A resource allocation method for sea-air-land heterogeneous links

Abstract

The application discloses a resource allocation method of sea-air-land heterogeneous link. According to data collection task requirements and node performance, the application establishes a heterogeneous link resource allocation optimization problem, which is decoupled into a water radio link sub-problem and an underwater acoustic link sub-problem; the deployment position of a UAV in the water radio link sub-problem is scheduled; the NOMA scheme is used in the communication between the water surface node and the UAV node by the UAV to optimize the communication time and the communication power; the UAV transmission power is optimized to obtain the optimal communication time of the water link; the underwater link optimization sub-problem is solved by using the water link communication time, the underwater acoustic channel allocation is associated with the node transmission power, and the residual energy of the node is taken as a weight value to solve the maximum weight matching problem to obtain an underwater link channel allocation optimization scheme. The resource allocation of the whole heterogeneous link is completed in combination with the water and underwater link optimization solution. The method optimizes the communication resource allocation in the sea-air-land heterogeneous link and prolongs the survival time of the link.

Description

Technical Field

[0001] This invention belongs to the field of wireless network technology, especially the field of heterogeneous network link technology, and specifically relates to a resource allocation method for sea, air and land heterogeneous links. Background Technology

[0002] As a crucial component of the integrated air-space-ground-sea network, the development of this network is of great significance. It can be deployed in various application scenarios, including marine resource exploration, maritime security, and disaster early warning. Reliable and efficient air-space-sea communication links can more effectively collect marine data and facilitate the rational development of marine resources. The openness of the ocean and the unique characteristics of its environment determine that the maritime security situation is exceptionally complex and severe; a well-designed integrated air-space-sea network can better serve the nation's needs for ensuring maritime security. Faster marine data collection and processing enable more rapid and accurate detection of marine disasters, timely early warning, and protection of national and public property and personal safety.

[0003] The integrated sea-air-land network contains heterogeneous links: the above-water portion transmits information via radio frequency communication, while the underwater portion transmits data via various underwater wireless communications. Within a complete data collection cycle, data collected by underwater sensors can be transmitted through these heterogeneous links to a land-based data center for subsequent analysis and research. The operation of these heterogeneous links serving underwater data collection requires the scheduling of resources across multiple links. Considering a complete data collection cycle: underwater nodes that have collected data select suitable surface aggregation nodes and available underwater acoustic channels to transmit the data; surface nodes, after collecting acoustic signals, convert them into radio frequency signals and transmit them to UAV nodes as quickly as possible; UAV nodes should coordinate with the distribution of surface nodes to select suitable deployment locations for better data reception, and they also need to consider the location of the data center for better data forwarding. Data transmission between nodes involves the scheduling of link resources; rational resource scheduling can better accomplish underwater data collection tasks and has practical significance. Summary of the Invention

[0004] The purpose of this invention is to provide a resource allocation method for heterogeneous links in a sea-air-land integrated network to meet the task requirements of collecting marine data.

[0005] The heterogeneous sea-air-land link consists of multiple underwater sensor nodes, multiple surface convergence nodes, an aerial drone node, and a land-based data center; the surface portion of the heterogeneous link uses radio frequency communication, while the underwater portion uses acoustic communication.

[0006] The specific allocation method is as follows:

[0007] Step (1) Determine the data collection time T and the data transmission size based on the task. Combined with the communication power limitations of the underwater sensor node USN, the surface convergence node SN, and the aerial UAV node UAV, establish a heterogeneous link resource allocation optimization problem: optimize the underwater channel allocation index set. RF channel allocation index set USN transmit power set SN transmit power set UAV transmit power set USN to SN data transmission time set SN to UAV data transmission time set UAV to ground-based data center GBS data transmission time set And the location χ of the drone, to maximize the minimum remaining energy of all USNs in each data collection cycle. in:

