A node clustering routing method and device for underwater wireless sensor networks

Abstract

This invention relates to a node clustering routing method and device for underwater wireless sensor networks. The underwater wireless sensor network includes a sink node, a base station, and multiple dynamic underwater nodes. The routing method is applied to the sink node and includes: establishing a connection with the base station; for each underwater node, forming multiple clusters through clustering; for each cluster, determining the corresponding cluster head node using a hybrid optimization algorithm combining fruit fly and firefly algorithms, and then establishing a connection with the corresponding cluster head node; determining the target cluster where the target underwater node is located, determining the optimal path from the cluster head node of the target cluster to the target underwater node using an adaptive step-size firefly algorithm, and then establishing a connection with the target underwater node based on the optimal path, thus establishing a route from the base station to the target underwater node. Compared with existing technologies, this invention has advantages such as strong search capability and fast convergence speed.

Description

Technical Field

[0001] This invention relates to the field of sensor network technology, and in particular to a node clustering routing method and device for underwater wireless sensor networks. Background Technology

[0002] Underwater wireless sensor networks (UWSNs) consist of mobile and fixed nodes that interact via an acoustic network. In recent years, UWSNs have received widespread attention in fields such as disaster management, water quality prediction, environmental monitoring, and underwater navigation. UWSNs comprise a large number of sensors placed in rivers and oceans to observe the underwater environment.

[0003] However, underwater sensors are energy-constrained, and charging / replacing batteries is very cumbersome, making energy efficiency a major challenge. Clustering and multi-hop routing protocols are considered energy-saving solutions for UWSNs. However, due to factors such as underwater current, low bandwidth, high water pressure, propagation delay, and error probability, traditional wireless network cluster-based routing protocols are not feasible in UWSNs.

[0004] Topology control using a clustering model can balance energy utilization, extend network lifespan, and reduce communication interference. Different noise sources and water currents limit the bandwidth of UWSNs. For long-distance transmission in UWSNs, sea waves, which can propagate for kilometers at higher power and lower frequencies, are utilized. Underwater networks face connectivity issues and high bit error rates due to multipath interference from marine networks. Underwater transmission is unreliable due to water temperature, multipath noise effects, and Doppler diffusion. Furthermore, multipath effects cause input signal attenuation. Therefore, routing becomes expensive and challenging in UWSNs.

[0005] Chinese patent application CN201710127088.7 discloses an improved minimum weighted clustering method for wireless sensor networks, belonging to the field of wireless sensor communication technology. Based on the relationship between the remaining energy of a node and the distance between two nodes, a cost function is proposed to quantify the cost of communication between two nodes. The density of sensor nodes is introduced into the weighting formula, which improves the applicability of the clustering method. In order to reduce energy consumption, a method is proposed to select a certain proportion of sensors as cluster heads and nodes that transmit data within the cluster, which improves the efficiency of cluster head selection.

[0006] The aforementioned applications improve the efficiency of cluster head selection and reduce energy consumption. However, existing methods still suffer from drawbacks such as poor local search capability and slow search space convergence speed.

[0007] In recent years, new metaheuristic and natural heuristic algorithms have been developed and widely used in UWSN, but most of them suffer from unreliability, immature performance, incomplete verification, and overuse. Summary of the Invention

[0008] The purpose of this invention is to overcome the shortcomings of the existing technology and provide a node clustering routing method and device for underwater wireless sensor networks. First, a cluster head is determined based on a fruit fly and firefly hybrid optimization algorithm. Then, a firefly algorithm with an adaptive step size is used to determine the routing path from the cluster head to the target underwater node. Finally, a route from the base station to the target underwater node is established.

[0009] The objective of this invention can be achieved through the following technical solutions:

[0010] One aspect of the present invention provides a node clustering routing method for an underwater wireless sensor network, the underwater wireless sensor network including a sink node, a base station, and multiple dynamic underwater nodes, the routing method being applied to the sink node, specifically including the following steps:

[0011] Establish a connection with the base station;

[0012] Establish connections with the cluster head nodes corresponding to each sub-cluster. The process of determining the cluster head nodes is as follows: for each underwater node, multiple sub-clusters are formed by clustering. For each sub-cluster, the corresponding cluster head node is determined by a hybrid optimization algorithm of fruit flies and fireflies.

