Method for multi-target ship communication path planning based on maritime waveguide channel map

Abstract

The application provides a kind of multi-target ship communication path planning method based on maritime waveguide channel map, it is related to wireless communication technical field, contain the following steps: in the case of long-distance transmission at sea, the single-input single-output ship motion model and communication model are established;CGM is established according to existing gas phase data, evaporation waveguide model and electromagnetic wave propagation model;An optimization problem model is proposed;NSGA-II-PSO-VS algorithm is proposed;The running results of NSGA-II algorithm and NSGA-II-PSO-VS algorithm under the same time are compared;Simulation experiments are carried out under different evaporation waveguide heights, which proves the rationality and applicability of the proposed algorithm. For data-driven multi-objective path planning problem based on channel map to obtain user information, the motion of the ship and the communication are modeled, an improved heuristic algorithm is proposed, which improves and combines the traditional two algorithms. Compared with the traditional algorithm, it shows good performance in many aspects, and provides help for data-driven multi-objective path planning problem.

Description

Technical Field

[0001] This invention relates to the field of wireless communication technology, and in particular to a multi-target ship communication path planning method based on a marine waveguide channel map. Background Technology

[0002] Evaporated waveguides are an atmospheric phenomenon where changes in atmospheric refractive index within the waveguide layer cause superrefractive effects in electromagnetic waves. This results in the waves reflecting back and forth between the waveguide layer and the sea surface, concentrating most of the wave energy, reducing path loss, and enhancing the received signal power. In ship-to-shore communication, evaporated waveguide communication offers the advantage of overcoming traditional line-of-sight limitations, enabling connections with distant coastal base stations and helping to reduce costs. As a key technology in environmental cognitive communication, the Channel Knowledge Graph (CKM) is constructed by fusing a large amount of historical data from all terminals in the area to reflect the characteristics of the local wireless environment. Prior environmental knowledge can be obtained in advance based on terminal location or virtual location information, avoiding repeated online environmental sensing and channel acquisition, thereby enabling rapid real-time prediction and inference of channel knowledge and improving communication and sensing performance. Based on the Channel Knowledge Graph (CKM), a Channel Gain Map (CGM) can be constructed, combining evaporated waveguide models, electromagnetic propagation models, and real-time meteorological data (such as weather radar and ocean buoys) to establish a marine evaporated waveguide channel map and perform route planning for ships to minimize transmission and arrival times.

[0003] Existing path planning methods focus on unmanned aerial vehicles (UAVs). These mainly include convex optimization methods, deep reinforcement learning methods, and heuristic algorithms. Convex optimization methods are the most common UAV path planning methods, including Successive Convex Approximation (SCA), Sequential Convex Optimization (SCP), and Block Coordinate Descent (BCD). These methods simplify the solution process by successively transforming a complex non-convex problem into a series of easier-to-solve convex optimization subproblems and performing alternating optimization operations. However, for some non-convex problems and constraints, transformation may be difficult, and the optimization problem needs to satisfy certain structural properties (such as convexity or blockiness). Deep reinforcement learning methods use deep learning networks to learn exploration and trajectory planning strategies, such as Deep Q-Networks (DQN). However, in complex problems, they require high-quality training data and have high maintenance costs. Heuristic algorithms include Genetic Algorithms (GA), Differential Evolutionary Algorithms (DE), and Ant Colony Algorithms (ACO). These methods can ignore the concavity and convexity of the optimization problem and are suitable for large-scale, nonlinear, non-convex, or discrete combinatorial optimization problems that are difficult for traditional algorithms to solve efficiently. However, in most cases, they only achieve suboptimal solutions. Summary of the Invention

[0004] The purpose of this invention is to solve the technical problem that algorithms in the prior art can only obtain suboptimal solutions in most cases.

[0005] To achieve the above objectives, the present invention adopts the following technical solution:

[0006] A multi-objective ship communication path planning method based on marine waveguide channel maps includes the following steps:

[0007] S1: In the case of long-distance transmission at sea, establish a single-input single-output ship motion model and communication model;

[0008] S2: Establish CGM based on existing gas phase data, evaporation waveguide model and electromagnetic wave propagation model;

[0009] S3: Propose an optimization problem model;

[0010] S4: Improve and combine the NSGA-II algorithm and the PSO algorithm to propose the NSGA-II-PSO-VS algorithm;

[0011] S5: Compare the running results of the NSGA-II algorithm and the NSGA-II-PSO-VS algorithm under the same time; conduct simulation experiments at different evaporation waveguide heights to prove the rationality and applicability of the proposed algorithm.

[0012] Preferably, the specific steps of S4 are as follows:

[0013] S4.1: Establish a population by using the direction angle of each time slot as a variable.