[0008] a m,k,n This is an identifier for whether the nth underwater channel is used when the m-th USN node sends data to the k-th SN node, a m,k,n ∈{0,1}, a m,k,n =1 means yes, a m,k,n =0 indicates no; b k,l This is an indicator for whether the l-th radio frequency channel is selected when the k-th SN node sends data to the UAV node, b k,l ∈{0,1},b k,l =1 indicates yes, b k,l =0 indicates no; m=1,…,M, where M is the number of USN nodes; k=1,…,K, where K is the number of SN nodes; n=1,…,N, where N is the number of underwater channels; l=1,…,L, where L is the number of radio frequency channels;

[0009] p m p represents the transmit power of the m-th USN node. k p represents the transmit power of the k-th SN node. uav This indicates the transmit power of the UAV node;

[0010] T m,k This represents the data transmission time when the m-th USN node sends data to the k-th SN node. T represents the path propagation time for the m-th USN node to send data to the k-th SN node; k This indicates the data transmission time when the k-th SN node sends data to the UAV node. T represents the path propagation time for the k-th SN node to send data to the UAV node; UAVThis indicates the data transmission time when the UAV node sends data to the land-based data center. This indicates the path propagation time for data sent from a UAV node to a land-based data center;

[0011] This represents the remaining energy of the m-th USN node after the s-th data collection cycle.

[0012] The specific constraints of the resource allocation optimization problem include:

[0013] Constraint a: For any USN node and any SN node, For any SN node and any underwater acoustic channel M k This is the set of USN nodes that send data to the k-th SN.

[0014] Constraint b: For any radio frequency channel, C max This represents the maximum number of SN nodes allowed to reuse the same radio frequency subchannel in a non-orthogonal frequency division multiple access (NOMA) scheme; for any SN node,

[0015] Constraint c: For any USN node, This represents the maximum transmit power of the USN node; for any SN node, This represents the maximum transmit power of the SN node; for the UAV node, This indicates the maximum transmit power of the UAV node;

[0016] Constraint d: This represents the maximum time required for the k-th SN node and UAV node to complete data transmission.

[0017] Constraint e: For any USN node and any SN node, r m,k,n R represents the rate at which the m-th USN node sends data to the k-th SN node. m,k Indicates the amount of data sent; for any SN node, r k,l R represents the rate at which the k-th SN node sends data to the UAV node. k Indicates the amount of data sent; r UAV η represents the rate at which the UAV node sends data to the ground-based data center GBS, and η represents the rate of change of the total data collected by the UAV node from the SN node.

[0018] Constraint 'a' indicates that in underwater transmission, USNs in the same cluster occupy different underwater acoustic channels, and each USN can only occupy one channel.

[0019] Constraint b indicates that in data transmission from SN to UAV, the subchannel can be composed of a maximum of C. max Each SN is shared, and each SN can only be assigned one radio frequency sub-channel.

[0020] Constraint c ensures that the transmit power of the USN, SN, and UAV does not exceed the maximum allowable value.

[0021] Constraint d ensures that the data collection task is completed within a defined data collection time T.

[0022] Constraint e ensures that all data can be successfully sent to the corresponding receiving node.

[0023] Step (2) will The problem is decoupled into an above-water radio frequency link sub-problem and an underwater acoustic link sub-problem.

[0024] The surface radio frequency link consists of two segments: from the surface aggregation node to the UAV node and from the UAV node to the land-based data center. The sub-problem of the surface radio frequency link is to optimize the communication time of the link by changing the location of the UAV node, thereby minimizing the data transmission time used by the surface radio frequency link. Specific constraints include: and Specifically:

[0025] (2-1) Determine the deployment location of the UAV nodes:

[0026] The deployment location of the drone after the s-th data collection cycle: d k,UAV This represents the distance from the k-th SN node to the UAV node. Let represent the remaining energy of the k-th SN node after the s-th data collection cycle, (·). θ This indicates the weight of the remaining energy's influence.

[0027] When deploying drones, the remaining energy of each surface node should be considered, and the design principle should be to balance the remaining energy of all SN nodes in each data collection cycle. Since SN nodes farther from the drone typically consume more energy than other nodes when transmitting the same amount of data, the drone should be positioned closer to SN nodes with less remaining energy.