[0013] Based on the optimal path, a connection with the target underwater node is established, and a route is established from the base station to the target underwater node. The process of determining the optimal path is as follows: determine the target cluster where the target underwater node is located, and determine the optimal path from the cluster head node of the target cluster to the target underwater node through the adaptive step size firefly algorithm.

[0014] As a preferred technical solution, the process of determining the corresponding cluster head node using a hybrid optimization algorithm combining fruit flies and fireflies includes the following steps:

[0015] Select one or more underwater nodes and initialize them as cluster head nodes;

[0016] For each remaining underwater node, the corresponding first fitness function value is calculated based on the normalized distance between the underwater node and each cluster head node, the distance between the cluster head node and the base station, and the energy utilization rate of the broadcast data packet.

[0017] The underwater nodes are sorted based on the first fitness function value. For underwater nodes whose serial number is less than or equal to a preset value, the corresponding first fitness function value is updated using the firefly optimization algorithm. For underwater nodes whose serial number is greater than or equal to the preset value, the corresponding first fitness function value is updated using the fruit fly algorithm. This step is repeated multiple times.

[0018] The final cluster head node is determined based on the final first fitness function value.

[0019] As a preferred technical solution, the first fitness function value is obtained using the following formula:

[0020] F c =W1×dist CM +W2×dist CB +W3×E To

[0021] Where W1, W2, and W3 are weighting coefficients, dist CM It is the sum of the distances between the underwater node and the cluster head nodes of all sub-clusters, dist CB E is the distance between the cluster head node and the base station. T0 This indicates the energy utilization rate of network broadcast data packets.

[0022] As a preferred technical solution, the normalization process is implemented using the sigmoid function.

[0023] As a preferred technical solution, the process of determining the optimal path from the cluster head node to the target underwater node using the adaptive step-size firefly algorithm includes the following steps:

[0024] Obtain the set of alternative routes from the cluster head node of the target cluster to the target underwater node, calculate the second fitness function value corresponding to each alternative path, and use the firefly algorithm for iteration. During the iteration process, determine the current firefly's movement step size based on the number of iterations and the distance between the current firefly and the optimal firefly.

[0025] The optimal path is determined based on the final second fitness function value.

[0026] As a preferred technical solution, during the iteration process, the distance between the current firefly and the optimal firefly is inversely proportional to the current firefly's movement step size.

[0027] As a preferred technical solution, the current movement step size of the firefly is obtained using the following formula:

[0028]

[0029] Where, x i (t) represents the position of firefly i in the t-th iteration, x b (t) represents the position of the optimal firefly in the t-th iteration, D i (t) represents any number in a uniform distribution, N t Indicates the maximum number of iterations. Let be an exponentially decaying function, and e represent the natural logarithm.

[0030] As a preferred technical solution, the second fitness function value is obtained using the following formula:

[0031]

[0032] Among them, F r Let b represent the fitness function, h represent the delay, β represent the distance, and μ represent the trust level.

[0033] In another aspect, an electronic device is provided, comprising: one or more processors and a memory, wherein the memory stores one or more programs, the one or more programs including instructions for executing the above-described node clustering routing method for underwater wireless sensor networks.

[0034] In another aspect, the present invention provides a computer-readable storage medium including one or more programs executable by one or more processors of an electronic device, the one or more programs including instructions for performing the above-described node clustering routing method for underwater wireless sensor networks.

[0035] Compared with the prior art, the present invention has the following advantages:

[0036] (1) Improved Search Capability: This invention first determines the cluster head node using a hybrid optimization algorithm combining fruit fly and firefly algorithms. Then, it uses an adaptive step-size firefly algorithm to determine the optimal path from the cluster head node of the target cluster to the target underwater node, thereby establishing a route from the base station to the target underwater node. Existing schemes using firefly optimization to determine cluster head nodes suffer from slow computation speed and poor local search capability, while fruit fly optimization schemes suffer from slow convergence speed in the search space. Unlike existing methods, this invention uses a hybrid optimization algorithm combining fruit fly and firefly algorithms to determine the cluster head, improving search capability while reducing negative search capability and improving convergence. Based on the intra-frame distance and inter-frame distance from the cluster head to the node, it achieves multiple objectives such as minimizing latency and maximizing energy, thereby improving the energy efficiency and lifetime of the UWSN.