[0014] S4.2: Calculate the fitness function value of each individual in the initial population, and rank the individuals according to the non-dominated ranking and crowding degree;

[0015] S4.3: Perform selection, crossover, and mutation operations on the population, based on the crossover rate A. c and the rate of variation A m A certain number of crossover and mutated offspring are generated and merged with the original population to form a composite population Q.

[0016] S4.4: Perform non-dominated sorting on the composite population Q. Individuals of the same dominance level are sorted according to crowding distance. The sorted composite population is then filled into the next generation population according to the initial population size.

[0017] S4.5: When the algorithm iterates a certain number of times, it enters the modified particle swarm optimization algorithm. First, it uses some of the previous parameters and sets the specific parameters of the particle swarm optimization algorithm.

[0018] S4.6: Calculate the fitness function value of the population, and establish an external repository to store the location information and fitness value in the repository. Initialize the repository and retain only the individuals with the highest dominance level.

[0019] S4.7: Update the individual's speed and location information;

[0020] S4.8: Repeat this process until the maximum number of iterations is met.

[0021] Preferably, in step S4.1, when establishing the initial population, a certain number of randomly initialized individuals are first randomly generated, and then a certain number of individuals satisfying the constraints are generated according to the angle limit of each time slot. The two are combined into an initial population containing Np individuals.

[0022] Preferably, in S4.2, the i-th individual A is defined. i Distance of congestion i for:

[0023]

[0024] In the above formula, N obj Let f be the number of objective functions. The population is sorted according to the value of the j-th objective function. j (i+1) and f j (i-1) represent the objective function values ​​of the previous and next individuals for the i-th individual, respectively, f j (A max ) and f j (A min ) are the maximum and minimum values ​​of the j-th objective function, respectively.

[0025] Preferably, the individual selection method in S4.3 adopts a tournament selection approach. First, individuals with higher dominance levels are selected. When dominance levels are the same, individuals with lower crowding levels are selected based on crowding distance.

[0026] The crossover method is a simulated binary crossover, which obtains the next generation individual by crossing two parent individuals:

[0027]

[0028] In the above formula, and For two offspring individuals, and For two parent individuals, β represents...

[0029]

[0030] In the above formula, η c It is a custom crossover distribution parameter, where U is a random number between 0 and 1;

[0031] During the mutation process, polynomial mutation was employed, and the original method was modified as follows:

[0032]

[0033] Where A u A represents the upper limit of variation for an individual. l μ represents the lower limit of variation for an individual.

[0034]

[0035] In the above formula, η m For the variation distribution parameters, μ2 is and The smaller of the two values ​​is compared. Furthermore, Γ is represented as:

[0036]

[0037] Here, V represents the dimensionality of an individual. In the initial population, V = M. During subsequent iterations, V is made equal to the largest value in the population, M². Dimensions exceeding V are ignored during crossover and mutation processes. Therefore, the dimensionality of individuals in the population changes continuously with the number of iterations. Due to this dimensionality change mechanism, even when the dimensionality decreases... The value gradually increases;

[0038] As the number of iterations increases, the mutation probability also increases, thereby increasing the likelihood of escaping local optima. Before the mutation operation, considering that the ship's speed and angle will not undergo abrupt changes, a portion of the parent generations are randomly selected for a smoothing operation.

[0039] Preferably, the velocity and position information in S4.7 are defined as follows:

[0040]

[0041] A α+1 =A α +v α

[0042] A penalty function is set according to the constraints. A new fitness function is calculated, and if constraints are violated, the penalty function is added to the fitness function. The external repository is updated with new individuals, added to the repository, and a non-dominated sort is performed, removing dominated individuals from the repository. Then, the individuals in the updated repository are sorted according to crowding distance, and the portion with the largest crowding distance is selected as the historical best position. Candidate.

[0043] Preferably, the specific steps of S1 are as follows: The location of the base station is set to (0,0,0), a three-dimensional coordinate system is established with the base station as the origin, and the antenna location of the base station is set to (0,0,h). r ), where h r Let the height of the antenna be the center of gravity. Consider the ship's height as its motion in three-dimensional space. Therefore, let the ship start from its initial position (x0, y0, h0) and head towards its destination position (x0, y0, h0). d ,y d ,h d The position of the ship at any time t is:

[0044] X(t)=(x s (t),y s (t),h s (t)), 0≤t≤T

[0045] In the above formula, T represents the maximum sailing time, the ship's speed at any time t is denoted by v(t), and the direction of the speed is expressed as an angle. Therefore, the ship's speed in the x and y directions are expressed as follows:

[0046]

[0047] Assume the ship is at an initial orientation angle Starting from the initial position, its turning ability is limited by the following:

[0048]

[0049] In the above formula, δt is the time difference between the current time and the previous time. The maximum turning angle of the ship between these two moments depends on the magnitude of δt. At any moment t, the distance between the ship and the base station is expressed as:

