A Method for Designing Hybrid Wireless Network Paths to Maximize Charging Efficiency

Abstract

This invention discloses a method for designing the shortest path in a hybrid wireless network that maximizes charging efficiency. This method targets a novel hybrid charging network composed of multiple sensor nodes and multiple energy recharging stations in a wireless charging network. Taking into account the random deployment of energy stations and sensor nodes, the method designs a charging efficiency maximization model based on the water injection algorithm and the Transmission Split (TSP) algorithm. Furthermore, considering charging time constraints, the method designs a closed-loop path for maximizing charging efficiency under time constraints based on two strategies: sensor removal and sensor compression. This method maximizes charging efficiency in a rechargeable hybrid sensor network.

Description

Technical Field

[0001] The present invention belongs to the technical field of wireless charging, and in particular relates to a hybrid wireless network path design method for maximizing charging efficiency. Background Art

[0002] With the continuous development of communication technology and the communication industry, radio frequency (RF) has become ubiquitous in the environment. The collection and scheduling of RF to provide efficient energy supply for IoT systems has become a current research hotspot. Traditional IoT device nodes are generally powered by batteries, but their energy storage is limited and their service life is short. Therefore, the energy supply problem has become a key bottleneck for the rapid development of IoT applications. Nowadays, wireless charging technology has been widely studied and applied, such as rechargeable smartphones, wearable devices, access authentication and urban sensing. A popular direction for solving energy supply problems in the field of wireless charging technology is path scheduling in wireless rechargeable sensor networks (WRSN). WRSN alleviates the energy limitation problem of traditional wireless sensor networks through mobile chargers. It is usually implemented by one or more mobile chargers carrying energy from energy supply stations, finding an effective path in the network to traverse sensor nodes and charge them.

[0003] In existing research, when a mobile charger runs low on energy, only one or a few refueling stations are used to refuel the mobile charger. Even solutions that utilize multiple refueling stations tend to pre-determine their deployment locations, ignoring the possibility that refueling stations can be randomly deployed, just like sensor nodes. In this case, the refueling station can be considered a special sensor node.

[0004] With the widespread deployment of wireless access points (WAPs) such as WiFi and 5G base stations, the number of energy recharge stations can be as numerous as the number of sensors. A large number of energy recharge stations and WRSNs together form another new type of rechargeable hybrid sensor network (h-WRSN). The sensor nodes in this hybrid network can be wirelessly charged using mobile chargers, and the mobile chargers can also be wirelessly charged by the energy recharge stations in the h-WRSN.

[0005] Furthermore, in conventional charging processes, the distance between the mobile charger and the device is often ignored; in fact, it has a significant impact on charging efficiency and path design. In h-WRSN, the mobile charger can be charged by the RF transmitter in the energy recharging station while on the move and collect available energy to support its continued operation without having to return to a base station or warehouse for recharging.

[0006] A mobile charger starts at the warehouse, collects data from all sensors, and charges some of them. During the charging process, if the mobile charger finds itself running low on energy, it draws energy from an energy recharging station. This process repeats until all sensor nodes have been visited. The mobile charger eventually returns to the warehouse. This process is similar to the route-finding process of a traveling salesman, and can be compared to the quota traveling salesman problem.

[0007] The water injection algorithm is a widely used optimal power allocation strategy. Applying it to sensor energy allocation maximizes efficiency. The TSP algorithm was originally developed to solve the problem of "given a series of cities and the distances between each pair of cities, find the shortest closed-loop path for a traveling salesman to visit each city once and return to the starting city." The charging process in this paper is similar to the traveling salesman problem, so the TSP algorithm can be used to solve the shortest closed-loop charging path. Given the time constraints inherent in real-world problems, a hybrid wireless network path that maximizes charging efficiency was designed using two strategies: sensor removal and sensor compression, to address this time constraint. Summary of the Invention

[0008] The purpose of the present invention is to propose a hybrid wireless network path design method for maximizing charging efficiency. For a new hybrid charging network composed of multiple sensor nodes and multiple energy supply stations in a wireless charging network, the random deployment of energy stations and sensor nodes is taken into consideration. A charging efficiency maximization model is designed based on the water injection algorithm and the TSP algorithm. Taking into account the charging time constraint, a closed-loop path method for maximizing charging efficiency under time constraints is designed based on the two strategies of sensor removal and sensor compression, thereby maximizing the charging efficiency of the rechargeable hybrid sensor network.

[0009] The present invention specifically comprises the following steps:

[0010] Step 1: Construct a graph G based on a set of rechargeable sensors V and a set of energy recharging stations U. Given a node v0 in the graph G representing the starting point of the mobile charger, the mobile charger starts from v0 and uses the TSP algorithm to find a shortest closed-loop path P that traverses all sensor nodes and eventually returns to the starting point v0. Fixed energy recharging stations are placed around the path.