[0037] (2) Fast convergence speed: The adaptive step size firefly algorithm determines the optimal path from the cluster head node of the target cluster to the target underwater node. The traditional firefly optimization algorithm uses a fixed step size, which leads to slow convergence speed. Unlike existing methods, this invention uses an adaptive step size and comprehensively considers parameters such as trust, energy, delay and distance to determine the optimal path for data transmission between the transmitter and the destination through the cluster head in the underwater wireless sensor network, which can greatly improve the overall network efficiency. Attached Figure Description

[0038] Figure 1 This is a flowchart of the wireless sensor network node clustering and multi-hop routing protocol optimization method in the embodiment;

[0039] Figure 2 Diagram of an underwater wireless sensor network model;

[0040] Figure 3 Flowchart of the optimization algorithm for fruit flies and fireflies. Detailed Implementation

[0041] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0042] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, "multiple" means two or more, unless otherwise explicitly specified.

[0043] Example 1

[0044] like Figure 1 This embodiment provides a node clustering routing method for underwater wireless sensor networks, including the following steps:

[0045] S1. Establish an underwater sensor network model and design an energy model according to the energy strategy of the underwater acoustic channel to obtain data information for S2 cluster analysis. The specific steps are as follows:

[0046] S11, Network Model as follows Figure 2 As shown, a dynamic network consists of N sparsely and randomly distributed nodes in an L×L×L space, collecting sensing data about the water medium via underwater sensors. The detected parameters are pressure, temperature, and current. The underwater sensors are equipped with acoustic modems, allowing them to communicate with another node underwater. A convergence node (SN) is located at the water surface and is equipped with both radio frequency (RF) and acoustic modems; the SN acoustic modem collects data from the underwater sensors, while the RF modem transmits the data to a base station (BS). This network scenario is assumed to be connected to a network. Because the underwater sensors are movable under the influence of water currents with velocities of approximately 1-3 m / s, their topology changes rapidly.

[0047] S12. Because the characteristics of sound waves in underwater broadcasting media differ from those of radio waves, the energy model for wireless sensor networks (WSNs) cannot be applied to underwater wireless sensor networks (UWSNs). Applying the energy strategy of the underwater acoustic channel to this method, the energy used to transmit k bits of data at a data rate R over a distance d is calculated as follows.

[0048]

[0049] Where E elec P represents the power consumption of routing 1 bit of data. Tx This indicates the power transmitted.

[0050] The power consumption of the receiver when receiving k bits of data via radio is as follows.

[0051] B Rx (k)=kP r

[0052] Let P r The power consumption based on device constants and fused k-bit data is as follows.

[0053] E DA (k)=k×K DA0

[0054] Where E DA0 This represents the energy consumed to fuse 1 bit of data, taken as 5 nJ / bit. Since the nodes are mobile due to the water flow, they are allowed arbitrary movement during operation. The current speed is 1-3 m / s.

[0055] S13. After the network model is deployed, the base station broadcasts a beacon signal to all nodes in the detection area. Each node then receives the transmitted signal, which is used to calculate its estimated distance to the base station. Next, the detector sends handshake messages to all nodes within its communication radius to gather information about its neighbors.

[0056] S2. Perform cluster analysis and set the fitness function for route selection. The fitness function is set based on four factors: trust level, remaining energy, delay, and distance. The specific steps are as follows:

[0057] S21. Cluster head election is based on the Firefly Algorithm (GSO) and Fruit Fly Algorithm (FFOA) hybrid optimization algorithm (FGSF). The fitness function considers the distance between the sensor node and all cluster heads (CH), the distance between the CH and the base station (BS), and the energy utilization rate of network broadcast data packets. The selected CH nodes in the network should all satisfy the maximum value provided by the fitness function, which is as follows.

[0058] F c=W1×dist CM +W2×dist CB +W3×E To

[0059] Where W1, W2, and W3 are the weighting coefficients that change to determine the priority order of the three factors, and W1 + W2 + W3 = 1. CM The distance between the sensor node and the CH of all clusters is represented by the following formula.

[0060]

[0061] Among them CM ij CH represents a member node from cluster j. j Let CH represent the CH from cluster j. In all clusters, N j Each node.