[0028] The horizontal deployment location of the UAV is determined by a gradient descent algorithm starting from the centroid of the polygon.

[0029] This paper analyzes the UAV positioning problem by referencing the Fermat point of a polygon in mathematics. The Fermat point of a polygon is a point selected inside or outside a polygon such that the sum of the distances from that point to each corner of the polygon is minimized. The UAV position is calculated using weighted polygon Fermat point calculations, substituting the residual energy of the water surface nodes as weights in the calculation. Since the Fermat point is located close to the centroid of the polygon, a gradient descent algorithm starting from the centroid is used to calculate the UAV's horizontal deployment position.

[0030] Determine the drone altitude D is an adjustment factor, and the SN node that achieves the maximum distance is the k-th node. * One SN node, For the kth * The horizontal distance from each SN node to the UAV node For the kth * Horizontal distance from each SN node to the land-based data center For the kth * Vertical distance from each SN node to the land-based data center.

[0031] The flight altitude of the UAV is determined by the altitude of the experimental sea area and the land-based data center, as well as the degree of obstruction between nodes. The adjustment factor D, as a parameter affecting the realization of the line-of-sight transmission channel model, represents the degree of obstruction on the communication link.

[0032] The altitude of the UAV node will change with the horizontal movement of the UAV, minimizing the propagation distance of the water link and thus reducing propagation time. At this point, the UAV has the shortest propagation time to the land-based base. This confirmed the exact location of the drone node deployment.

[0033] (2-2) Optimize the communication time and transmission power from the surface node to the UAV node:

[0034] After determining the location of the UAV node, the radio frequency sub-channels of the SN nodes are allocated. All SN nodes are arranged in ascending order according to their remaining energy, and a group of SN nodes with symmetrical head and tail are selected to reuse the same sub-channel.

[0035] Decouple the communication link from the SN node to the UAV node. and satisfy

[0036] First of all The optimal value is initialized to ψ (0) The transmission time of the SN node becomes constant, that is...

[0037] Check if the transmit power of the SN node meets the requirements. If satisfied, then ψ (0)Decrease to a new value ψ (1) Otherwise, ψ (0) Increase to a new value ψ (1) Repeated testing;

[0038] When the termination condition |ψ is met (t) -ψ (t-1) When |≤ε, the test terminates, where ε is a set threshold; the current ψ (t) As one of the optimal values, ψ,T is thus obtained. k ,p k The optimal values ​​are expressed as follows:

[0039] (2-3) Total communication time over water:

[0040] Optimize the surface radio frequency link to minimize data transmission time. and satisfy and

[0041] The UAV node achieves its maximum transmission rate and shortest data transmission time when its transmission power is at its maximum. The shortest transmission time at this point is: Finding the optimal solution to the underwater radio frequency link optimization problem

[0042] Step (3) Optimize the channel matching of the underwater sensor node (USN) to reduce the energy consumption of node data transmission and achieve the minimum remaining energy maximization. Specific constraints include: and Specifically:

[0043] T m,k The optimal value should be In T m,k Based on obtaining the optimal value, consider To minimize the transmit power of the USN node; for a given m-th USN node and k-th SN node, there exists a channel n * , making And when n≠n * At that time, a m,k,n =0: Combining the signal-to-interference-plus-noise ratio formula and the communication rate formula, we get:

[0044]

[0045] That is, the channel allocation matrix a = [a m,k,n ] M×K×N The power matrix P = [p] of USN m,k ] M×KCorrespondingly, the power matrix of the USN node is obtained through the channel allocation matrix.

[0046] It can be further broken down into K subproblems; specifically: and satisfy and a k This represents the channel allocation matrix of the cluster consisting of the current k-th SN node. This represents the channel allocation matrix excluding the cluster formed by the current k-th SN node.