[0011] Step 2: Take each sensor node of path P as the center of the circle and set the given radius ri The set of connection edges between any energy refueling station and sensor nodes within i is E1, and the set of connection edges between any sensor nodes is E2, obtaining the graph G(E1, E2, V, U);

[0012] Step 3. Let the sensor v i with a radius r i The set of energy refueling stations within the range is N i , that is, a set of energy refueling station nodes within the radius r i is established for each sensor;

[0013] Step 4. For any two sensors v i and v j on the path P, for the connection edge e ij between them (i < j), find the nearest energy refueling station u k , u k is called the nearest energy refueling station.

[0014] Calculate the real-time energy consumption of the mobile charger through the energy loss formula. When the mobile charger needs to replenish energy on a certain edge of the path P, insert the nearest energy refueling station of this edge into the path P;

[0015] Step 5. When the mobile charger reaches any energy refueling station on the path P, the energy refueling station will charge it fully;

[0016] Step 6. When the mobile charger charges the sensor nodes, all sensor nodes between the two nearest energy refueling stations are allocated energy according to the water filling algorithm; first, sort the sensor sequence in non-decreasing order of the remaining energy, then allocate energy to the first sensor in the non-decreasing sequence until its remaining energy is equal to that of the second sensor in the non-decreasing sequence. Finally, consider the first and second sensors with equal remaining energy as a whole in the subsequent energy allocation, and repeat the above steps until all nodes are allocated or the available energy is exhausted; finally, output an optimal charging benefit closed-loop path P′ that includes all sensors and some energy refueling stations.

[0017] For the above method of designing the path of the hybrid wireless network to maximize the charging benefit, there are the following time constraints, which specifically include the following steps:

[0018] Step 1. Establish a set of energy refueling station nodes within a given radius for each sensor in the graph G(E1, E2, V, U);

[0019] Step 2: The mobile charger traverses all sensors in the graph G(E1, E2, V, U) from a given starting point and finds an optimal charging efficiency closed-loop path P′ that includes all sensors and some energy refueling stations;

[0020] Step 3: Calculate the time t(P′) consumed by path P′;

[0021] Step 4: Charge the sensors on path P' at the first charging ratio. The non-decreasing sorting generates a sequence S1. Each element of the S1 sequence contains two members: sensor node v i and its first charge ratio

[0022] Step 5: Assign the sensors on path P′ to the same sensor group g according to the same residual energy. m , and then these sensor groups are charged according to their second charging ratio Non-decreasing sorting generates sequence S2, each element of S2 contains two members: sensor group g m and

[0023] Step 6. If t(P′) <t s , t s If the time threshold is given, the path P′ is returned directly; a path including all sensors and some energy supply stations is obtained and can be s A closed-loop path to maximize charging benefits under constraints;

[0024] Otherwise, compare the first charging ratio of the first element of sequence S1 and sequence S2 in turn and the second charge ratio if Greater than or equal to Then calculate the time it takes in S2 to charge the first sensor group to the same energy as the second sensor group And order n is the number of sensors in the first sensor group, then delete the first element in the S2 sequence; if Less than Remove the first sensor S1 on path P′, obtain a new path P″, calculate t(P″), update t(P′) = t(P″) and P′ = P″; repeat step 6 until it returns to path P′.

[0025] The present invention takes into account the hybrid charging network composed of multiple energy supply stations and multiple sensor nodes, and designs a charging benefit maximization method with the goal of minimizing the total cost of mobile charging and maximizing the total charging benefit. The TSP algorithm can be used to quickly solve a shortest closed-loop path that traverses all sensor nodes, and then the corresponding energy supply station is inserted into this shortest path according to the power consumption of the mobile charger on the path. In this way, the complex hybrid node path solving problem is simplified to a problem of adding and deleting path nodes. At the same time, the water injection algorithm is used to more reasonably allocate energy, so that the charging of the sensor can achieve maximum benefit. In real life, the entire path traversal and charging process takes time, so it is necessary to consider time constraints. The present invention designs two strategies to alleviate the charging restrictions under time constraints, and uses different strategies for different charging situations, so that the designed method has more practical application significance. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 Schematic diagram of the design of the charging benefit maximization model.

[0027] Figure 2 Schematic diagram of the charging benefit maximization model with time constraints. DETAILED DESCRIPTION

[0028] The present invention will be further described below with reference to the accompanying drawings.

[0029] In the two-dimensional space A, a group of m energy supply stations are randomly deployed, denoted as U = {u j ,j=1,…,m}, a group of n sensor nodes, denoted as V={v i ,i=1,…,n}, and a mobile charger. The mobile charger has a capacity of B u Each sensor node also has a rechargeable battery with a capacity of B s , the initial residual energy is c i Rechargeable battery. Energy supply station u j and sensor node v i The position after deployment is fixed.