[0062] dist CB The distance between the CH and the base station (BS) is expressed by the following formula.

[0063] dist CB =d(CH j ,BS)

[0064] BS refers to the location of BS. E To The energy utilization rate of network broadcast data packets is expressed by the following formula.

[0065]

[0066] Where c represents the number of clusters, N i This represents the number of nodes in the i-th cluster. This represents the energy consumed by the i-th cluster head node in sending data packets. The energy consumed by the j-th member node in the i-th cluster to send a data packet to the cluster head node can be derived from the energy model in S12.

[0067] S22. Use the normalization function sigmoid to eliminate the scale differences among the three factors, expressed as:

[0068]

[0069] S3. The concept of the Firefly Optimization Algorithm (FFOA) is integrated into the Fruit Fly Optimization Algorithm (GSO), combining the best characteristics of both algorithms to propose the FGSF algorithm, which is then used for cluster head election. The specific steps are as follows:

[0070] S31. Based on the fitness function used in S2 to select the optimal cluster head node, the Fruit Fly and Firefly Hybrid Optimization Algorithm (FGSF) is used iteratively. Although GSO can handle nonlinear, multi-model problems, this algorithm is slow in computation and has poor local search capabilities. Similarly, the FFOA algorithm also suffers from slow convergence speed in the search space. FGSF incorporates the concept of FFOA into GSO, overcoming the problems of both traditional algorithms and possessing the best characteristics of both. It achieves multiple optimization objectives such as minimizing time delay and maximizing energy, improving search capability while reducing negative search capability, thus improving convergence. The specific steps of FGSF are as follows.

[0071] S32. Initialize the FGSF algorithm parameters. Nodes are randomly distributed in the search space, and then the nodes are assigned similar fluorescein intensities and similar decision domains GM0.

[0072] S33. Calculate the fitness value of each node based on the fitness function used in S2 to select the best cluster head node, and sort and number all sensor nodes according to their fitness values.

[0073] S34. If the number is less than or equal to 6, then the firefly optimization algorithm is selected for updating.

[0074] In the GSO algorithm, the swarm consists of a group of fireflies randomly distributed throughout the search space. Each firefly has a property called "luciferin," which represents its luminescence intensity. Furthermore, the fireflies are located in the decision domain... Fireflies move according to their own decision-making rules. The mechanism of interaction between fireflies is as follows: if a firefly w is within the neighborhood of another firefly g, and w's luciferin intensity is higher than g's, then g will move closer to w. Luciferin intensity is positively correlated with the fitness of the current location. Therefore, the optimal location is the location of the firefly with the highest luciferin intensity.

[0075] The neighborhood is generally defined as along... The local decision domain, which extends to the radial sensor range. The firefly selects a neighboring firefly with the highest luciferin value and moves towards it. The GSO algorithm consists of four phases:

[0076] (1) Initialization: Fireflies are randomly distributed in the search space. Then, nodes are assigned similar luciferin intensities and similar decision domains GM0.

[0077] (2) Luminescent Intensity Update: The luciferin intensity of a firefly reflects the fitness of its position. Higher luciferin intensity indicates that the firefly is closer to the optimal position, i.e., the optimal target value. Conversely, lower luciferin intensity indicates a poorer target value. During the iteration process, the firefly's position changes continuously, and the luciferin intensity is updated accordingly. Let X be the position of the g-th firefly at time t. g (t), then its corresponding objective function value is J(X) g (t)). Furthermore, J(X) g Substitute (t) into LU g (t), defined in formula LU g (t)=(1-v)LU g (t-1)+η(J(X g In (t)), v represents the fluorescein decay constant 0 < t < 1, and η represents the fluorescein improvement constant.

[0078] (3) Movement: In this stage, each firefly selects its neighbor based on two conditions and follows it with a certain probability. These two conditions are: the neighbor firefly w is located within the decision domain of firefly g, and the luciferin intensity of w is higher than that of g. Let PT... gj (t) represents the firefly g moving from N. g The probability that the initial neighbor w moves is given by (t), and its expression is:

[0079]

[0080] Once the firefly moves, its position is updated. The evaluation of the position update is based on the formula. Given, where size represents the step size.