[0047] a is obtained through the maximum weight matching method. k Specifically:

[0048] First, in two disjoint sets and Establish a bipartite graph between them Edge set Weight set This represents the set of available underwater acoustic channels;

[0049] Let any edge That is, a connection can be formed between any two nodes from two sets, indicating that the i-th USN node is assigned the j-th underwater acoustic channel;

[0050] Calculate the transmit power and remaining energy of each USN node. This represents the set of USN nodes using the j-th underwater acoustic channel, yielding the initial weights.

[0051] The initial weight is the minimum value among the sets of remaining energy obtained after each node in the current k-th cluster has tried using the j-th channel. Based on this, the maximum weight matching problem is solved to obtain the optimal allocation scheme for the underwater acoustic channels.

[0052] Step (4) Optimize heterogeneous link resource allocation:

[0053] Extend the link lifespan, and combine steps (2) and (3) to allocate communication resources for the surface radio frequency link and underwater acoustic link, thereby completing the allocation of overall heterogeneous link resources.

[0054] This invention utilizes optimization theory to solve two optimal resource scheduling problems: one for surface and one for underwater environments. This achieves overall resource scheduling optimization for heterogeneous sea-air-land links, improving resource utilization and extending link lifetime. The invention emphasizes the deployment location of unmanned aerial vehicles (UAVs), leveraging their free movement and rapid deployment capabilities to act as relay nodes connecting surface nodes and land-based data centers, thus optimizing communication time in surface links. In the underwater link section, this invention analyzes the energy consumption of different nodes transmitting the same amount of data using different underwater acoustic channels. The remaining energy is used as a weight to analyze the suitability of the current channel selection, and a maximum weight matching problem is solved to obtain the optimal underwater channel allocation. Attached Figure Description

[0055] Figure 1 Deployment diagram of heterogeneous link nodes for sea, air, and land;

[0056] Figure 2 This is a flowchart of the method of the present invention. Detailed Implementation

[0057] The present invention will be further described below with reference to specific embodiments and accompanying drawings. Figure 1 To illustrate the deployment diagram of the heterogeneous sea-air-land link nodes in this implementation example, there are 36 underwater nodes and 6 surface nodes, with 10 available underwater acoustic channels. The resource allocation method for the sea-air-land heterogeneous link is detailed below. Figure 2 As shown:

[0058] Step (1) Based on the task, determine the data collection time T and the data transmission size R. Combining the communication power limitations of the underwater sensor node USN, the surface convergence node SN, and the aerial UAV node UAV, establish a heterogeneous link resource scheduling optimization problem:

[0059]

[0060]

[0061]

[0062]

[0063]

[0064]

[0065]

[0066]

[0067]

[0068]

[0069]

[0070]

[0071] a m,k,n ∈{0,1},b k,l ∈{0,1}(1.1l)

[0072] The goal of this optimization problem is to optimize resource allocation indicators. and Transmit power and Data transmission time and And the location χ of the drone, to maximize the minimum remaining energy of all USNs in each data collection cycle;

[0073] Among them, C max This indicates the maximum number of SNs that can reuse the same RF subchannel in a NOMA scheme. and Let C represent the maximum transmit power of USN, SN, and UAV, respectively. In problem (1.1), constraints (1.1a) and (1.1b) state that in underwater transmission, USNs in the same cluster occupy different underwater acoustic channels, and each USN can only occupy one channel. Constraints (1.1c) and (1.1d) state that in data transmission from SN to UAV, the subchannel can be a maximum of C max Each SN shares a channel, and each SN can only be assigned one RF subchannel. Constraints (1.1e), (1.1f), and (1.1g) ensure that the transmit power of the USN, SN, and UAV does not exceed the maximum allowed value. Constraint (1.1h) ensures that data collection from all USNs to the GBS can be completed within a given data collection time T, where Ensure that the SN with the longest transmission time can successfully complete data transmission. Constraints (1.1i), (1.1j), and (1.1k) guarantee that all data can be successfully sent to the corresponding receiving node. m,k,n Whether it is 1 indicates whether the nth channel is selected when the mth USN sends data to the kth SN, b k,l Whether the value is 1 indicates whether the l-th channel is selected when the k-th SN sends data to the UAV;