[0050]

[0051] Let h denote the channel fading coefficient, and x represent the transmission channel. Therefore, the signal y received at the base station is represented as:

[0052] y = hx + n

[0053] Where n is the noise vector, and the channel fading coefficient h has a large-scale fading coefficient h L and small-scale fading coefficient h S Composition, represented as:

[0054] h = h L h S

[0055] Ignoring small-scale fading, we only consider large-scale fading; assume the transmit power is P. t Then the received power P r It can be represented as:

[0056] P r =|h L | 2 P t =L p G t G r P t

[0057] In the above formula, L p The path loss is determined by the distance d(t) between the base station and the ship, G. t For the transmit antenna gain, G r For the receiving antenna gain, the path loss L in free space is... p Defined as:

[0058] L p =L fs = 32.45 + 20lg(d(t)) + 20lg(f)

[0059] In the above formula, the distance d(t) is in kilometers, and f is the carrier frequency in MHz. Under the influence of evaporating waveguides, the path loss at sea exhibits different characteristics. Using the parabolic equation method, the propagation of electromagnetic waves in the evaporating waveguide environment is modeled, and its path loss L... p It can be represented as

[0060]

[0061] In the above formula, u represents the field strength, r represents the Earth's radius, and λ represents the carrier wavelength. Since the transmission distance is usually very small relative to the Earth's radius, the above formula can be simplified to:

[0062] L ed =-20lg|u|+20lg(4π)+10lg(d(t))-30lg(λ)

[0063] Therefore, the channel capacity is

[0064]

[0065] In the above formula, B is the channel width and n0 is the noise power spectral density. Therefore, at any time t, the maximum achievable transmission rate R(t) can be expressed as:

[0066]

[0067] Preferably, in step S2, the parabolic equation method is used to construct the CGM in the evaporating waveguide environment. Since the position of the ship at any time t is X(t), the large-scale channel gain at that position is defined as...

[0068] The space is divided into many grids, each with a width of Δd. The channel state remains unchanged within the grid. The chosen Δd should be of an appropriate size to ensure that the channel gain difference between adjacent grids is not too large. The vertical grid height Δh should be smaller than the horizontal grid width Δd.

[0069] Assume the farthest transmission distance from the base station is N. le Δd, where N le The integer is given, the obtained CGM is symmetric about the base station center, and the height of the evaporating waveguide is h. eva Then h eva =N ve Δh, where N ve It is also an integer, and the coordinates of each network can be represented as:

[0070] s(i,j,k)=(iΔd,jΔd,kΔh),i∈I N ,j∈J N ,k∈K N

[0071] In the above formula, I N ={-N le ,-N le +1,…,N le -1,N le},I N With J N Same, K N ={0,1,…,N ve -1,N ve The channel gain at any position s(i,j,k) is} Equal to received power P r L in the formula p Therefore, based on the grid where the ship is located at any given time, the large-scale channel gain between the base station and the ship can be obtained.

[0072] Preferably, the objective in S3 is to optimize the ship's trajectory to minimize its transmission and arrival times. The minimum transmission time is defined as T1, and the minimum arrival time as T2. The ship's trajectory is represented as K = {X(t), t ∈ [0, T2]}, and the optimization problem is expressed as...

[0073] P1):

[0074] stC1:T1≤T2

[0075] C2:T2≤T

[0076] C3:

[0077] C4: X(0) = (x0, y0, h0)

[0078] C5: X(T2)=(x d ,y d ,h d )

[0079] C6:

[0080] In the above formula, C1 means that the ship must transmit all data before reaching its destination, and the transmission time is less than the arrival time.

[0081] C2 indicates that the ship's sailing time must be less than the specified time;

[0082] C3 indicates that the amount of data to be transmitted within the transmission time must be greater than or equal to the amount of data to be transmitted;

[0083] C4 represents the starting constraint;

[0084] C5 represents the endpoint constraint;

[0085] C6 indicates the ship's steering constraint;

[0086] Constraints C3 and C6 are difficult to express in continuous time; therefore, the entire voyage is decomposed into multiple time slots, each with a length of Δt. The number of time slots used for data transmission is M1, the number of time slots required to reach the destination is M2, and the number of time slots for the maximum voyage time is... The ship's trajectory is K = {X} s (0),X s (1), ..., X s (M2)}, the formulas for the ship's position at any time t and the formulas for the limitations on the ship's turning ability are modified as follows:

[0087] X(i)=(x s (i),y s (i),h s (i)), 0≤i≤M2

[0088]

[0089] Assuming the ship's speed is constant and ignoring the effects of wave fluctuations, the displacement within a single time slot is expressed as:

[0090] ||X(i+1)-X(i)||=vΔt, 0≤i≤M2-1

[0091] The optimization problem (P1) is rewritten as:

[0092] P2):

[0093] stC1:M1≤M2

[0094] C2:M2≤M

[0095] C3:

[0096] C4: X(0) = (x0, d0, h0)

[0097] C5: X(M2)=(x d ,y d ,h d )

[0098] C6:

[0099] Preferably, after completing S4, the running results of the NSGA-II algorithm and the NSGA-II-PSO-VS algorithm at the same time are compared; simulation experiments are conducted at different evaporation waveguide heights to prove the rationality and applicability of the proposed algorithm.

