An electromagnetic spectrum allocation method for a drone swarm communication system

Abstract

This invention discloses an electromagnetic spectrum allocation method for a drone swarm communication system, comprising the following steps: frequency allocation for the drone swarm communication network; time allocation for the drone swarm communication network; and power control for the drone swarm communication network. Addressing the prominent supply-demand imbalance of spectrum resources and the self-interference and mutual interference problems among drone swarms, this application proposes a spectrum allocation method for drone swarm communication networks based on non-cooperative game theory from three aspects: frequency domain, time domain, and energy domain (power domain), and conducts simulation analysis. Simulation results show that the proposed spectrum allocation method for drone swarm communication networks can effectively save spectrum resources, avoid self-interference and mutual interference among drone swarms, and has the advantages of strong anti-interference and low energy consumption.

Description

Technical Field

[0001] This invention relates to the field of spectrum management technology, and in particular to an electromagnetic spectrum allocation method for a drone swarm communication system. Background Technology

[0002] Traditional electromagnetic spectrum allocation adopts a "unified planning, hierarchical management, and manual operation" approach. This involves lower-level units reporting spectrum requests, while higher-level units allocate and assign spectrum resources based on task assignments, issuing plans and organizing coordination. However, with the continuous emergence of large-scale, multi-type frequency-using equipment, the traditional spectrum allocation model, primarily driven by manual experience and models, is clearly no longer suitable for military development needs. Spectrum allocation should shift towards a "dynamic, intelligent, and autonomous" model.

[0003] Problem 1: The supply and demand imbalance of spectrum resources is prominent. The scarcity of spectrum resources has always been a major factor restricting the effectiveness of UAVs in combat. Currently, UAV operations still adopt a static spectrum allocation mode of "one UAV, one channel," which is clearly unable to meet the spectrum resource requirements of large-scale UAV swarms in the future.

[0004] Question 2: Severe Interference Issues. Besides the scarcity of spectrum resources, self-interference and mutual interference between radiation sources and the complexity of the electromagnetic environment are also significant factors restricting UAV operations. UAV swarm networks are wireless communication networks composed of a large number of closely spaced nodes; self-interference and mutual interference between nodes, as well as interference from other weapon platforms, are key issues that need to be addressed. Summary of the Invention

[0005] The technical problem to be solved by the present invention is how to provide an electromagnetic spectrum allocation method for a drone swarm communication system that can effectively save spectrum resources, avoid self-interference and mutual interference between drone swarms, and has the advantages of strong anti-interference and low energy consumption.

[0006] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is: an electromagnetic spectrum allocation method for a drone swarm communication system, characterized by comprising the following steps:

[0007] S1: Frequency allocation for drone swarm communication network;

[0008] S2: Time allocation for drone swarm communication network;

[0009] S3: Power control for drone swarm communication network.

[0010] A further technical solution involves the following steps in the specific method for allocating frequencies in the drone swarm communication network:

[0011] The drone swarm adopts a hierarchical mesh topology structure. The hierarchy of the drone nodes determines the communication frequency of the nodes. Drone nodes at the same hierarchy can use the same communication frequency. The communication range of higher-level drone nodes is greater than that of lower-level drone nodes.

[0012] 1) The lowest-level drone nodes use the same communication frequency to exchange information and transmit information to the cluster head node;

[0013] 2) Intermediate-level drone nodes are both the cluster heads of lower-level drone nodes and cluster members of higher-level nodes. Therefore, they can use two different communication frequencies: one frequency is used to communicate with cluster members within the lower-level cluster, and the other frequency is used to maintain communication with nodes at the same level and the cluster head node at the higher level.

[0014] 3) The highest-level nodes can also use two different communication frequencies: one frequency is used to communicate with lower-level cluster members within the same cluster family, and the other communication frequency is used to exchange information with space-based, airborne, or ground-based command and information systems.

[0015] A further technical solution involves the following steps in the specific method for time allocation in the drone swarm communication network:

[0016] S2-1: First, the cluster head drone node randomly and non-repeatingly allocates time slots to its m cluster member nodes;

[0017] S2-2: Through inter-machine information exchange, each node will send its own time slot T i It passes the timeslots {T} to neighboring nodes and receives the timeslots {T} from all neighboring nodes. j},j∈φ i Each node calculates its own effect coefficient E. i If there is a time conflict with a neighboring node, then E i =0; if there is no time conflict with neighboring nodes, then E i =1;

[0018] S2-3: Through inter-machine information exchange, each node will transmit its own effect coefficient E. i It is passed to the cluster head node and neighboring nodes, and receives the slot effect coefficients {E} from all neighboring nodes. j},j∈φ i Each node calculates its own effect function u. i (T i ,T -i The highest-level cluster head node obtains the effect coefficients of all nodes and calculates the overall efficiency coefficient E of the entire network.