[0074] Step (2) In order to effectively solve problem (1.1), it is decoupled into two sub-problems corresponding to the two transmission segments of surface radio frequency and underwater acoustics. Solve the surface radio frequency link sub-problem; specifically:

[0075]

[0076]

[0077]

[0078]

[0079]

[0080]

[0081]

[0082] b k,l ∈{0,1}. (1.2g)

[0083] The maritime link consists of two segments: from the surface aggregation node to the drone node and from the drone node to the land-based data center. Communication time on the maritime link is optimized by scheduling the drone node locations. Specifically:

[0084] (2-1) Optimize drone node deployment:

[0085] When deploying the drone, the remaining energy of each surface node is considered, and balancing the remaining energy of all SNs in each data collection cycle is used as a design criterion. Since SNs farther from the drone typically consume more energy than other nodes when transmitting the same amount of data, the drone should be positioned closer to SNs with less remaining energy. The drone's deployment location after the s-th data collection cycle is as follows:

[0086]

[0087] This paper analyzes the UAV positioning problem by referencing the Fermat point of a polygon in mathematics. The Fermat point of a polygon is a point selected inside or outside a polygon such that the sum of the distances from that point to each corner of the polygon is minimized. The UAV position is calculated using weighted polygon Fermat point calculations, substituting the remaining energy of the water surface nodes as weights in the calculation. Since the Fermat point is located close to the centroid of the polygon, a gradient descent algorithm starting from the centroid is used to calculate the UAV's horizontal deployment position.

[0088] Based on the node distribution during implementation, the UAV's horizontal position is (2.971, 3.021), in km, with coordinate axes referenced. Figure 1 .

[0089] The flight altitude of the UAV is determined by the altitude of the experimental sea area and the land-based data center, as well as the degree of obstruction between nodes. An adjustment factor D is introduced, which serves as a parameter affecting the realization of the line-of-sight transmission channel model and represents the degree of obstruction on the communication link. The maximum horizontal distance from the surface node to the land-based data center is calculated. and vertical distance Let k be the number of SNs that achieve the maximum distance. * Calculate the distance from the drone node to k * Horizontal distance The altitude of the drone is obtained as follows

[0090] Based on the node distribution during implementation, the drone's altitude is determined to be 181 meters.

[0091] The altitude of the UAV node will change with the horizontal movement of the UAV, minimizing the propagation distance of the water link and thus reducing propagation time. At this point, the UAV has the shortest propagation time to the land-based base. This confirmed the specific locations for the drone node deployment;

[0092] Based on the node distribution during implementation, the UAV deployment location is (2.971, 3.021, 0.181), in km, with coordinate axes referenced. Figure 1 .

[0093] (2-2) Time-power optimization of the segment from the water surface node to the UAV node:

[0094] The UAV's location was determined using (2-1), and the allocation of RF subchannels for the SNs began. To derive an effective allocation scheme, it was assumed that in the NOMA scheme, a subchannel can be reused by at most two SNs. The remaining energy of all SNs was considered during subchannel allocation. All SNs were arranged in ascending order based on their remaining energy, and a pair of SNs with symmetrical beginnings and ends were selected to reuse the same subchannel. At this point, the communication link from the SN to the UAV was decoupled.

[0095]

[0096]

[0097]

[0098] Design a binary search-based algorithm to search for the optimal solution to the problem. The algorithm is described as follows: First, initialize the optimal value of the objective function of problem (1.4) to ψ. (0) Based on this, the transmission time of SN becomes constant, that is... The corresponding transmit power of SN can be further calculated according to (1.4b), and the obtained transmit power p can be checked by constraint (1.4a). k Does it meet the requirements? If constraint (1.4a) is met, then the optimal result ψ is... (0) It will decrease to a new value ψ (1) Otherwise ψ (0) It will increase to a new value ψ (1)Repeat this calculation process. When the termination condition |ψ is met... (t) -ψ (t-1) When |≤ε, the calculation ends, and ψ,T can be obtained. k ,p k The optimal values ​​are expressed as follows:

[0099] Depending on the implementation scenario, the optimal communication time ψ * The value is 0.419, in seconds. The power configuration of the surface node is [0.77W 0.65W 0.79W 0.78W 0.66W 0.77W].

