Channel time slot allocation method for wearable wireless sensor networks supporting WPT technology

Abstract

This invention provides a channel time slot allocation method for wearable wireless sensor networks supporting WPT technology, belonging to the field of wearable wireless sensor network technology. This invention proposes a channel time slot allocation scheme based on particle swarm optimization (PSO). This includes using WPT technology to collect node energy to establish an optimal node power model, transforming it into a geometric problem to solve for the optimal transmit power, and based on this model, analyzing the optimization objective of the time slot allocation model in WWSN, solving it using the PSO algorithm, and verifying the proposed algorithm through simulation. Furthermore, performance comparisons were conducted with two other typical channel allocation protocols in terms of energy consumption and latency, demonstrating significant performance advantages. This invention can address the problem of wasted time slot resources by dynamically adjusting the allocation of time slots in the channel according to the data packet size and the maximum data packet capacity of each time slot, achieving time slot reuse to improve channel bandwidth utilization and enhance overall network performance.

Description

Technical Field

[0001] This invention relates to the field of wearable wireless sensor network technology, and more particularly to a method for allocating channel time slots in wearable wireless sensor networks that supports WPT technology. Background Technology

[0002] Wearable wireless sensor network (WWSN) technology for medical monitoring has developed rapidly in recent years, providing technical support for the early detection and prevention of chronic diseases in healthcare systems and showing promising development prospects. Traditional battery-powered WWSNs often face the problem of the entire system failing and malfunctioning due to battery depletion. Furthermore, periodic battery replacement is not easy. Therefore, this invention introduces WPT technology to provide a continuous power supply to the WWSN, thereby extending the sensor's lifespan. In practical WWSN applications, some sensor nodes carry data packets that do not reach the maximum capacity of a time slot, resulting in inefficient time slot utilization and wasted time slot resources. Therefore, traditional fixed time slot allocation schemes are not suitable for most communication scenarios. One way to improve time slot utilization is to introduce dynamism into superframes. Dynamic allocation schemes help to utilize superframes more effectively. A common method to achieve dynamism in superframes is to change the length of the time slot. Although the above methods play an important role in improving channel utilization, simply changing the time slot length also has certain problems. For example, in practical application scenarios, it is difficult to determine the amount of time slot length to change. Uniformly changing the fixed time slot length cannot meet the heterogeneity characteristics of sensor nodes. Therefore, the above allocation scheme is not suitable for the WWSN scenario. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to address the shortcomings of the prior art by providing a channel time slot allocation method for wearable wireless sensor networks that supports WPT technology. This method can address the problem of wasted time slot resources by dynamically adjusting the allocation of time slots in the channel based on the data packet size and the maximum capacity of the data packets that a time slot can hold, thereby achieving time slot reuse, improving channel bandwidth utilization, and enhancing the overall network performance.

[0004] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows: A method for allocating channel time slots in a wearable wireless sensor network supporting WPT technology includes the following steps: Step 1: Construct the network topology; This paper studies the channel time slot allocation problem of WWSN based on a single-hop star topology. It is assumed that a star topology WWSN consists of a coordinator node and N sensor nodes. Each sensor node is within the listening range of the other sensor nodes in the network, and the two communicate bidirectionally. Step 2: To address the WWSN time slot allocation problem, an optimal node power model is constructed using the WPT (Wide Power Transfer) technique to collect energy. Step 3: Use MATLAB's CVX toolbox to solve the node power optimization model and obtain the optimal transmit power; Step 4: Determine the optimization objective of the time slot allocation model; Step 5: Based on the optimization objective, iteratively optimize the number of time slots required for node allocation using the particle swarm optimization algorithm.