[0019] S2-4: E for each effect coefficient iFor nodes with a time update strategy of 0, the affected node i selects the time T with the smallest number in the set of time strategies that does not conflict with its neighbors. min And broadcast it to the cluster head node, the cluster head node will then implement the original policy T within the cluster. min The time slot of the member node is updated to the time slot T of node i. i This update strategy avoids time conflicts between node i and its neighboring nodes, as well as with its member nodes within the same cluster. If node i's neighboring nodes have already used up all available time slots, then node i's time slot will not be updated in this round.

[0020] S2-5: Broadcast the updated timeslot to neighboring nodes, and repeat steps S2-2 to S2-4;

[0021] S2-6: When the utility functions of all nodes reach their maximum, i.e., Nash equilibrium is reached, the iteration stops, and the time slot of each node is sent to the cluster head to obtain the final overall efficiency coefficient E.

[0022] A further technical solution is that the specific method for power control of the UAV swarm communication network includes the following steps:

[0023] S3-1: First, initialize the transmit power of all drone nodes in the network;

[0024] S3-2: The network system sends the time slot, transmit power, and location information of each node at the same level, as well as the cluster head location information of the cluster family to which node i belongs, from top to bottom according to the hierarchy.

[0025] S3-3: Each node obtains the transmit power and location of nodes in the same time slot as itself at the same level, and then calculates its own signal-to-interference ratio γ. i and effect function u i (p i ,p -i );

[0026] S3-4: Update the node's transmit power according to the iterative formula, and broadcast the updated transmit power to neighboring nodes and cluster head nodes through inter-machine information exchange;

[0027] S3-5: Compare the transmit power of node i after the iteration. Compared with the transmit power before iteration If there are any changes, continue repeating steps S3-2 to S3-4. If there are basically no changes, end the iteration and finally obtain the transmit power strategy for each node.

[0028] A further technical solution is that the signal-to-interference ratio γ i The calculation method includes the following steps:

[0029] Suppose that in a certain layer of a drone swarm network, there are l drone nodes operating in the same time slot, where node i has a transmit power of p. i The signal-to-interference ratio (SIR) of node i at the receiver is γ. i Typically, the channel gain used by a node is In the formula, d is the communication distance from node i to the cluster head, A is the constant gain, which refers to the channel gain when the distance is 1 meter; c ij Let represent the spreading code correlation coefficient between node i and node j, then the interference power of other nodes on node i is: σ 2 The background noise power of the electromagnetic environment is represented by γ, and the signal-to-interference ratio at node i is also represented by γ. i for:

[0030]

[0031] The signal-to-interference ratio γ of each node i at the receiver i It should be greater than or equal to the minimum threshold of the signal-to-interference ratio. That is, satisfy This ensures the communication quality of node i; simultaneously, the transmit power p of each node i... i Both should be less than the maximum transmit power. That is, satisfy

[0032] A further technical solution involves making the utility function of the non-cooperative power control game in the drone swarm communication network as follows:

[0033]

[0034] In the formula, p -i For power strategy selection of nodes other than node i, For the target signal-to-interference ratio, a i and b i The adjustment cost factor for signal-to-interference ratio and transmit power;

[0035] Throughout the game, each node i continuously adjusts its transmission power to maximize its utility function. At that time, Nash equilibrium is reached.

[0036] A further technical solution is that the calculation method of the iterative formula includes the following steps;

[0037] To obtain the Nash equilibrium solution, the utility function is partially differentiated with respect to...

[0038]

[0039] Setting the above expression to zero and substituting expression (1) into the above expression, we get

[0040]

[0041]

[0042] Solving equations (5) and (1) together, we can obtain...

[0043]

[0044] The iterative formula for the transmit power can be obtained using Newton's iterative algorithm.

[0045]

[0046] The beneficial effects of adopting the above technical solution are as follows: Addressing the prominent contradiction between the supply and demand of spectrum resources for UAV swarms and the problems of self-interference and mutual interference among swarm UAVs, a spectrum allocation method for UAV swarm communication networks based on non-cooperative game theory is proposed from three aspects: frequency domain, time domain, and energy domain (power domain), and simulation analysis is conducted. Simulation results show that the spectrum allocation method for UAV swarm communication networks proposed in this application can effectively save spectrum resources, avoid self-interference and mutual interference among swarm UAVs, and has the advantages of strong anti-interference and low energy consumption. Attached Figure Description

[0047] The present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments.