[0100] (2-3) Total communication time over water:

[0101] After determining the UAV deployment location, the optimal communication time ψ for the SN to UAV segment is... * Subsequently, the waterborne link optimization problem changed as follows:

[0102]

[0103]

[0104]

[0105] Optimization problem (1.5) is a linearly bounded problem. The UAV achieves the maximum transmission rate and the shortest data transmission time when its transmission power is maximized. In summary, the optimal solution τ for the water link optimization problem is obtained. * ,

[0106] Based on the implementation scenario, the optimal solution to the waterborne link optimization problem is τ. * The value is 0.433, and the unit is seconds. The drone's power configuration is 1W.

[0107] Step (3) After obtaining the optimal solution for the communication time of the surface radio frequency link, solve the underwater acoustic link subproblem; specifically:

[0108]

[0109]

[0110]

[0111]

[0112]

[0113]

[0114] am,k,n ∈{0,1} (1.6f)

[0115] The goal of this optimization problem is to optimize the channel matching of underwater nodes, reduce the energy consumption of node data transmission, and extend the link lifetime; specifically:

[0116] T m,k The optimal value should be m∈M k In T m,k Based on obtaining the optimal value, consider minimizing the USN transmit power by satisfying (1.6e); for a given m-th USN and k-th SN, there exists a channel n * , making And when n≠n * At that time, a m,k,n =0:

[0117]

[0118] Combining the signal-to-interference-plus-noise ratio formula and the communication rate formula, we can obtain:

[0119]

[0120] That is, the channel allocation matrix a = [a m,k,n ] M×K×N The power matrix P = [p] of USN m,k ] M×K Correspondingly, the power matrix of USN can be obtained through the channel allocation matrix;

[0121] Considering the constraints from the perspective of the USN cluster, the problem can be further decomposed into K subproblems; specifically:

[0122]

[0123]

[0124]

[0125] Problem (1.9) can be solved using the maximum weight matching method. The following is the analysis process of modeling the problem as a bipartite graph; specifically:

[0126] First, in two disjoint sets and Establish a bipartite graph between them in and Let represent the edge set and the weight set respectively; secondly, let That is, a connection can be formed between any two nodes from two sets, indicating that the i-th USN is assigned the j-th channel; the transmit power and remaining energy of each USN can be calculated. Let USN be the set of underwater acoustic channels using the j-th channel. The initial weights can be obtained as follows:

[0127]

[0128] The physical meaning of this weight is the minimum value among the sets of remaining energy obtained after each node in the current k-th cluster has tried using the j-th channel. Based on this, the maximum weight matching problem is solved to obtain the optimal channel allocation scheme;

[0129] Based on the implementation scenario, the optimal allocation scheme for underwater link channels is as follows:

[0130]

[0131] Step (4) Optimize heterogeneous link resource scheduling:

[0132] By combining steps (2) and (3) to schedule communication resources for the surface and underwater links, the link lifetime is extended, and the overall heterogeneous link resource scheduling is optimized.

[0133] Based on the implementation scenario, the heterogeneous link resource scheduling is optimized as follows:

[0134] The drone is deployed at (2.971, 3.021, 0.181), in km, with the coordinate axis referenced to 2;

[0135] The communication power of the surface node is [0.77W 0.65W 0.79W 0.78W 0.66W 0.77W];

[0136] The communication power of the drone node is 1W;

[0137] The optimal channel allocation for underwater links is as follows:

[0138]

[0139]