[0005] Furthermore, the construction of the node power optimal model in step 2 is as follows: Assume there are N sensor nodes in a wearable wireless sensor network. Sensor nodes are in an active state when transmitting data and in a sleep state when not transmitting data. The total energy consumption required for one frame in the wearable wireless sensor network is expressed as the sum of the energy consumption required by the node in its active state during data transmission and the energy consumption required in its sleep state when not transmitting data, i.e.: (1) in, Let be the transmit power of node i during data transmission; Let be the transmission delay of node i during the data transmission process; Let be the power of node i in sleep mode; is the duration that node i is in sleep mode; n is the number of nodes; Assuming that nodes enter a dormant state after transmitting all the data packets they carry; in the optimal node power model, the node's transmit power must be greater than the network's minimum energy consumption threshold and less than the network's maximum transmit power, the sum of the transmit power of all sensor nodes must be less than the upper limit power of all nodes, and the energy consumed by a node in one frame must be less than the energy collected from the environment using WPT technology and the node's energy consumption. The sum of remaining energy after the transmission of the previous frame; under the above constraints, the optimal node power model is as follows: (2) (3) (4) (5) (6) (7) in, , These are the minimum power consumption limit threshold for the network and the maximum transmit power that the network can withstand, respectively. This represents the upper limit power for all nodes; The maximum latency for all nodes during the data transmission process. For nodes i Total energy consumed For nodes i Energy harvested from the environment using WPT technology For nodes Residual energy after transmission; The calculation method is as follows: (8) in, , , , , , Represents voltage. The mutual inductance between the transmitting and receiving coils, where m represents the number of coils. and This is the equivalent resistance of the transmitting and receiving coils. and Let be the equivalent inductance of the transmitting and receiving coils. For transmission efficiency, , The operating angular frequency, ; The calculation method is the sum of the energy consumed by the node in sending data and the energy consumed by the node in receiving data, as shown in the following formula: (9) in, The received power during data transmission at node i. The data reception delay during the data transmission process of node i; node i The energy consumed while asleep is shown below: (10) sensor nodes i The total energy consumed is expressed as: (11) Therefore, the optimal node power model is transformed into the following geometric programming problem: (12).

[0006] Furthermore, in step 4, the optimization objective of the time slot allocation model is the minimum number of time slots required for all nodes to complete transmission, as shown below: (13) (14) (15) (16) in, Equation (14) indicates that the number of time slots used for channel allocation is less than or equal to the number of existing time slots. The ratio of the superframe length to the length of each time slot indicates the maximum number of time slots that can be allocated. Equation (15) indicates that the size of the data packet carried by the node is not greater than the maximum value of the data packet that the time slot can accommodate. Equation (16) indicates that each node will be allocated to one and only one time slot.

[0007] Furthermore, the specific steps of the particle swarm optimization algorithm in step 5 for iteratively optimizing the number of time slots required for node allocation are as follows: Step 5.1: Initialize the parameters in the network, including setting the maximum number of evolutions. T Number of sensor nodes, population size, and individual and population extreme value exchange contraction factors. k Learning factors c1 and c2; Step 5.2: Randomly generate population locations p Randomly generated population initialization speed v And initialize the velocity matrix and initialize the historical best position of each individual particle. and the group's historical best position ; Step 5.3: The fitness function is the objective function value, which is the number of time slots used by each allocation scheme. Calculate the fitness value for each individual. Record the best historical position sequence of an individual; sort the fitness function of each individual in ascending order, and find the optimal fitness value of the population and the corresponding position sequence; Step 5.4: The particle updates its velocity and position; updating the position is equivalent to updating the path sequence; compare the current fitness function value. Compared to its fitness function value during its last evolution If the current fitness function value is better than the fitness function value at the time of the last evolution, then update the individual's historical best position; otherwise, keep it as is. Step 5.5: Compare particles pairwise to obtain the result after the [step 5.5]. T The global optimal position after secondary evolution Then, compared with the previous iteration Compare the results; if the fitness value is better at this point, then update. Otherwise, remain unchanged; Step 5.6: If the optimal fitness value of the population tends to stabilize or reaches the maximum number of iterations, the algorithm ends and the optimal time slot allocation is obtained.

[0008] The beneficial effects of adopting the above technical solution are as follows: The wearable wireless sensor network channel time slot allocation method supporting WPT technology provided by this invention adopts a single-hop star topology. By establishing a global minimum energy consumption model, it optimizes the transmission power of nodes and introduces the concept of time slot sharing by comparing the size of the data packets carried by each node with the maximum remaining capacity of the time slot, ensuring the effective utilization of transmission time slots by nodes. Simultaneously, by setting sensor nodes to a sleep state when not transmitting data, the overall energy consumption of the wearable wireless sensor network is effectively reduced. Furthermore, the time slot allocation of the entire network is optimized, extending the network's lifespan. Attached Figure Description

[0009] Figure 1 This is a schematic diagram of the network topology provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of the superframe structure for the data transmission stage provided in an embodiment of the present invention; Figure 3 A data transmission flowchart provided for an embodiment of the present invention; Figure 4 This is a schematic diagram illustrating the utilization of node time slots according to an embodiment of the present invention. Figure 5 The algorithm optimization process provided in the embodiments of the present invention; Figure 6 A comparison chart of the relationship between the number of nodes and energy consumption provided in an embodiment of the present invention; Figure 7 A comparison chart of the relationship between the number of nodes and latency provided in an embodiment of the present invention. Detailed Implementation

[0010] The specific embodiments of the present invention will be described in further detail below with reference to the accompanying drawings and examples. The following examples are for illustrative purposes only and are not intended to limit the scope of the invention.