[0048] Figure 1 This is a schematic diagram of the multi-layer mesh topology of the UAV swarm communication network in an embodiment of the present invention;

[0049] Figure 2 This is a timing diagram of information transmission and reception by nodes in the UAV swarm communication network in an embodiment of the present invention;

[0050] Figure 3a This is a graph showing the relationship between the signal-to-interference ratio and utility of network nodes in an embodiment of the present invention.

[0051] Figure 3b This is a graph showing the relationship between the transmission power and utility of network nodes in an embodiment of the present invention.

[0052] Figure 4 This is a graph showing the change in the utility function of the third-layer network nodes in this embodiment of the invention;

[0053] Figure 5 This is a graph showing the overall performance coefficient variation of the third-layer network in this embodiment of the invention;

[0054] Figure 6 This is a time allocation diagram of the third-layer network nodes in an embodiment of the present invention;

[0055] Figure 7This is a graph showing the change in the utility function of the fourth-layer network node in this embodiment of the invention;

[0056] Figure 8 This is a graph showing the overall performance coefficient variation of the fourth layer network in this embodiment of the invention;

[0057] Figure 9 This is a time allocation diagram of the fourth-layer network nodes in this embodiment of the invention;

[0058] Figure 10 This is a graph showing the signal-to-interference ratio (SIR) variation at the receivers of all nodes in the third-layer network in this embodiment of the invention.

[0059] Figure 11 This is a graph showing the transmission power variation of all nodes in the third-layer network in this embodiment of the invention.

[0060] Figure 12 This is a graph showing the signal-to-interference ratio (SIR) variation at the receivers of all nodes in the fourth-layer network in this embodiment of the invention.

[0061] Figure 13 This is a graph showing the transmission power variation of all nodes in the fourth layer network in this embodiment of the invention;

[0062] Figure 14 This is a flowchart of the method described in an embodiment of the present invention. Detailed Implementation

[0063] 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 a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0064] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0065] This invention discloses an electromagnetic spectrum allocation method for a drone swarm communication system. The method is based on game theory algorithms and performs adaptive spectrum allocation from three aspects: frequency domain, time domain, and energy domain (power domain). Figure 14 As shown.

[0066] The steps described above will be explained in detail below with reference to specific content.

[0067] Frequency allocation for drone swarm communication networks:

[0068] Large-scale UAV swarm warfare networks involve numerous nodes. If each node were to directly interact with the command information system or exchange information with each other, it would undoubtedly require high network overhead and a large amount of spectrum resources. A hierarchical mesh topology for UAV swarms, where the hierarchy of UAV nodes determines their communication frequency, allows nodes at the same hierarchy to use the same communication frequency, significantly conserving spectrum resources. Figure 1 As shown: ① The lowest-level UAV nodes use the same communication frequency to exchange information and transmit information to the cluster head node; ② The intermediate-level nodes are both the cluster head of the lower-level nodes and the cluster members of the upper-level nodes, so they can use two different communication frequencies: one frequency is used to communicate with cluster members within the lower-level cluster, and the other frequency is used to maintain communication with nodes at the same level and the cluster head node of the upper level; ③ The highest-level nodes can also use two different communication frequencies: one frequency is used to communicate with cluster members within the lower-level cluster, and the other frequency is used to exchange information with space-based, airborne, or ground-based command and information systems.

[0069] Drone swarms employing this hierarchical cluster-based network topology only require the highest-level leader drone to communicate with space-based, airborne, or ground-based command and information systems, selecting the appropriate frequency range and bandwidth as needed. For information exchange between drone network nodes with extremely short communication distances, infrared, visible light, and ultraviolet spectrum resources can be used.

[0070] This frequency allocation method for drone swarm communication networks greatly saves spectrum resources, alleviates the supply and demand imbalance of spectrum resources (especially the microwave band where the supply and demand imbalance is prominent), and at the same time enhances the network's anti-interference capability and reduces the network's energy consumption.

[0071] Drone swarm communication network time allocation:

[0072] Time-division multiplexing is an effective method for allocating electromagnetic spectrum resources and reducing mutual interference between drone swarms. The cluster head node allocates different time slots for data transmission to cluster member nodes. Member nodes send status information, task information, and other data to the cluster head according to the specified time. The cluster head node receives information transmitted from all members within the cluster. After a round of data transmission is completed, the collected information is fused and compressed to reduce data redundancy, and the processed information is sent to the next higher-level cluster head member, such as... Figure 2 As shown.