Claims

1. A resource allocation method for a sea-air-land heterogeneous link, applicable to sea-air-land heterogeneous links whose task is marine data collection, wherein the sea-air-land heterogeneous link consists of multiple underwater sensor nodes, multiple surface aggregation nodes, one aerial UAV node, and a land-based data center, the surface portion of the heterogeneous link uses radio frequency communication, and the underwater portion uses acoustic communication; characterized in that, The specific allocation method is as follows: Step (1) Determine the data collection time T and the data transmission size based on the task. Combined with the communication power limitations of the underwater sensor node USN, the surface convergence node SN, and the aerial UAV node UAV, establish a heterogeneous link resource allocation optimization problem: optimize the underwater channel allocation index set. RF channel allocation index set USN transmit power set SN transmit power set UAV transmit power set USN to SN data transmission time set SN to UAV data transmission time set UAV to ground-based data center GBS data transmission time set And the location χ of the drone, to maximize the minimum remaining energy of all USNs in each data collection cycle. in: a m,k,n This is an identifier for whether the nth underwater channel is used when the m-th USN node sends data to the k-th SN node, a m,k,n ∈{0,1}, a m,k,n =1 means yes, a m,k,n =0 indicates no; b k,l This is an indicator for whether the l-th radio frequency channel is selected when the k-th SN node sends data to the UAV node, b k,l ∈{0,1},b k,l =1 indicates yes, b k,l =0 indicates no; m = 1,…,M, where M is the number of USN nodes; k = 1,…,K, where K is the number of SN nodes; n = 1,…,N, where N is the number of underwater channels; l = 1,…,L, where L is the number of radio frequency channels; p m p represents the transmit power of the m-th USN node. k p represents the transmit power of the k-th SN node. uav T represents the transmit power of the UAV node; m,k This represents the data transmission time when the m-th USN node sends data to the k-th SN node. T represents the path propagation time for the m-th USN node to send data to the k-th SN node; k This indicates the data transmission time when the k-th SN node sends data to the UAV node. T represents the path propagation time for the k-th SN node to send data to the UAV node; UAV This indicates the data transmission time when the UAV node sends data to the land-based data center. This indicates the path propagation time for data sent from a UAV node to a land-based data center; This represents the remaining energy of the m-th USN node after the s-th data collection cycle; The specific constraints of the resource allocation optimization problem include: Constraint a: For any USN node and any SN node, For any SN node and any underwater acoustic channel M k This is the set of USN nodes that send data to the k-th SN. Constraint b: For any radio frequency channel, C max This represents the maximum number of SN nodes allowed to reuse the same radio frequency subchannel in a non-orthogonal frequency division multiple access (NOMA) scheme; for any SN node, Constraint c: For any USN node, This represents the maximum transmit power of the USN node; for any SN node, This represents the maximum transmit power of the SN node; for the UAV node, This indicates the maximum transmit power of the UAV node; Constraint d: This represents the maximum time required for the k-th SN node and UAV node to complete data transmission. Constraint e: For any USN node and any SN node, r m,k,n R represents the rate at which the m-th USN node sends data to the k-th SN node. m,k Indicates the amount of data sent; for any SN node, r k,l R represents the rate at which the k-th SN node sends data to the UAV node. k Indicates the amount of data sent; r UAV η represents the rate at which the UAV node sends data to the land-based data center GBS, and η represents the rate of change of the total data collected by the UAV node from the SN node; Step (2) will The problem is decoupled into an above-water radio frequency link sub-problem and an underwater acoustic link sub-problem; The subproblem of the waterborne radio frequency link is to optimize the communication time of the waterborne link by changing the location of the UAV nodes, thereby minimizing the data transmission time used by the waterborne radio frequency link. Specific constraints include: and Specifically: (2-1) Determine the deployment location of the UAV nodes: The deployment location of the drone after the s-th data collection cycle: d k,UAV This represents the distance from the k-th SN node to the UAV node. Let represent the remaining energy of the k-th SN node after the s-th data collection cycle, (·). θ Indicates the weight of the remaining energy's influence; The horizontal deployment position of the UAV is determined by a gradient descent algorithm starting from the centroid of the polygon; Determine the drone altitude D is an adjustment factor, and the SN node that achieves the maximum distance is the k-th node. * One SN node, For the kth * The horizontal distance from each SN node to the UAV node For the kth * Horizontal distance from each SN node to the land-based data center For the kth * Vertical distance from each SN node to the land-based data center; The altitude of the UAV node will change with the horizontal movement of the UAV, minimizing the propagation distance of the water link and thus reducing propagation time. At this point, the UAV has the shortest propagation time to the land-based base. This confirmed the specific locations for the drone node deployment; (2-2) Optimize the communication time and transmission power from the surface node to the UAV node: Arrange all SN nodes in ascending order according to their remaining energy, and select a symmetrical group of SN nodes to reuse the same subchannel; Decouple the communication link from the SN node to the UAV node. and satisfy First of all The optimal value is initialized to ψ (0) The transmission time of the SN node becomes constant, that is... Check if the transmit power of the SN node meets the requirements. If satisfied, then ψ (0) Decrease to a new value ψ (1) Otherwise, ψ (0) Increase to a new value ψ (1) Repeated testing; When the termination condition |ψ is met (t) -ψ (t-1) When |≤ε, the test terminates, where ε is a set threshold; the current ψ (t) As one of the optimal values, ψ,T is thus obtained. k ,p k The optimal values ​​are expressed as follows: (2-3) Total communication time over water: Optimize the surface radio frequency link to minimize data transmission time. and satisfy and The UAV node achieves its maximum transmission rate and shortest data transmission time when its transmission power is at its maximum. The shortest transmission time at this point is: Finding the optimal solution to the underwater radio frequency link optimization problem Step (3) Optimize the channel matching of the underwater sensor node (USN) to reduce the energy consumption of node data transmission and achieve the minimum remaining energy maximization. Specific constraints include: and Specifically: T m,k The optimal value is In T m,k Based on obtaining the optimal value, consider To minimize the transmit power of the USN node; for a given m-th USN node and k-th SN node, there exists a channel n * , making And when n≠n * At that time, a m,k,n =0: Combining the signal-to-interference-plus-noise ratio formula and the communication rate formula, we get: That is, the channel allocation matrix a = [a m,k,n ] M×K×N The power matrix P = [p] of USN m,k ] M×K Correspondingly, the power matrix of the USN node is obtained through the channel allocation matrix; It can be further broken down into K subproblems; specifically: and satisfy and a k This represents the channel allocation matrix of the cluster consisting of the current k-th SN node. This represents the channel allocation matrix excluding the cluster formed by the current k-th SN node; a is obtained through the maximum weight matching method. k The optimal allocation scheme for the underwater acoustic channel is derived. Step (4) Optimize heterogeneous link resource allocation: Extend the link lifespan, and combine steps (2) and (3) to allocate communication resources for the surface radio frequency link and underwater acoustic link, thereby completing the allocation of overall heterogeneous link resources.