[0100] The aforementioned multi-objective propagation path planning method based on marine evaporating waveguide channel maps, for data-driven multi-objective path planning problems that obtain user information based on channel maps, models the motion and communication of ships and proposes an improved heuristic algorithm. It improves and combines two traditional intelligent algorithms and shows good performance in many aspects compared with traditional algorithms, thus providing assistance for data-driven multi-objective path planning problems.

[0101] Compared with existing technologies, the present invention has the following advantages:

[0102] Overall, this invention employs a heuristic algorithm, NSGA-II-PSO-VS, which is an improvement upon the NSGA-II algorithm. It provides a solution for data-driven ship trajectory planning. This algorithm demonstrates superior performance compared to existing algorithms in terms of convergence speed, number of solutions, and solution range. Attached Figure Description

[0103] Figure 1 This is a system block diagram of a multi-target ship communication path planning method based on a marine waveguide channel map according to an embodiment of the present invention;

[0104] Figure 2This is a channel gain map constructed in a multi-objective ship communication path planning method based on a marine waveguide channel map according to one embodiment of the present invention.

[0105] Figure 3 This is a comparison chart of the results of the NSGA-II algorithm and the proposed NSGA-II-PSO-VS algorithm in one embodiment of the present invention;

[0106] Figure 4 This is a trajectory diagram of the optimal point at different waveguide heights in one embodiment of the present invention. The optimal point is the point with the same arrival time and transmission time weights, that is, the point closest to the origin. Detailed Implementation

[0107] The present invention will be further described in detail below with reference to specific embodiments.

[0108] Please see Figure 1 A multi-objective ship communication path planning method based on marine waveguide channel maps includes the following steps:

[0109] S1: In the case of long-distance transmission at sea, establish a single-input single-output ship motion model and communication model;

[0110] Specifically, in one embodiment, the location of the base station is set to (0,0,0), a three-dimensional coordinate system is established with the base station as the origin, and the location of the base station's antenna is set to (0,0,h). r ), where h r Let the height of the antenna be the center of gravity. Consider the ship's height as its motion in three-dimensional space. Therefore, let the ship start from its initial position (x0, y0, h0) and head towards its destination position (x0, y0, h0). d ,y d ,h d The position of the ship at any time t is:

[0111] X(t)=(x s (t),y s (t),h s (t)), 0≤t≤T (1)

[0112] In the above formula, T represents the maximum sailing time, the ship's speed at any time t is denoted by v(t), and the direction of the speed is expressed as an angle. Therefore, the ship's speed in the x and y directions are expressed as follows:

[0113]

[0114] Due to factors such as a ship's size, draft, and speed, its maneuverability is limited when turning. Assuming the ship starts at an initial azimuth angle... Starting from the initial position, its turning ability is limited by the following:

[0115]

[0116] In the above formula, δt is the time difference between the current time and the previous time. The maximum turning angle of the ship between these two moments depends on the magnitude of δt. At any moment t, the distance between the ship and the base station is expressed as:

[0117]

[0118] Since the communication system is a single-input single-output system, and h represents the channel fading coefficient, and x represents the transmission channel, the signal y received at the base station is represented as:

[0119] y = hx + n (6)

[0120] Where n is the noise vector, and the channel fading coefficient h has a large-scale fading coefficient h L and small-scale fading coefficient h S Composition, represented as:

[0121] h = h L h S (7)

[0122] In the case of long-distance transmission over vast oceans and single-antenna systems, the impact of small-scale fading is negligible compared to large-scale fading; therefore, small-scale fading is ignored, and only large-scale fading is considered. Assume the transmit power is P. t Then the received power P r It can be represented as:

[0123] P r =|h L | 2 P t =L p G t G r P t (8)

[0124] In the above formula, L p The path loss is determined by the distance d(t) between the base station and the ship, G. t For the transmit antenna gain, G r For the receiving antenna gain, the path loss L in free space is... p Defined as:

[0125] L p =L fs =32.45+20lg(d(t))+20lg(f) (9)

[0126] In the above formula, the distance d(t) is in kilometers, and f is the carrier frequency in MHz. Under the influence of evaporating waveguides, the path loss at sea exhibits different characteristics. Using the parabolic equation method, the propagation of electromagnetic waves in the evaporating waveguide environment is modeled, and its path loss L... p It can be represented as