[0081] (4) Neighborhood Range Update: After updating the firefly positions, update the neighborhood range. If the neighborhood range only covers a small density of fireflies, then the neighborhood range will be increased; otherwise, the neighborhood range will be decreased. The update formula is in the formula... The expression is given, where λ represents a constant parameter. Here, nu t It is used as a parameter to control the number of neighbors.

[0082] S35. If the number is greater than 6, then select the fruit fly optimization algorithm for updating.

[0083] FFOA is based on the behavior of fruit flies searching for food. The algorithm consists of four stages:

[0084] (1) Initialization phase: In this phase, the fruit flies are randomly dispersed into X groups. a xis and Y a xis, as shown in the equation. Where rv is a random value. A random value represents a "random vector" sampled from a uniform distribution.

[0085] X i =X a xis+rv

[0086] Y i =Y a xis+rv

[0087] (2) Path construction stage: Here, the distance and odor concentration value of each fruit fly are determined according to the formula.

[0088]

[0089]

[0090] Distance i This represents the distance between the i-th individual and the food location. The judgment value representing the concentration of flavor is the reciprocal of the distance.

[0091] (3) Fitness evaluation stage: The fitness formula is defined as follows.

[0092]

[0093] smell best index best =max(smell) i )

[0094] In the formula, smell i The smell concentration value represents the characteristic fruit fly odor concentration. best and index best These represent the highest element and its corresponding exponent in different dimensions of the odor vector, respectively, max(smell) i () indicates the maximum odor concentration in fruit flies.

[0095] (4) Movement phase: The fruit fly gives the optimal value of odor concentration and flies toward that location.

[0096] BEST SMELL = semll best

[0097] X a xis = X(index) best )

[0098] Y a xis = Y(index) best )

[0099] S36. Find the candidate cluster head node with the smallest fitness function as the best cluster head node, and the remaining unselected nodes become ordinary nodes.

[0100] S37. Each cluster head node performs a broadcast operation so that each ordinary node can select the nearest cluster head node and join its cluster. This step continues until all ordinary nodes have joined a cluster. At this point, there will be multiple clusters in the network, each containing one cluster head node and several ordinary nodes. Furthermore, the cluster head node and ordinary nodes of each cluster are different.

[0101] S4. Set the fitness function for route selection. The fitness function is set based on four factors: trust level, remaining energy, delay, and distance. The specific steps are as follows:

[0102] S41. Route path finding is based on the Adaptive Step-Size Firefly Optimization (SAS-GSO) algorithm. The fitness function considers four factors: trust level, remaining energy, delay, and distance. The optimal route path satisfies the minimum value provided by the fitness function, which is expressed as follows:

[0103]

[0104] Among them, F r Let b represent the fitness function, h represent the delay, β represent the distance, and μ represent trust.

[0105]

[0106] Among them, b K This represents the remaining energy of the Kth node.

[0107]

[0108] Where 'a' is the distance between the node and the cluster head, and 'm' represents the node's transmission rate.

[0109]

[0110] Where β(K,T) is the Euclidean distance between the Kth node and the Tth node, and η represents the normalization factor.

[0111]

[0112] Where η represents the normalization factor, and DT, RT, and HT represent the data transmission time, routing time, and processing time between nodes, respectively.

[0113] S5. After examining the impact of the movement step size on the GSO algorithm, an adaptive step size was used to replace the fixed step size in the traditional GSO algorithm, resulting in the Adaptive Step Size Firefly Algorithm (SAS-GSO). The SAS-GSO algorithm was then used to find the optimal routing path. The specific steps are as follows:

[0114] S51. Based on the fitness function used in S4 to select the optimal cluster head node, an adaptive step-size firefly algorithm (SAS-GSO) is used for iteration. In the original GSO algorithm, individual fireflies move towards brighter neighbors with a fixed step size to approach the optimal solution. However, this approach has some drawbacks. On the one hand, when individual fireflies are close to the optimal solution, the fixed step size may cause them to overshoot the optimal solution and fail to converge; on the other hand, if the step size is too small, it will reduce the movement speed of individual fireflies and the convergence speed of the algorithm. Therefore, a fixed step size parameter makes it difficult for the algorithm to achieve ideal results. This invention considers two factors affecting the step size, namely the number of rounds in the firefly algorithm and the distance between the firefly in the nth round and the optimal firefly, and derives the adaptive step-size firefly algorithm (SAS-GSO), the steps of which are as follows.

