respectively. max , T qdmax They represent the maximum values ​​of the expected transmission time and queuing delay in the network respectively. ω1 and ω2 are parameters for adjusting the weights of the expected transmission time and queuing delay on the link cost. Here, both values ​​are 0.

5.

4. The method as claimed in claim 1, wherein in (4), when the ant selects the next node, the probability selection formula is improved. 4a) The ant colony algorithm is prone to fall into a local optimal solution after a few iterations. Therefore, after the algorithm has been running for a period of time, the method of dynamically expanding the transition probability should be adopted to expand the search space and increase the probability of finding a better solution. 4b) At the beginning of the algorithm, the probability selection is performed according to the heuristic factor and the pheromone on the path. The probability selection formula is as follows: is the probability that the Kth ant chooses node j from node i, τ ij (t) is the pheromone concentration between node i and node j, η ik (t) is the heuristic factor, which is the inverse of the path cost between ij. 4c) When the optimal path is the same for five consecutive iterations, the path selection probability is modified, and the balance parameter q0 (0<q0<1) is designed. Then, a random number q is generated using the random function rand(). The improved probability selection formula is as follows: Where L n The number of nodes that can be selected as the next hop for node i.

5. The method according to claim 1, wherein in (5), the pheromone update strategy is improved as follows: 5a) Pheromone update formula: t ij (t+1)=(1-ρ)*τ ij (t)+Δτ ij (t) Where ρ is the pheromone volatility factor, Δτ ij (t) is the pheromone increment on path ij. 5b) Pheromone increment calculation formula: Where Q is a constant, l min is the cost of the optimal path at the end of this cycle. At the same time, in order to prevent the pheromone concentration on the local path from being too small or too large, the pheromone concentration on the path is set to be maintained at (τ min , τ max ) range. 5c) Total path cost on a path Cost m =∑Cost i→j Where i, j are all the nodes that ant m has walked through.

Images