[0127]

[0128] In the above formula, u represents the field strength, r represents the Earth's radius, and λ represents the carrier wavelength. Since the transmission distance is usually very small relative to the Earth's radius, the above formula can be simplified to:

[0129] L ed =-20lg|u|+20lg(4π+10lg(d(t))-30lg(λ) (11)

[0130] Therefore, the channel capacity is

[0131]

[0132] In the above formula, B is the channel width and n0 is the noise power spectral density. Therefore, at any time t, the maximum achievable transmission rate R(t) can be expressed as:

[0133]

[0134] S2: Establish the CGM based on existing gas phase data, evaporation waveguide model, and electromagnetic wave propagation model; for details, please refer to [link to relevant documentation]. Figure 2 In one embodiment, the parabolic equation method is used to construct the CGM in an evaporating waveguide environment. Since the position of the ship at any time t is X(t), the large-scale channel gain at that position is defined as...

[0135] Due to limitations in spatial continuity and storage capacity, it is impossible to store large-scale channel gains at all locations. Therefore, it is necessary to divide the space into many grids, each with a width of Δd, within which the channel state remains constant. Since there are no obstacles in the long-distance transmission over sea in an evaporating waveguide environment, the large-scale channel gain in the horizontal direction will not change drastically over short distances. Therefore, the chosen Δd should be of an appropriate size to ensure that the channel gain difference between adjacent grids is not too large.

[0136] Compared to long-distance horizontal transmission, the variation in transmit and receive antenna heights significantly impacts large-scale channel gain, as typical marine evaporative waveguides range from 10 to 40 meters in height. Therefore, the vertical grid height Δh is smaller than the horizontal grid width Δd.

[0137] Assume the farthest transmission distance from the base station is N. le Δd, where N le The integer is given, the obtained CGM is symmetric about the base station center, and the height of the evaporating waveguide is h. eva Then h eva =N ve Δh, where N ve It is also an integer, and the coordinates of each network can be represented as:

[0138] s(i,j,k)=(iΔd,jΔd,kΔh),i∈I N ,j∈J N ,k∈K N (14)

[0139] In the above formula, I N ={-N le ,-N le +1,…,N le -1,N le},I N With J N Same, K N ={0,1,…,N ve -1,N ve The channel gain at any position s(i,j,k) is} equals L in formula (8) p Therefore, based on the grid where the ship is located at any given time, the large-scale channel gain between the base station and the ship can be obtained.

[0140] S3: Propose an optimization problem model;

[0141] Our goal is to optimize the ship's trajectory to minimize its transit and arrival times. Let the minimum transit time be defined as T1, the minimum arrival time as T2, and the ship's trajectory as K = {X(t), t ∈ [0, T2]}. The optimization problem is expressed as:

[0142] (P1):

[0143] stC1:T1≤T2

[0144] C2:T2≤T

[0145] C3:

[0146] C4: X(0) = (x0, y0, h0)

[0147] C5: X(T2)=(x d ,y d ,h d )

[0148] C6:

[0149] In the above formula, C1 means that the ship must transmit all data before reaching its destination, and the transmission time is less than the arrival time.

[0150] C2 indicates that the ship's sailing time must be less than the specified time;

[0151] C3 indicates that the amount of data to be transmitted within the transmission time must be greater than or equal to the amount of data to be transmitted;

[0152] C4 represents the starting constraint;

[0153] C5 represents the endpoint constraint;

[0154] C6 indicates the ship's steering constraint;

[0155] Constraints C3 and C6 are difficult to express in continuous time; therefore, the entire voyage is decomposed into multiple time slots. Each time slot has a length of Δt, M1 is used for data transmission, M2 is required to reach the destination, and the maximum voyage time is determined by the number of time slots required. The ship's trajectory is K = {X} s (0),X s (1), ..., X s (M2)}, modify formulas (1) and (4) as follows:

[0156] X(i)=(x s (i),y s (i),h s (i)), 0≤i≤M2 (16)

[0157]

[0158] Furthermore, the ship's trajectory approximates a straight line within a single time slot, where the angular direction of the velocity remains constant. To simplify the problem, we assume the ship's velocity to be uniform and ignore the effects of wave undulation; therefore, the displacement within a single time slot is expressed as:

[0159] ||X(i+1)-X(i)||=vΔt, 0≤i≤M2-1 (18)

[0160] The optimization problem (P1) is rewritten as:

[0161] P2):

[0162] stC1:M1≤M2

[0163] C2:M2≤M

[0164] C3:

[0165] C4: X(0) = (x0, d0, h0)

[0166] C5: X(M2)=(x d ,y d ,h d )

[0167] C6:

[0168] S4: Improve and combine the NSGA-II algorithm and the PSO algorithm to propose the NSGA-II-PSO-VS algorithm;

[0169] In one embodiment, S4 includes the following steps:

[0170] S4.1: Establish a population by using the direction angle of each time slot as a variable. When establishing the initial population, first randomly generate a certain number of initial individuals, then generate a certain number of individuals that satisfy the constraints based on the angle of each time slot. Combine the two into a population containing N individuals. p An initial population of a certain number.