[0115] S52. Initialize the firefly deployment. Nodes are randomly distributed in the search space. Initialize the luciferin value and decision domain GM0 of each node.

[0116] S53. Calculate the fitness value of each node as the basis for updating its fluorescein value. The fluorescein value of each node is equal to the fluorescein value at the previous moment plus a certain extraction ratio of the node's current fitness value, and then subtracts a certain proportion of the fluorescein value that has evaporated over time.

[0117] LU g (t)=(1-v)LU g (t-1)+η(J(X g (t)))

[0118] S54. Each node selects an optimal target node based on its own and its neighboring nodes' fluorophore levels and distances, and moves towards it by a certain step size.

[0119] In the GSO algorithm, the step size is specified as a fixed value. Since the choice of step size is crucial to the effectiveness of the firefly algorithm, this invention considers two factors affecting the step size: the number of rounds in the firefly algorithm and the distance between the firefly in the nth round and the optimal firefly. If the i-th firefly is far from the optimal solution, the step size increases; conversely, the step size decreases. In the nth round, when the i-th firefly represents the optimal solution, its step size is 0. After examining the impact of the step size on the GSO algorithm, the SAS-GSO algorithm is derived, as follows:

[0120]

[0121] Where each x i (t) is exactly assigned to a s i (t), D i (t) represents any number in a uniform distribution, Nt x represents the maximum number of iterations. b (t) represents the position of the optimal firefly in round t. It is an exponentially decaying function, where e represents the base of the natural logarithm. As the number of iterations increases, the step size gradually decreases, but it will not be zero.

[0122] Firefly g with a commitment probability PT gj (t) from N g The move of the w neighbor starting from (t) is expressed as follows.

[0123]

[0124] Once the firefly moves, its position is updated using the following formula:

[0125]

[0126] S55. Update the neighborhood range after updating the node position. Dynamically adjust the decision radius of each firefly based on the density of neighboring fireflies. During iteration, if the density of neighboring fireflies is low, expand the decision radius to find more neighbors; if the density of neighboring fireflies is high, shrink the decision radius. The update formula is as follows:

[0127]

[0128] S56. Repeat steps S53-S55 until the preset number of iterations or convergence condition is reached.

[0129] S57. In the final firefly swarm, select the path with the highest fitness value as the optimal path.

[0130] Compared with existing methods, this method has the following advantages:

[0131] (1) This invention designs a metaheuristic-based clustering routing protocol for Underwater Wireless Sensor Networks (UWSNs). Due to the characteristics of underwater currents, low bandwidth, high water pressure, propagation delay, and error probability, traditional clustering routing protocols in wireless networks are not suitable for underwater wireless sensor networks. The clustering and routing process can be regarded as a nondeterministic polynomial-time (NP) difficult optimization problem, which can be solved using metaheuristic methods. The main objective of this clustering routing protocol is to select the cluster head (CH) and the optimal path to the destination node. The application of this technology helps to significantly improve the energy efficiency and lifetime of UWSNs.

[0132] (2) This invention is the first to use the hybrid optimization algorithm FGSF, which combines Fruit Fly Optimization (FFOA) and Firefly Optimization (GSO), in cluster head election of underwater wireless sensor networks. Although GSO can handle nonlinear and multimodal problems, it suffers from slow computation speed and poor local search capability. Meanwhile, the FFOA algorithm also suffers from slow convergence speed in the search space. Therefore, to overcome the problems of traditional algorithms, the proposed algorithm combines the best characteristics of both algorithms, enabling it to improve search capability while reducing negative search capability and improving convergence. FGSF integrates the concept of FFOA into GSO, achieving multiple objectives such as minimizing latency and maximizing energy based on the intra-frame distance and inter-frame distance from the cluster head to the node.

[0133] (3) This invention uses the SAS-GSO algorithm to select the best route. The SAS-GSO algorithm improves the problem of slow convergence speed caused by the use of fixed step size in the traditional GSO algorithm. It uses an adaptive step size and comprehensively considers parameters such as trust, energy, delay and distance to determine the optimal path for data transmission between the transmitter and the destination through the cluster head in the underwater wireless sensor network, which can greatly improve the overall network efficiency.