[0011] This embodiment first briefly introduces the topology, node model, and superframe composition of wearable wireless sensor networks. Next, a minimum node power model is established, which is transformed into a geometric problem to solve for the optimal transmit power. Then, based on this model, a particle swarm optimization (PSO)-based channel time slot allocation scheme is proposed to optimize the time slot allocation model, and the PSO algorithm is used to solve it. The specific method of this embodiment is described below.

[0012] like Figure 1As shown, this embodiment illustrates its network topology. Wearable wireless sensor networks (WWSNs) employ different topologies to meet varying network requirements in different network scenarios. Commonly used topologies include star and tree topologies. However, the star topology is relatively simple, and can be further divided into single-hop and multi-hop star topologies. Considering the short distance between sensor nodes and coordinator nodes in wearable wireless sensor networks, single-hop transmission is sufficient to meet network requirements. Furthermore, this single-hop transmission method is more direct and requires no routing information. Therefore, this embodiment studies the channel time slot allocation problem of the WWSN based on a single-hop star topology. Sensor nodes send data to the coordinator node via a single-hop method, and the coordinator node collects and manages this physiological data. In this embodiment, it is assumed that a star-topology WWSN consists of one coordinator node and N sensor nodes, each sensor node being within the listening range of the other sensor nodes in the network, and the two communicate bidirectionally.

[0013] like Figure 2 As shown, this embodiment illustrates the data frame structure. This embodiment designs a TDMA (Time Division Multiple Access) scheme based on superframes. During the data transmission phase, the WWSN communicates periodically in units of time frames. Each time frame consists of multiple time slots, with a guard interval inserted between every two consecutive time slots. This is to ensure that time slots do not interfere with each other during continuous transmission. The length of a superframe is... The length of each time slot is expressed as... Superframe structure such as Figure 2 As shown, each node in WWSN will be allocated a time slot, during which the node can communicate with the coordinator node.

[0014] like Figure 3 As shown in the figure, this embodiment illustrates its communication process. The coordinator node broadcasts beacon frames to each sensor node. The beacon frame contains the time slot information allocated by the coordinator node to each sensor node, i.e., the time slot allocation result for each node. Sensor nodes can use this result to determine their own time slot number. This broadcasting method saves latency costs across the entire network. During the beacon period, all nodes open their information receiving modules to receive the beacon frame sent by the coordinator node, completing synchronization. After receiving the beacon frame, the sensor nodes transmit information to the coordinator node in their respective time slots. After completing data reception, the coordinator node replies with an ACK frame to the sensor nodes, thus completing the entire communication process.

[0015] like Figure 4As shown in the diagram, this embodiment illustrates its node time slot utilization. If each time slot is evenly allocated to each node, it's highly likely that a node's actual time slot (including data frames and response frames) will be less than the allocated time slot, even occupying less than one-third of the total time slot length. These nodes with smaller data packets will have unused time slot capacity, meaning their allocated time slots will not be fully utilized. Reducing the time slot length might alleviate this waste, but focusing solely on this is a one-sided approach. It fails to consider the negative impacts of reduced time slot length, neglecting the increased proportion of control information. This situation will still result in reduced time slot utilization in the WWSN. Furthermore, reducing the time slot length in the channel will also reduce network capacity and the amount of data that can be transmitted, contradicting the design intent of this embodiment. Therefore, a suitable time slot allocation scheme is crucial for meeting the data heterogeneity requirements of wearable wireless sensor networks. A new time slot allocation scheme is needed that allows nodes to fill as many time slots as possible with a fixed time slot length.