[0171] S4.2: Calculate the fitness function value of each individual in the initial population, and rank the individuals according to the non-dominated ranking and crowding degree. The i-th individual A... i Distance of congestion i for:

[0172]

[0173] In the above formula, N obj Let f be the number of objective functions. The population is sorted according to the value of the j-th objective function. j (i+1) and f j (i-1) represent the objective function values ​​of the previous and next individuals for the i-th individual, respectively, f j (A max ) and f j (A max ) are the maximum and minimum values ​​of the j-th objective function, respectively.

[0174] S4.3: Perform selection, crossover, and mutation operations on the population, based on the crossover rate A. cand the rate of variation A m A certain number of crossover and mutation offspring are generated and merged with the original population to form a combined population Q.

[0175] Individual selection adopts a tournament selection method. First, individuals with higher dominance levels are selected. When dominance levels are the same, individuals with lower crowding levels are selected based on crowding distance.

[0176] The crossover method is simulated binary crossover (SBX), which obtains the next generation individual by crossing two parent individuals:

[0177]

[0178] In the above formula, and It consists of two offspring individuals. and For two parent individuals, β can be represented as

[0179]

[0180] In the above formula, η c The crossover distribution parameter is a user-defined parameter, where U is a random number between 0 and 1. During the mutation process, multinomial mutation is used, and the original method has been modified:

[0181]

[0182] Where A u A represents the upper limit of variation for an individual. l μ represents the lower limit of variation for an individual.

[0183]

[0184] In the above formula, η m For the variation distribution parameters, μ2 is and The smaller of the two values ​​is compared. Furthermore, Γ is represented as:

[0185]

[0186] Here, V represents the dimensionality of an individual. In the initial population, V = M. During subsequent iterations, V is made equal to the largest value in the population, M². Dimensions exceeding V are ignored during crossover and mutation processes. Therefore, the dimensionality of individuals in the population changes continuously with the number of iterations. Due to this dimensionality change mechanism, when the dimensionality decreases... The value gradually increases. As the number of iterations increases, the mutation probability also increases, thus increasing the likelihood of escaping local optima. Before the mutation operation, considering that the ship's speed and angle will not undergo abrupt changes, a portion of the parent generations is randomly selected for a smoothing operation, i.e.

[0187] S4.4: Perform a non-dominated sort on the composite population Q, sorting individuals of the same dominance level according to crowding distance. The sorted composite population is then used to fill the next generation population based on the initial population size.

[0188] S4.5: When the algorithm has iterated a certain number of times, it enters the modified particle swarm optimization algorithm. First, it uses some of the previous parameters and sets the specific parameters of the particle swarm optimization algorithm.

[0189] S4.6: Calculate the fitness function value of the population and establish an external repository. Store the location information and fitness values ​​in the repository and initialize the repository, retaining only individuals with the highest dominance level.

[0190] S4.7: Update individual speed and location information

[0191]

[0192] A α+1 =A α +v α (27)

[0193] A penalty function is set according to the constraints. A new fitness function is calculated, and if constraints are violated, the penalty function is added to the fitness function. The external repository is updated with new individuals, added to the repository, and a non-dominated sort is performed, removing dominated individuals from the repository. Then, the individuals in the updated repository are sorted according to crowding distance, and the subset with the largest crowding distance is selected as the historical best position. Candidate.

[0194] S4.8: Repeat this process until the maximum number of iterations is met.

[0195] S5: Compare the running results of the NSGA-II algorithm and the NSGA-II-PSO-VS algorithm under the same time; conduct simulation experiments at different evaporation waveguide heights to prove the rationality and applicability of the proposed algorithm.

[0196] Please see Figure 3 and Figure 4In step S5, the NSGA-II algorithm and the proposed NSGA-II-PSO-VS algorithm are run for the same amount of time under the same parameters, with four cases selected: evaporation waveguide heights of 20m, 25m, 30m, and 35m, and the results are compared.

[0197] Next, simulation experiments were conducted again at these four waveguide heights. Since the obtained solution is an optimal set, the transmission time and arrival time were given the same weights, the optimal point was selected, and the trajectory diagram was obtained.