[0073] (I) System Description

[0074] A drone swarm employing a hierarchical mesh network topology experiences interference only from drone nodes operating at the same time and frequency within the same layer. Assuming the drone swarm network has n drone nodes, and a given node i belongs to a cluster with m members, then the available time slot T for node i is... i ={T1,T2,…,T m Define the effect coefficient E. i This indicates whether node i effectively performs its combat function, i.e., whether it is interfered with by nodes within the cluster and by its neighbors.

[0075]

[0076] The overall efficiency coefficient E of the entire drone swarm network system represents whether all nodes in the network are subject to interference.

[0077]

[0078] Clearly, the overall effect coefficient E reaches its maximum value n when all nodes in the network are undisturbed and can effectively exert their combat effectiveness.

[0079] Therefore, the next problem to be solved is to rationally design the time slot of each node so that each UAV swarm network node is not disturbed by other nodes and the overall effect coefficient E of the system reaches its maximum value.

[0080] (II) Non-cooperative time-division multiplexing game model

[0081] Game theory is a powerful mathematical tool for studying competition and cooperation, and for developing optimal strategies. Based on whether a binding cooperative agreement can be reached when participants interact, game theory can be divided into cooperative games and non-cooperative games. A key issue in non-cooperative game research is how participants choose strategies to initially maximize their own interests when their interests or gains are interdependent. This aligns perfectly with the distributed adaptive characteristics of the UAV swarm network nodes in this application.

[0082] The non-cooperative time-division multiplexing game model can be represented as G=[N,{T i} i∈N ,{u i} i∈N ], where the participants N = {1,2,...,n} is the set of all nodes in the drone swarm network, and the policy space {T i} i∈N Represents the time policy vector of all nodes, {u i} i∈N This represents the utility function for each node.

[0083] For a distributed drone swarm network topology, from the perspective of system performance, the utility function of a drone node is defined as follows:

[0084]

[0085] In the formula T -i For the selection of time strategies for nodes other than node i, φ i Let T represent all the neighboring nodes of node i. Therefore, the first term of the utility function represents the time slot T selected by node i. i The first term represents the utility of node i itself, and the second term represents the influence of its neighbors, which is the sum of the utilities of all neighboring nodes. The utility function reflects the local utility of each node i. In the time-division multiplexing game, each node tries to maximize its own utility function. That is, reaching Nash equilibrium.

[0086] (III) Algorithm Iteration Design

[0087] Nash equilibrium is a state of strategy combination for all participants, where each participant's strategy is the optimal strategy chosen based on the strategies of all other participants. If node i chooses a time slot that is neither the same as its own cluster members nor the same as its neighboring cluster members, then node i can maximize its effect function. The iterative algorithm for the non-cooperative time-division multiplexing game model is as follows:

[0088] Step 1: First, the cluster head node randomly and non-repeatingly allocates time slots to its m cluster member nodes.

[0089] Step 2: Through inter-machine information exchange, each node will send its own time slot T i It passes the timeslots {T} to neighboring nodes and receives the timeslots {T} from all neighboring nodes. j},j∈φ i Each node calculates its own effect coefficient E. i If there is a time conflict with a neighboring node, then E i =0; if there is no time conflict with neighboring nodes, then E i =1.

[0090] Step 3: Through inter-machine information exchange, each node will send its own effect coefficient E. i It is passed to the cluster head node and neighboring nodes, and receives the slot effect coefficients {E} from all neighboring nodes. j},j∈φ i Each node calculates its own effect function u. i (T i ,T -i The highest-level cluster head node obtains the effect coefficients of all nodes and calculates the overall efficiency coefficient E of the entire network.

[0091] Step 4: Each effect coefficient E i For nodes with a time update strategy of 0, the affected node i selects the time T with the smallest number in the set of time strategies that does not conflict with its neighbors. min And broadcast it to the cluster head node, the cluster head node will then implement the original policy T within the cluster. min The time slot of the member node is updated to the time slot T of node i. i This update strategy avoids time conflicts between node i and its neighbors, as well as with other nodes within the same cluster. If node i's neighbors have already used all available time slots, then node i's time slot will not be updated in this round.

[0092] Step 5: Broadcast the updated timeslot to neighboring nodes, and repeat steps 2-4.

[0093] Step 6: When the utility functions of all nodes reach their maximum, i.e., Nash equilibrium is reached, the iteration stops, and the time slot of each node is sent to the cluster head to obtain the final overall efficiency coefficient E.