[0016] This embodiment addresses the aforementioned problems in depth. Combining time slot allocation and time slot sharing, a channel time slot allocation algorithm is designed, allowing two or more sensor nodes to share a single time slot for data transmission, reducing the waste of time slot resources. Furthermore, in practical applications, some nodes in wearable wireless sensor networks need to be implanted within the body, strictly limiting the transmit power of nodes in the network. Therefore, this embodiment first establishes a Network Power Optimization Model (NPOM) to optimize the transmit power of nodes, minimizing the transmit power of a single frame in TDMA access mode. Based on the transmit power optimization, a channel time slot allocation scheme that satisfies the constraints is further found to improve channel utilization, reduce node energy consumption, and optimize the overall network performance.

[0017] The construction of the optimal node power model is as follows: Assume there are N sensor nodes in a wearable wireless sensor network. Sensor nodes are in an active state when transmitting data and in a sleep state when not transmitting data. The total energy consumption required for one frame in the wearable wireless sensor network is expressed as the sum of the energy consumption required by the node in its active state during data transmission and the energy consumption required in its sleep state when not transmitting data, i.e.: (1) in, Let be the transmit power of node i during data transmission; Let be the transmission delay of node i during the data transmission process; Let be the power of node i in sleep mode; Let be the duration that node i is in sleep mode; n is the number of nodes.

[0018] Assuming nodes enter a dormant state after transmitting all their data packets; in the optimal node power model, the node's transmit power must be greater than the network's minimum energy consumption threshold and less than the network's maximum transmit power, and the sum of all sensor node transmit power must be less than the upper limit power of all nodes. To ensure sensor nodes satisfy the energy neutrality theorem, the energy consumed by a node in a frame must be less than the energy collected from the environment using WPT technology and the node's own energy consumption. The sum of remaining energy after the transmission of the previous frame. Under the above constraints, the optimal node power model is as follows: (2) (3) (4) (5) (6) (7) in, , These are the minimum power consumption limit threshold for the network and the maximum transmit power that the network can withstand, respectively. This represents the upper limit power for all nodes; The maximum latency for all nodes during the data transmission process. For nodes i Total energy consumed For nodes i Energy harvested from the environment using WPT technology For nodes The remaining energy after transmission.

[0019] The calculation method is as follows: (8) in, , , , , , Represents voltage. The mutual inductance between the transmitting and receiving coils, where m represents the number of coils. and This is the equivalent resistance of the transmitting and receiving coils. and Let be the equivalent inductance of the transmitting and receiving coils. For transmission efficiency, The operating angular frequency, .

[0020] The calculation method is the sum of the energy consumed by the node in sending data and the energy consumed by the node in receiving data, as shown in the following formula: (9) in, The received power during data transmission at node i. The data reception delay during the data transmission process of node i.

[0021] In the node energy consumption optimization model proposed in this embodiment, the sensor node After transmitting all the data packets it carries, it enters sleep mode. In this mode, the duty cycle is reduced, avoiding additional information exchange and energy consumption, thus improving overall energy efficiency. Node i The energy consumed while asleep is shown below: (10) sensor nodes i The total energy consumed is expressed as: (11) Therefore, the optimal node power model is transformed into the following geometric programming problem: (12) The node optimal power model (NPOM) proposed in this embodiment is a geometric programming problem with an optimal solution. It can be solved using the CVX toolbox in MATLAB to obtain the optimal transmit power.

[0022] The optimization objective of the time slot allocation model is: Based on the above analysis, the optimization objective of the time slot allocation model can be determined as the minimum number of time slots required for all nodes to complete transmission, as shown below: (13) (14) (15) (16) In equation (13) Equation (14) indicates that the number of time slots used for channel allocation is less than or equal to the number of existing time slots. The ratio of the superframe length to the length of each time slot indicates the maximum number of time slots that can be allocated. Equation (15) indicates that the size of the data packet carried by the node is not greater than the maximum value of the data packet that the time slot can accommodate. Equation (16) indicates that each node will be allocated to one and only one time slot.

[0023] Particle Swarm Optimization (PSO) is a nature-inspired swarm metaheuristic algorithm, considered one of the leading swarm intelligence algorithms widely used in hybrid technologies. PSO simulates the social behavior of bird, cattle, and fish swarms, starting with a randomly distributed set of particles (potential solutions) and attempting to improve the solution based on a fitness function. Since its inception, the algorithm has undergone numerous improvements and is now applied in many fields.