[0198] In summary, this application discloses a multi-objective propagation path planning method based on marine evaporating waveguide channel maps. For data-driven multi-objective path planning problems that obtain user information based on channel maps, it models the motion and communication of ships and proposes an improved heuristic algorithm. It improves and combines two traditional intelligent algorithms and shows good performance in many aspects compared with traditional algorithms, thus providing assistance for data-driven multi-objective path planning problems.

Claims

1. A multi-objective ship communication path planning method based on marine waveguide channel maps, characterized in that: Includes the following steps: S1: In the case of long-distance transmission at sea, establish a single-input single-output ship motion model and communication model; S2: Establish CGM based on existing gas phase data, evaporation waveguide model and electromagnetic wave propagation model; S3: Propose an optimization problem model; S4: Improve and combine the NSGA-II algorithm and the PSO algorithm to propose the NSGA-II-PSO-VS algorithm; The specific steps of S4 are as follows: S4.1: Establish a population by using the direction angle of each time slot as a variable. S4.2: Calculate the fitness function value of each individual in the initial population, and rank the individuals according to the non-dominated ranking and crowding degree; S4.3: Perform selection, crossover, and mutation operations on the population, based on the crossover rate A. c and the rate of variation A m A certain number of crossover and mutated offspring are generated and merged with the original population to form a composite population Q. S4.4: Perform non-dominated sorting on the composite population Q. Individuals of the same dominance level are sorted according to crowding distance. The sorted composite population is then filled into the next generation population according to the initial population size. S4.5: When the algorithm iterates a certain number of times, it enters the modified particle swarm optimization algorithm. First, it uses some of the previous parameters and sets the specific parameters of the particle swarm optimization algorithm. S4.6: Calculate the fitness function value of the population, and establish an external repository to store the location information and fitness value in the repository. Initialize the repository and retain only the individuals with the highest dominance level. S4.7: Update the individual's speed and location information; S4.8: Repeat this process until the maximum number of iterations is met; S5: Compare the running results of the NSGA-II algorithm and the NSGA-II-PSO-VS algorithm under the same time; conduct simulation experiments at different evaporation waveguide heights to prove the rationality and applicability of the proposed algorithm.

2. The multi-objective ship communication path planning method based on marine waveguide channel maps according to claim 1, characterized in that: In S4.1, when establishing the initial population, a certain number of randomly initialized individuals are first generated, and then a certain number of individuals that meet the constraints are generated according to the angle limit of each time slot. The two are combined into an initial population containing Np individuals.

3. The multi-objective ship communication path planning method based on marine waveguide channel maps according to claim 2, characterized in that: In S4.2, the i-th individual A is defined. i Distance of congestion i for: ， In the above formula, N obj Let f be the number of objective functions. The population is sorted according to the value of the j-th objective function. j (i+1) and f j (i-1) represent the objective function values ​​of the preceding and following individuals for the i-th individual, respectively, f j (A max ) and f j (A min ) are the maximum and minimum values ​​of the j-th objective function, respectively.

4. The multi-objective ship communication path planning method based on marine waveguide channel maps according to claim 3, characterized in that: In S4.3, the individual selection method adopts a tournament selection approach. First, individuals with higher dominance levels are selected. When dominance levels are the same, individuals with lower crowding levels are selected based on crowding distance. The crossover method is a simulated binary crossover, which obtains the next generation individual by crossing two parent individuals: ， ， ， Represented as ， In the above formula, η c It is a custom crossover distribution parameter, where U is a random number between 0 and 1; During the mutation process, polynomial mutation was employed, and the original method was modified as follows: ， Where A u A represents the upper limit of variation for an individual. l μ represents the lower limit of variation for an individual. ， In the above formula, η m For the variation distribution parameters, ， μ2 is and The smaller of the two values ​​is compared. Furthermore, Г is represented as: ， Here, V represents the dimensionality of an individual. In the initial population, V = M. During subsequent iterations, V is made equal to the largest value in the population, M². Dimensions exceeding V are ignored during crossover and mutation. Therefore, the dimensionality of individuals in the population changes continuously with the number of iterations. Due to this dimensionality change mechanism, even when the dimensionality decreases... The value gradually increases; As the number of iterations increases, the mutation probability also increases, thereby increasing the likelihood of escaping local optima. Before the mutation operation, considering that the ship's speed and angle will not undergo abrupt changes, a portion of the parent generations are randomly selected for a smoothing operation. 。 5. The multi-objective ship communication path planning method based on marine waveguide channel maps according to claim 4, characterized in that: The velocity and position information in S4.7 are defined as follows: ， ， The process involves setting a penalty function based on constraints; calculating a new fitness function, and adding the penalty function to the fitness function if constraints are violated; updating the external repository with new individuals, adding them to the repository, and performing a non-dominated sort to remove dominated individuals from the repository; then, sorting the individuals in the updated repository according to crowding distance, and selecting the portion with the largest crowding distance as the historical best position. Candidate.