[0094] This iterative algorithm combines the characteristics of non-cooperative game theory with those of greedy algorithms. Each updating node assumes the policies of its neighbors remain unchanged, seeking the time slot that maximizes its local efficiency and minimizes its time slot number. This approach ensures the system eventually converges to a Nash equilibrium while minimizing the number of time slots used.

[0095] Drone swarm communication network power control:

[0096] In a network, nodes can achieve better communication quality by increasing their transmission power, but this also interferes with other nodes, thus reducing their communication quality. Furthermore, increasing transmission power also increases energy consumption. Therefore, choosing a scientifically sound power control strategy is crucial to ensure that each node can achieve good communication quality with relatively low transmission power.

[0097] (I) System Description

[0098] After frequency and time allocation, interference in a drone swarm network mainly comes from drone nodes at the same level that are relatively far apart and choose to share the same time slot. Assume that in a certain level of the drone swarm network, there are *l* drone nodes operating in the same time slot, where node *i* has a transmit power of *p*. i The signal-to-interference ratio (SIR) of node i at the receiver is γ. i (The signal-to-interference ratio (SIR) is the most important indicator for measuring communication quality.) Typically, the channel gain used by a node is... In the formula, d is the communication distance from node i to the cluster head, and A is a constant gain, referring to the channel gain at a distance of 1 meter. ij Let represent the spreading code correlation coefficient between node i and node j, then the interference power of other nodes on node i is: σ 2 This indicates the background noise power of the electromagnetic environment.

[0099] The signal-to-interference ratio γ of node i i for

[0100]

[0101] The signal-to-interference ratio γ of each node i at the receiver i It should be greater than or equal to the minimum threshold of the signal-to-interference ratio. That is, satisfy This ensures the communication quality of node i. Simultaneously, the transmit power p of each node i... i Both should be less than the maximum transmit power. That is, satisfy

[0102] (II) Non-cooperative power control game model

[0103] The non-cooperative power control game model can be represented as G=[N,{P i} i∈N ,{u i} i∈N ], where the participants N = {1,2,...,n} is the set of all nodes in the drone swarm network, and the policy space {P i} i∈N Represents the power policy vector of all nodes, {u i} i∈N This represents the utility function for each node.

[0104] When designing the utility function, both transmit power and signal-to-interference ratio (SIR) must be considered. When the transmit power of node i remains constant, a higher SIR at the receiver results in better communication quality for node i. Conversely, when the SIR at the receiver remains constant, a higher transmit power does not necessarily improve communication quality, but it increases energy consumption. It is worth noting that when the transmit power of node i remains constant, after the SIR at the receiver increases to a certain extent, the utility will not continue to increase but will gradually approach a constant. like Figures 3a-3b As shown.

[0105] Let the utility function of the non-cooperative power control game in the drone swarm communication network be:

[0106]

[0107] In the formula, p -i For power strategy selection of nodes other than node i, For the target signal-to-interference ratio, a i and b i This is the adjustment cost factor for signal-to-interference ratio and transmit power.

[0108] Throughout the game, each node i continuously adjusts its transmission power to maximize its utility function. At that time, Nash equilibrium is reached.

[0109] (III) Iterative Algorithm

[0110] To obtain the Nash equilibrium solution, take the partial derivative of the utility function with respect to:

[0111]

[0112] Setting the above expression to zero and substituting expression (4) into the above expression, we get:

[0113]

[0114]

[0115] Solving equations (8) and (4) together, we get:

[0116]

[0117] The iterative formula for the transmit power can be obtained using Newton's iterative algorithm:

[0118]

[0119] The iterative algorithm for the non-cooperative power control game model of drone swarm networks is as follows:

[0120] Step 1: First, initialize the transmit power of all nodes in the network.

[0121] Step 2: The network system sends the time slot, transmit power, and location information of each node at the same level from top to bottom according to the hierarchy, as well as the cluster head location information of the cluster family to which node i belongs.

[0122] Step 3: Each node obtains the transmit power and location of nodes in the same time slot as itself at the same level, and then calculates its own signal-to-interference ratio γ. i and effect function u i (p i ,p -i ).

[0123] Step 4: Update the transmit power of the node according to the iterative formula (10), and broadcast the updated transmit power to neighboring nodes and cluster head nodes through inter-machine information exchange.

[0124] Step 5: Compare the transmit power of node i after the iteration. Compared with the transmit power before iteration If there are any changes, continue to repeat Step 2-Step 4. If there are basically no changes, end the iteration and finally obtain the transmit power strategy of each node.