[0024] The Particle Swarm Optimization (PSO) algorithm mainly comprises five basic elements: particles, population, fitness function, individual optimal positions, and global optimal positions. A single particle is defined as a potentially feasible solution to the optimization problem in the search space. In each iteration, the population is updated by updating the velocity and position of each individual particle. These updates are based on the individual's optimal position. and the global optimal position Particles possess memory capabilities, remembering their individual optimal position, global optimal position, and velocity found during the search process. In the model, particles are affected by their own position, but... In the model, the particle's position is influenced by the optimal position found by any member of the entire population. In short, This is the largest single particle to date to have found its optimal position. It is the highest level of The globally optimal position found by all particles in the dimensional search space during the search process.

[0025] The particle updates its velocity and position in accordance with the methods given in formulas (17) and (18): (17) (18) In equation (17) c1 and c2 represent inertia weights, ensuring the particle search speed is controlled within a reasonable range. c1 and c2 are particle learning factors; their magnitudes determine the particle's preferred selection. still .

[0026] This embodiment utilizes the particle swarm optimization algorithm to iteratively optimize the number of time slots required for node allocation. The algorithm's search process is as follows: The fitness value is the number of time slots used by each position sequence scheme after maximizing time slot sharing. After initializing the population parameters, the position matrix, shrinkage factor, and learning factor are initialized. Each particle in the population acts as a microparticle in the search space, retaining its original sequence with a certain probability when updating its position and velocity. After the velocity and position are updated, a new position sequence is formed. The fitness value of the new position sequence after each update is calculated, and the arrangement of each node in the sequence is recorded.

[0027] Since the fitness function in this embodiment is the number of time slots, rather than optimizing a specific function value in the traditional sense, and each scheme corresponds to a different positional order of all nodes, with different sequences using different numbers of time slots, the traditional particle swarm optimization (PSO) algorithm's speed update method is adaptively improved by combining the basic principles of the PSO algorithm and the objective function proposed in this embodiment. The main elements and variables of the PSO algorithm are as follows: (1) The particle swarm `pop` represents the sequence of node positions after time-slot sharing, i.e., all solutions. Let the particle swarm population size be... There are , and the number of nodes is Each particle represents a positional sequence of all nodes arranged in order. A 3D integer matrix, where one particle corresponds to one solution. Particle swarm. .

[0028] (2) The fitness value is the objective function value, which is the number of time slots used for each allocation scheme. In this embodiment, it is quantified as a counting process of the number of time slots used. .

[0029] (3) Individual optimal position The optimal value is determined by comparing the fitness values ​​of particles in each iteration and sorting them according to the number of time slots. Record Acquisition The sequence of positions corresponding to the particles.

[0030] (4) Optimal position of the group It is the globally optimal value obtained by comparing the fitness values ​​after the algorithm reaches the maximum number of generations. Record Acquisition The sequence of positions corresponding to the particles.

[0031] The core idea of ​​the particle swarm optimization algorithm is to enable all birds in the population to know the location of food sources, and for particles to pass messages to each other to move closer to the optimal fitness value. In this embodiment, the fitness value is the number of time slots, which is the sequence of nodes corresponding to each particle. The new position sequence is updated using the exchange sequence of velocity records. As can be seen from equation (17), the particle velocity is determined by the previous generation particle velocity, individual velocity-related variables, and population velocity-related variables. The previous generation particle velocity term is determined by the weighting factor. The determination of individual velocity-related variables and group velocity-related variables is based on probability factors. In this embodiment, individual and group corrected velocities are set. The calculation method is as follows: Particles continuously move closer to their historical best position. Indicates whether the current particle is with Sequence swapping is performed, and the criteria for determining the swap are the generated random number and the shrinkage factor. With learning factors The product of the two products is compared and the judgment process is shown in equation (19).

[0032] (19) Similarly, Indicates whether the current particle is with The sequence is swapped, and the judgment process is shown in equation (20).

[0033] (20) Particles continuously approach the optimal solution according to a certain probability based on a learning factor, and the velocity recording sequence is updated accordingly. At this point, the current position is updated according to the correction speed, the fitness value is recorded, and the first generation loop ends.

[0034] In summary, to optimize the number of network time slots, the channel time slot allocation scheme for wearable wireless sensor networks proposed in this embodiment is based on the particle swarm optimization algorithm. The specific steps of the algorithm are as follows: Step 1: Initialize network parameters: Set the maximum number of evolutions. The number of sensor nodes is set to 20, the population size is 5, and the individual and population extreme value exchange shrinkage factor is set. Learning factor and .