6. The multi-objective ship communication path planning method based on marine waveguide channel maps according to claim 5, characterized in that: The specific steps of S1 are as follows: Set the location of the base station to (0,0,0), establish a three-dimensional coordinate system with the base station as the origin, and set the antenna location of the base station to (0,0,h). r ), where h r Let the height of the antenna be the altitude of the ship. Considering the ship's altitude as a point of motion in three-dimensional space, let the ship start from its initial position (x0, y0, h0) and head towards its destination position (x0, y0, h0). d ,y d ,h d The position of the ship at any time t is: ， In the above formula, T represents the maximum sailing time, the ship's speed at any time t is denoted by v(t), and the angle of the speed direction is represented by φ(t). Therefore, the ship's speed in the x and y directions are expressed as follows: ， ， Assuming the ship departs from its initial position with an initial directional angle φ(0), its turning ability is limited as follows: ， In the above formula, δt is the time difference between the current time and the previous time, and δφ is the maximum turning angle of the ship between these two times, the magnitude of which depends on the magnitude of δt. At any time t, the distance between the ship and the base station is expressed as: ； Let h denote the channel fading coefficient, and x represent the transmission channel. Therefore, the signal y received at the base station is represented as: ， Where n is the noise vector, and the channel fading coefficient h has a large-scale fading coefficient h L and small-scale fading coefficient h S Composition, represented as: ， Ignoring small-scale fading, we only consider large-scale fading; assume the transmit power is P. t Then the received power P r Represented as: ， In the above formula, L p The path loss is determined by the distance d(t) between the base station and the ship, G. t For the transmit antenna gain, G r For the receiving antenna gain, the path loss L in free space is... p Defined as: ， In the above formula, the distance d(t) is in kilometers, and f is the carrier frequency in MHz. Under the influence of evaporating waveguides, the path loss at sea exhibits different characteristics. Using the parabolic equation method, the propagation of electromagnetic waves in the evaporating waveguide environment is modeled, and its path loss L... p Represented as ， In the above formula, u represents the field strength, r represents the Earth's radius, and λ represents the carrier wavelength. Since the transmission distance is usually very small relative to the Earth's radius, the above formula simplifies to: ， Therefore, the channel capacity is, ， In the above formula, B is the channel width and n0 is the noise power spectral density. Therefore, at any time t, the maximum achievable transmission rate R(t) is expressed as: 。 7. The multi-objective ship communication path planning method based on marine waveguide channel maps according to claim 6, characterized in that: In S2, the parabolic equation method is used to construct the CGM in an evaporating waveguide environment. Since the position of the ship at any time t is X(t), the large-scale channel gain at that position is defined as... , The space is divided into many grids, each with a width of Δd. The channel state remains unchanged within the grid. The chosen Δd should be of an appropriate size to ensure that the channel gain difference between adjacent grids is not too large. The vertical grid height Δh should be smaller than the horizontal grid width Δd. Assume the farthest transmission distance from the base station is N. le Δd, where N le The integer is given, the obtained CGM is symmetric about the base station center, and the height of the evaporating waveguide is h. eva ,but Where, N ve It is also an integer, and the coordinates of each network are represented as: ， In the above formula, I N With J N same, any position The channel gain at that point is , equal to the received power P r L in the formula p Therefore, based on the grid where the ship is located at any given time, the large-scale channel gain between the base station and the ship can be obtained.

8. The multi-objective ship communication path planning method based on marine waveguide channel maps according to claim 7, characterized in that: In S3, the objective is to optimize the ship's trajectory to minimize its transmission and arrival times. The minimum transmission time is defined as T1, and the minimum arrival time as T2. The ship's trajectory is represented as follows: ， The optimization problem is represented as follows: ， ， ， ， ， ， ； In the above formula, C1 means that the ship must transmit all data before reaching its destination, and the transmission time is less than the arrival time. C2 indicates that the ship's sailing time must be less than the specified time; C3 indicates that the amount of data to be transmitted within the transmission time must be greater than or equal to the amount of data to be transmitted; C4 represents the starting constraint; C5 represents the endpoint constraint; C6 indicates the ship's steering constraint; Constraints C3 and C6 are difficult to express in the context of continuous time; therefore, the entire navigation process is decomposed into multiple time slots, each with a length of Δt. The number of time slots used for data transmission is M1, the number of time slots required to reach the destination is M2, and the number of time slots for the maximum flight time is [missing information]. , The ship's trajectory is ， The formulas for the ship's position at any time t and the formulas for the limitations on the ship's turning ability are modified as follows: ， ， Assuming the ship's speed is constant and ignoring the effects of wave fluctuations, the displacement within a single time slot is expressed as: ， The optimization problem (P1) is rewritten as: ， ， ， ， ， ， 。

Images