[0134] Example 2

[0135] This embodiment provides an electronic device, including: one or more processors and a memory, wherein the memory stores one or more programs, and the one or more programs include instructions for executing the node clustering routing method for underwater wireless sensor networks as described in Embodiment 1.

[0136] Example 3

[0137] This embodiment provides a computer-readable storage medium including one or more programs executable by one or more processors of an electronic device, the one or more programs including instructions for performing a node clustering routing method for an underwater wireless sensor network as described in Embodiment 1.

[0138] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present invention, and these modifications or substitutions should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A node clustering routing method for underwater wireless sensor networks, characterized in that, The underwater wireless sensor network includes a sink node, a base station, and multiple dynamic underwater nodes. The routing method is applied to the sink node and specifically includes the following steps: Establish a connection with the base station; Establish connections with the cluster head nodes corresponding to each sub-cluster. The process of determining the cluster head nodes is as follows: for each underwater node, multiple sub-clusters are formed by clustering. For each sub-cluster, the corresponding cluster head node is determined by a hybrid optimization algorithm of fruit flies and fireflies. A connection to the target underwater node is established based on the optimal path. A route is established from the base station to the target underwater node. The process of determining the optimal path is as follows: The target cluster where the target underwater node is located is determined; the optimal path from the cluster head node of the target cluster to the target underwater node is determined using an adaptive step-size firefly algorithm. The process of determining the corresponding cluster head node using a hybrid optimization algorithm combining fruit fly and firefly algorithms includes the following steps: Select one or more underwater nodes and initialize them as cluster head nodes; For each remaining underwater node, the corresponding first fitness function value is calculated based on the normalized distance between the underwater node and each cluster head node, the distance between the cluster head node and the base station, and the energy utilization rate of the broadcast data packet. The underwater nodes are sorted based on the first fitness function value. For underwater nodes whose serial number is less than or equal to a preset value, the corresponding first fitness function value is updated using the firefly optimization algorithm. For underwater nodes whose serial number is greater than or equal to the preset value, the corresponding first fitness function value is updated using the fruit fly algorithm. This step is repeated multiple times. The final cluster head node is determined based on the final first fitness function value. The process of determining the optimal path from the cluster head node to the target underwater node using the adaptive step-size firefly algorithm includes the following steps: Obtain the set of alternative routes from the cluster head node of the target cluster to the target underwater node, calculate the second fitness function value corresponding to each alternative path, and use the firefly algorithm for iteration. During the iteration process, determine the current firefly's movement step size based on the number of iterations and the distance between the current firefly and the optimal firefly. The optimal path is determined based on the final second fitness function value. The current movement step size of the firefly is obtained using the following formula: in, Indicates the first Firefly in round of iteration Location, Indicates the first The optimal firefly position in each iteration. Let represent any number that is uniformly distributed. Indicates the maximum number of iterations. It is an exponentially decaying function. Represents the natural logarithm. The second fitness function value is obtained using the following formula: in, Represents the fitness function. Indicates energy. Indicates a delay. Indicates distance, Indicates the level of trust.

2. The node clustering routing method for underwater wireless sensor networks according to claim 1, characterized in that, The first fitness function value is obtained using the following formula: in, , and These are weighting coefficients. It is the sum of the distances between the underwater node and the cluster head nodes of all sub-clusters. It is the distance between the cluster head node and the base station. This indicates the energy utilization rate of network broadcast data packets.

3. The node clustering routing method for underwater wireless sensor networks according to claim 1, characterized in that, The normalization process is implemented using the sigmoid function.

4. The node clustering routing method for underwater wireless sensor networks according to claim 1, characterized in that, During the iteration process, the distance between the current firefly and the best firefly is inversely proportional to the current firefly's step size.

5. An electronic device, characterized in that, include: One or more processors and a memory, wherein the memory stores one or more programs, the one or more programs including instructions for executing the node clustering routing method for underwater wireless sensor networks as described in any one of claims 1-4.

6. A computer-readable storage medium, characterized in that, Includes one or more programs that are executed by one or more processors of an electronic device, the one or more programs including instructions for executing the node clustering routing method for underwater wireless sensor networks as described in any one of claims 1-4.

Images