[0035] Step 2: Randomly generate population locations Randomly generated population initialization speed And initialize the velocity matrix and initialize the historical best position of each individual particle. and the group's historical best position ; Step 3: The fitness function is the objective function value, which is the number of time slots used by each allocation scheme. Calculate the fitness value for each individual. Record the best historical position sequence of each individual. Sort the fitness function of each individual in ascending order, and find the optimal fitness value of the population and its corresponding position sequence.

[0036] Step 4: The particle updates its velocity and position; updating the position is equivalent to updating the path sequence. The current fitness function value is then compared. Compared to its fitness function value during its last evolution If the current value is better, update the individual's historical best position; otherwise, leave it as is. Step 5: Compare particles pairwise to obtain the results after the [step 1]. The global optimal position after secondary evolution Then, compared with the previous iteration Compare the results; if the fitness value is better at this point, then update. Otherwise, remain unchanged; Step 6: If the optimal fitness value of the population tends to stabilize or the maximum number of iterations is reached, the algorithm ends.

[0037] This embodiment uses the particle swarm optimization algorithm to optimize the objective function. After iteration, it finds a set of node position sequences that minimizes the number of time slots used. The time slot optimization process is as follows: Figure 5 As shown. Figure 5 Taking 10, 20, and 30 sensor nodes as examples, the figure shows the number of time slots required to allocate all nodes before and after optimization using the algorithm proposed in this embodiment, under different numbers of nodes. As shown in the figure above, after iterations of the particle swarm optimization algorithm, the fitness function value in the algorithm, i.e., the number of time slots used for each allocation scheme, changes under different numbers of sensor nodes. All values ​​showed a downward trend compared to before optimization, demonstrating the excellent characteristics of the time slot allocation algorithm proposed in this embodiment in the process of optimizing the objective function.

[0038] like Figure 6As shown, the energy consumption of the particle swarm optimization (PSO)-based channel slot allocation algorithm proposed in this embodiment is lower than that of the other two protocols. This is because the algorithm optimizes the transmission power of the nodes, and the sensor nodes enter a sleep mode after transmitting all the data packets they carry. In this mode, the duty cycle can be reduced, avoiding additional information interaction and energy consumption, thereby improving overall energy efficiency. Sensor nodes in WWSN have multiple categories, and the vital signs of each type of node are completely different, thus exhibiting heterogeneity. In this embodiment, the priority level of nodes is defined according to the urgency of the data they carry. Nodes carrying urgent data are located in the high-priority sequence and have priority in transmission. During the algorithm execution, slots are allocated according to the order of high and low priority nodes, and the concept of slot sharing is used to achieve slot reuse, thereby improving the channel slot utilization rate. This embodiment uses the particle swarm optimization algorithm to optimize the objective function, reducing the number of slots used for allocation. After the algorithm is completed, an optimal node position sequence is found. This sequence uses the fewest slots, thus increasing the number of empty slots in the entire superframe and reducing overall energy consumption.

[0039] like Figure 7 The figure shows the trend of average latency as the number of nodes changes. Because the IEEE 802.15.6 standard MAC protocol uses a backoff mechanism, nodes often cannot access the channel in a timely manner, wasting time and thus incurring higher latency costs. The sleep-wake mechanism MAC protocol adopts a scheduling strategy similar to TDMA, dividing the overall network latency into wake-up latency and transmission latency. The channel allocation algorithm proposed in this embodiment has slightly higher latency compared to the MAC protocol based on the Wake-Up Radio mechanism. This is mainly because the coordinator node sets priorities for the sensor nodes, and the data synchronization between the sensor nodes and the coordinator node takes time.

[0040] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope defined by the claims of the present invention.