2. The resource allocation method for a heterogeneous sea-air-land link as described in claim 1, characterized in that: Constraint a indicates that in underwater transmission, USNs in the same cluster occupy different underwater acoustic channels, and each USN can only occupy one channel; Constraint b indicates that in data transmission from SN to UAV, the subchannel can be composed of a maximum of C. max Each SN is shared, and each SN can only be assigned one radio frequency sub-channel; Constraint c ensures that the transmit power of the USN, SN, and UAV does not exceed the maximum allowable value; Constraint d ensures that the data collection task is completed within the defined data collection time T; Constraint e ensures that all data can be successfully sent to the corresponding receiving node.

3. The resource allocation method for a heterogeneous sea-air-land link as described in claim 1, characterized in that, The method described above uses the maximum weight matching method to obtain a k Specifically: First, in two disjoint sets and Establish a bipartite graph between them Edge set Weight set This represents the set of available underwater acoustic channels; Let any edge That is, a connection can be formed between any two nodes from two sets, indicating that the i-th USN node is assigned the j-th underwater acoustic channel; Calculate the transmit power and remaining energy of each USN node. This represents the set of USN nodes using the j-th underwater acoustic channel, yielding the initial weights. The initial weight is the minimum value among the set of remaining energy obtained after each node in the current k-th cluster has tried using the j-th channel.

Images