[0125] Simulation analysis:

[0126] To verify the effectiveness of the spectrum allocation method for UAV swarm communication networks proposed in this chapter in alleviating the contradiction between spectrum resource supply and demand, avoiding mutual interference and self-interference between UAVs, and reducing energy consumption, simulation analysis was conducted using the MATLAB software platform. The simulation scenario deployed a swarm of 100 UAVs within a 500×500 two-dimensional planar grid. First, a four-layer distributed mesh network topology was constructed.

[0127] Frequency allocation:

[0128] Frequency allocation was performed on a four-layer distributed mesh topology composed of 100 drone swarms. According to the spectrum allocation method proposed above, the four-layer distributed mesh topology only requires communication channels in four frequency bands: frequency f1 within the microwave band (300MHz-300GHz), which is the main application band for drone communication systems and also the band with the most prominent spectrum resource supply-demand imbalance; and frequencies f2, f3, and f4 within the solar blind zone ultraviolet light (1.07-1.5PHz).

[0129] ① The first level consists of one UAV, which occupies the resources of two channels. One is a microwave channel frequency f1, used for information exchange with the command information system. This usually requires a large bandwidth. Different communication links are selected depending on whether the UAV is connected to a space-based, airborne, or ground-based control station, such as the common Ka-band, Ku-band, X-band, C-band, and L / S-band. The other is a solar blind zone ultraviolet light channel frequency f2, used for information exchange with the four cluster member UAVs in the second level to obtain data information from all UAV nodes.

[0130] ② The second-level group of four nodes are both members of the first-level cluster and cluster heads of the third-level nodes. They share the resources of two channels: one is the solar blind zone ultraviolet light channel frequency f2, used for information exchange with the first-level cluster head node and the second-level neighbor nodes; the other is the solar blind zone ultraviolet light channel frequency f3, used for information exchange with their respective cluster member nodes in the third level.

[0131] ③ The third level consists of 16 drones, divided into 4 clusters. The cluster head corresponds to each drone in the previous level and is also the cluster head of the corresponding node in the fourth level. They share the resources of two channels: one is the solar blind zone ultraviolet light channel frequency f3, used for information exchange with the cluster head node in the second level and the neighboring nodes in the third level; the other is the solar blind zone ultraviolet light channel frequency f4, used for information exchange with their respective cluster member nodes in the fourth level.

[0132] ④ The fourth level consists of 79 drones, divided into 16 clusters. Each cluster has 4 or 5 drones, and the cluster heads correspond to each drone in the previous level. They share a single solar blind zone ultraviolet light channel frequency f4 for information exchange with the cluster head nodes of the third level and neighboring nodes of the fourth level.

[0133] If frequency allocation is performed according to the current centralized control method for UAVs, a swarm of 100 UAVs would require 100 microwave channel resources. However, the frequency allocation method for UAV swarm communication network proposed in this application only requires 1 microwave channel resource, which greatly saves valuable microwave spectrum resources. Moreover, the use of wireless ultraviolet light channel frequencies between UAVs has significant advantages such as strong confidentiality, strong anti-interference, and green energy saving.

[0134] Time allocation

[0135] After frequency allocation, a swarm communication network of 100 UAVs in a four-layer structure is time-division multiplexed to avoid co-channel interference between nodes. According to the time-division multiplexing iterative algorithm, the allocated time slots for all network nodes are first initialized. The cluster head node of each cluster randomly and non-repeatingly allocates time slots to its cluster members. After initialization, nodes will not have time conflicts with members within their own cluster, but may have time conflicts with neighboring nodes of other clusters. In the simulation, the first and second layer network nodes of the four-layer distributed mesh topology composed of 100 UAVs do not have other clusters, so there are no time conflicts. The third and fourth layer network nodes engage in a game based on the iterative algorithm of a non-cooperative time-division multiplexing game model, continuously updating the time slots of the interfered nodes until a Nash equilibrium solution is reached.

[0136] The utility functions of third-level network nodes and the overall network performance coefficients are as follows: Figure 4 and Figure 5 As shown in the figure, after three iterations, the effect coefficients of each node and the overall network efficiency coefficient converge rapidly. The final overall efficiency coefficient E of the third-level network is 16, indicating that none of the 16 nodes in the third level have time conflicts and can perform normally. The final time slots allocated to the nodes in the third-level network are shown in the figure. Figure 6 As shown.