Claims

1. A method for allocating channel time slots in a wearable wireless sensor network supporting WPT technology, characterized in that: Includes the following steps: Step 1: Construct the network topology; This study investigates the channel time slot allocation problem of WWSN based on a single-hop star topology. A star topology WWSN consists of a coordinator node and N sensor nodes. Each sensor node is within the listening range of the other sensor nodes in the network, and the two communicate bidirectionally. Step 2: To address the WWSN time slot allocation problem, an optimal node power model is constructed using the WPT (Wide-Point Power Transfer) energy harvesting method. The specific construction of the optimal node power model is as follows: A wearable wireless sensor network contains N sensor nodes. Sensor nodes are in an active state when transmitting data and enter a sleep state when not transmitting data. The total energy consumption required for one frame in a wearable wireless sensor network is represented as the sum of the energy consumption required by the node in its active state during data transmission and the energy consumption required in its sleep state when not transmitting data, i.e.: (1)； in, Let be the transmit power of node i during data transmission; Let be the transmission delay of node i during the data transmission process; Let be the power of node i in sleep mode; is the duration that node i is in sleep mode; n is the number of nodes; Assuming that nodes enter a dormant state after transmitting all the data packets they carry; in the optimal node power model, the node's transmit power must be greater than the network's minimum energy consumption threshold and less than the network's maximum transmit power, the sum of the transmit power of all sensor nodes must be less than the upper limit power of all nodes, and the energy consumed by a node in one frame must be less than the energy collected from the environment using WPT technology and the node's energy consumption. The sum of remaining energy after the transmission of the previous frame; under the above constraints, the optimal node power model is as follows: (2)； (3)； (4)； (5)； (6)； (7)； in, , These are the minimum power consumption limit threshold for the network and the maximum transmit power that the network can withstand, respectively. This represents the upper limit power for all nodes; This is the superframe length, which is the maximum delay across all nodes during the data transmission process. For nodes i Total energy consumed For nodes i Energy harvested from the environment using WPT technology For nodes Residual energy after transmission; The calculation method is as follows: (8)； in, , , , , , Represents voltage. The mutual inductance between the transmitting and receiving coils, where m represents the number of coils. and This is the equivalent resistance of the transmitting and receiving coils. and Let be the equivalent inductance of the transmitting and receiving coils. For transmission efficiency, , The operating angular frequency, ; The calculation method is the sum of the energy consumed by the node in sending data and the energy consumed by the node in receiving data, as shown in the following formula: (9)； in, The received power during data transmission at node i. The data reception delay during the data transmission process of node i; node i The energy consumed while asleep is shown below: (10)； sensor nodes i The total energy consumed is expressed as: (11)； Therefore, the optimal node power model is transformed into the following geometric programming problem: (12)； Step 3: Use MATLAB's CVX toolbox to solve the node power optimization model and obtain the optimal transmit power; Step 4: Determine the optimization objective of the time slot allocation model; Step 5: Based on the optimization objective, iteratively optimize the number of time slots required for node allocation using the particle swarm optimization algorithm.

2. The method for channel time slot allocation in wearable wireless sensor networks supporting WPT technology according to claim 1, characterized in that: In step 4, the optimization objective of the time slot allocation model is the minimum number of time slots required for all nodes to complete transmission, as shown below: (13)； (14)； (15)； (16)； in, Equation (14) indicates that the number of time slots used for channel allocation is less than or equal to the number of existing time slots. The ratio of the superframe length to the length of each time slot indicates the maximum number of time slots that can be allocated. Equation (15) indicates that the size of the data packet carried by the node is not greater than the maximum value of the data packet that the time slot can accommodate. Equation (16) indicates that each node will be allocated to one and only one time slot.

3. The wearable wireless sensor network channel time slot allocation method supporting WPT technology according to claim 2, characterized in that: The specific steps of the particle swarm optimization algorithm in step 5 to iteratively optimize the number of time slots required for node allocation are as follows: Step 5.1: Initialize the parameters in the network, including setting the maximum number of evolutions. T Number of sensor nodes, population size, and individual and population extreme value exchange contraction factors. k Learning factors c1 and c2; Step 5.2: Randomly generate population locations p Randomly generated population initialization speed v And initialize the velocity matrix and initialize the historical best position of each particle. and the group's historical best position ; Step 5.3: The fitness function is the objective function value, which is the number of time slots used by each allocation scheme. Calculate the fitness value for each individual. Record the best historical position sequence of an individual; sort the fitness function of each individual in ascending order, and find the optimal fitness value of the population and the corresponding position sequence; Step 5.4: The particle updates its velocity and position; updating the position is equivalent to updating the path sequence; compare the current fitness function value. Compared to its fitness function value during its last evolution If the current fitness function value is better than the fitness function value at the time of the last evolution, then update the individual's historical best position; otherwise, keep it as is. Step 5.5: Compare particles pairwise to obtain the result after the [step 5.5]. T The global optimal position after secondary evolution Then, compared with the previous iteration Compare the results; if the fitness value is better at this point, then update. Otherwise, remain unchanged; Step 5.6: If the optimal fitness value of the population tends to stabilize or reaches the maximum number of iterations, the algorithm ends and the optimal time slot allocation is obtained.

Images