[0137] The utility function of the fourth-level network nodes and the overall network performance coefficient are as follows: Figure 7 and Figure 8 As shown in the figure, after 6 iterations, the effect coefficients of each node and the overall network efficiency coefficient converge rapidly. The final overall network efficiency coefficient E of the fourth layer is 79, indicating that none of the 79 nodes in the third layer experienced time conflicts and were able to perform effectively. The final time slots allocated to the nodes in the fourth layer are shown in the figure. Figure 9 As shown.

[0138] In the simulation, all nodes of the 100-drone swarm communication network were allocated available time slots and were compatible with each other without interference. This shows that the time-division multiplexing time allocation method for drone swarm communication networks based on the non-cooperative game model proposed in this application can effectively avoid interference between swarm drones and has a good convergence speed, enabling rapid realization of communication time allocation for large-scale drone swarm networks.

[0139] Power control

[0140] After frequency and time allocation, the interference in the 100 four-layer drone swarm communication network mainly comes from selecting drone nodes of the same layer that are relatively far apart in the same time slot. The transmission power of these nodes is adjusted so that each node can obtain good communication quality with a lower transmission power.

[0141] In the simulation, it is assumed that the constant gain A = 10 in the channel gain formula and the correlation coefficient c of the spreading code in the signal-to-interference ratio formula is... ij =1, Background noise σ 2 =1×10 -15 W, the minimum threshold for the signal-to-interference ratio. Target Information Relationship Maximum transmit power Adjustment cost factor a for the signal-to-interference ratio i =1, the cost factor b for adjusting the transmit power i =1.

[0142] Since none of the nodes in the first and second layers operate at the same frequency and time, the transmission power only needs to be adjusted based on the background noise, distance information, and target signal-to-noise ratio.

[0143] Assume the initial power of all nodes in the third and fourth layers. The iterative formula (10) updates the transmit power of the nodes iteratively until the Nash equilibrium solution is reached. The changes in signal-to-interference ratio and transmit power of all nodes in the third and fourth layers are as follows: Figures 10 to 13 As shown in the figure, the signal-to-interference ratio (SIR) and transmit power of all nodes converge rapidly, and the SIR of all nodes eventually approaches the target SIR. The transmit power of all nodes is less than the maximum transmit power. Moreover, since the fourth-layer network nodes are closer to their respective receiver nodes than the third-layer network nodes, the transmission power of the fourth-layer network nodes is significantly less than that of the third-layer network nodes.

[0144] In the simulation, the transmission power of all nodes in the 100-drone swarm communication network was adjusted. It can be seen that the power control method for drone swarm communication networks based on the non-cooperative game model proposed in this application can effectively adjust the transmission power of the swarm drones, reduce energy consumption, avoid interference between swarm drones, and has a good convergence speed.

[0145] To address the prominent supply-demand imbalance of spectrum resources for UAV swarms and the problems of self-interference and mutual interference among swarm UAVs, this paper proposes a spectrum allocation method for UAV swarm communication networks based on non-cooperative game theory from three aspects: frequency domain, time domain, and energy domain (power domain), and conducts simulation analysis. Simulation results show that the proposed UAV swarm communication network spectrum allocation method can effectively save spectrum resources, avoid self-interference and mutual interference among swarm UAVs, and has the advantages of strong anti-interference and low energy consumption.

Claims

1. A method for electromagnetic spectrum allocation in an unmanned aerial vehicle (UAV) swarm communication system, characterized in that... Includes the following steps: S1: Frequency allocation for drone swarm communication network; S2: Time allocation for drone swarm communication network; S3: Power control for drone swarm communication network; The specific method for frequency allocation in the UAV swarm communication network includes the following steps: The drone swarm adopts a hierarchical mesh topology structure. The hierarchy of the drone nodes determines the communication frequency of the nodes. Drone nodes at the same hierarchy can use the same communication frequency. The communication range of higher-level drone nodes is greater than that of lower-level drone nodes. 1) The lowest-level drone nodes use the same communication frequency to exchange information and transmit information to the cluster head node; 2) The intermediate-level drone nodes are both the cluster heads of the lower-level drone nodes and the cluster members of the upper-level nodes. Therefore, they can use two different communication frequencies: one frequency is used to communicate with the cluster members in the lower-level cluster, and the other frequency is used to maintain communication with the nodes in the same level and the cluster head nodes in the upper level. 3) The highest-level nodes can also use two different communication frequencies: one frequency is used to communicate with lower-level cluster members within the same cluster family, and the other communication frequency is used to exchange information with space-based, airborne, or ground-based command and information systems. The specific method for time allocation in the UAV swarm communication network includes the following steps: S2-1: First, the cluster head drone node randomly and non-repeatingly allocates time slots to its m cluster member nodes; S2-2: Through inter-machine information exchange, each node will send its own time slot T i It passes the timeslots to neighboring nodes and receives the timeslots from all neighboring nodes. Each node calculates its own effect coefficient E. i If there is a time conflict with a neighboring node, then E i =0; if there is no time conflict with neighboring nodes, then E i =1; S2-3: Through inter-machine information exchange, each node will transmit its own effect coefficient E. i It is passed to the cluster head node and neighboring nodes, and receives the slot effect coefficients from all neighboring nodes. Each node calculates its own effect function. The highest-level cluster head node obtains the effect coefficients of all nodes and calculates the overall efficiency coefficient E of the entire network. S2-4: E for each effect coefficient i For nodes whose update time strategy is 0, the affected node i selects the time T with the smallest number in the set of time strategies that does not conflict with its neighbors. min And broadcast it to the cluster head node, the cluster head node will then implement the original policy T within the cluster. min The time slot of the member node is updated to the time slot T of node i. i This update strategy avoids time conflicts between node i and its neighboring nodes, as well as with its member nodes within the same cluster. If node i's neighboring nodes have already used up all available time slots, then node i's time slot will not be updated in this round. S2-5: Broadcast the updated timeslot to neighboring nodes, and repeat steps S2-2 to S2-4; S2-6: When the utility functions of all nodes reach their maximum, i.e., Nash equilibrium is reached, the iteration stops, and the time slot of each node is sent to the cluster head to obtain the final overall efficiency coefficient E; The specific method for power control of the UAV swarm communication network includes the following steps: S3-1: First, initialize the transmit power of all drone nodes in the network; S3-2: The network system sends the time slot, transmit power, and location information of each node at the same level, as well as the cluster head location information of the cluster family to which node i belongs, from top to bottom according to the hierarchy. S3-3: Each node obtains the transmit power and location of nodes in the same time slot as itself at the same level, and then calculates its own signal-to-interference ratio γ. i and effect function ; S3-4: Update the node's transmit power according to the iterative formula, and broadcast the updated transmit power to neighboring nodes and cluster head nodes through inter-machine information exchange; S3-5: Compare the transmit power of node i after the iteration. Compared with the transmit power before iteration If there are any changes, continue repeating steps S3-2 to S3-4. If there are basically no changes, end the iteration and finally obtain the transmit power strategy for each node.

2. The electromagnetic spectrum allocation method for a drone swarm communication system as described in claim 1, characterized in that, The signal-to-interference ratio γ i The calculation method includes the following steps: Suppose that in a certain layer of a drone swarm network, there are l drone nodes operating in the same time slot, where node i has a transmit power of p. i The signal-to-interference ratio (SIR) of node i at the receiver is γ. i Typically, the channel gain used by a node is In the formula, d is the communication distance from node i to the cluster head, A is the constant gain, which refers to the channel gain when the distance is 1 meter; c ij Let represent the spreading code correlation coefficient between node i and node j, then the interference power of other nodes on node i is: ; The background noise power of the electromagnetic environment is represented by γ, and the signal-to-interference ratio at node i is also represented by γ. i for: （1） The signal-to-interference ratio γ at the receiver for each node i i It should be greater than or equal to the minimum threshold of the signal-to-interference ratio. That is, satisfying This is necessary to ensure the communication quality of node i; simultaneously, the transmit power p of each node i... i Both should be less than the maximum transmit power. That is, satisfying .

3. The electromagnetic spectrum allocation method for a drone swarm communication system as described in claim 2, characterized in that, Let the utility function of the non-cooperative power control game in the drone swarm communication network be: （2） In the formula, For power strategy selection of nodes other than node i, For the target signal-to-interference ratio, a i and b i The adjustment cost factor for signal-to-interference ratio and transmit power; Throughout the game, each node i continuously adjusts its transmission power to maximize its utility function. At that time, Nash equilibrium is reached.

4. The electromagnetic spectrum allocation method for a drone swarm communication system as described in claim 3, characterized in that, The calculation method of the iterative formula includes the following steps; To obtain the Nash equilibrium solution, the utility function is partially differentiated with respect to... （3） Setting the above expression to zero and substituting expression (1) into the above expression, we get （4） （5） Solving equations (5) and (1) together, we can obtain... （6） The iterative formula for the transmit power can be obtained using Newton's iterative algorithm: （7）。

Images