[0030] In the entire system, the energy supply station can send wireless radio frequency energy to the mobile charger, the sensor node can obtain wireless radio frequency energy from the mobile charger, and the energy supply station and the sensor node can communicate with the mobile charger.

[0031] The energy transfer formula of the energy recharge station is described by the Friis free space equation:

[0032] Received power

[0033] θr and θ t Represents the receiving power and transmitting power respectively, G s is the transmitting antenna gain of the energy supply station, G r is the receiving antenna gain of the mobile charger, λ is the wavelength, and d is the distance between the energy supply station and the mobile charger.

[0034] The final received power θ r Expressed as,

[0035] Among them L p is the polarization loss, η is the rectifier benefit, and β is the compensation parameter of the Friis free space equation in the case of antenna near-field transmission.

[0036] The receiving power of the energy receiving device in the wireless charging process is simplified to a charging efficiency attenuation model that is only related to the charging distance. Where α is a parameter determined by the hardware of the mobile charger and sensor node, such as antenna gain, antenna polarization loss, etc. The values of α and β are constants. Given d maximum value d max , when the sensor node v i The distance from the mobile charger is greater than d max , sensor node v i Cannot draw power from mobile charger.

[0037] The mobile charger charges the sensor and uses the charging benefit function r(·) to evaluate the charging benefit. The charging benefit function is a non-increasing submodular function of the remaining energy of the sensor, that is, r(x+Δ)-r(x)≥r(y+Δ)-r(y)ifx≤y;

[0038] x and y represent the remaining energy of the two sensor nodes respectively, Δ represents the charge amount of a charging process, and the maximum charge of the sensor is B s , then (x+Δ)≤B s and (y+Δ)≤B s . For sensor v i The benefit gained from charging is the marginal benefit r i =r(x+Δ)-r(x).

[0039] Given a position, if the position is consistent with the sensor v i The distance between them is less than d max , that is, the sensor v i Cover. Given a moving path P, let Indicates that the sensor v on path P i Covered segment, S p Denotes the set of sensors covering any segment of path P. Let θ rThe power of energy collected, θ t is the power of the transmitted energy, that is, the energy collected and transmitted per unit time. Let h represent the energy collected by the mobile charger, which depends on the distance between the mobile charger and the sensor, and the duration of maintaining this position. For time t, the distance is a function of time, denoted by d(t), so the energy collected on path P is

[0040] where d 0i (t) is the mobile charger and sensor v at time t i The distance between each path segment of path P The energy collected by the mobile charger is the duration dt and the distance d 0i (t). Assume that the mobile charger moves at a constant speed, let P represent the path of the mobile charger, and each path P is a series of continuous geometric positions in two-dimensional space. Let θ m is the power consumption per unit distance of the mobile charger, and the energy consumed on path P is p =θ m l(p), where l(p) represents the length of path P.

[0041] The time consumed by the entire system is mainly composed of three parts: the first part is the time it takes for the mobile charger to charge the sensor nodes or to charge itself; the second part is the time it takes for the mobile charger to communicate with the sensor nodes to collect the environmental information they perceive. Due to the fast data transmission, this time can be included in the first part. The third part is the time it takes for the mobile charger to visit all sensor nodes and the energy supply station.

[0042] Assuming the mobile charger keeps moving at a constant speed, the time loss is

[0043] Where α is a constant determined by the specific charging rate, To charge energy, Indicates the charging time; β is also a constant, which is determined by the specific average moving speed, d t is the path length of the movement, βd t It means that the movement takes time.

[0044] like Figure 1 As shown, a hybrid wireless network path design method for maximizing charging efficiency includes the following steps:

[0045] Step 1: Construct a graph G based on the set V of rechargeable sensors and the set U of energy supply stations. Given that the node v0 of the graph G represents the starting position of the mobile charger, the mobile charger starts from v0, uses the TSP algorithm to find the shortest closed-loop path P that traverses all sensor nodes and finally returns to the starting point v0, and places fixed-deployed energy supply stations around the path.

[0046] Step 2: Take each sensor node on the path P as the center, and set the given radius r i The set of edges connecting any energy supply station within the range and sensor nodes is E1, and the set of edges connecting any sensor nodes is E2, obtaining the graph G(E1, E2, V, U).

[0047] Step 3: Let the set of energy supply stations within the radius r of the sensor v i be N i , that is, establish a set of energy supply station nodes within the radius r for each sensor. i i

[0048] Step 4: For any two sensors v i and v j on the path P, for the connecting edge e ij (i < j), find the nearest energy supply station u k , u k is called the nearest energy supply station, and its definition is:

[0049] Calculate the real-time energy consumption of the mobile charger through the energy loss formula. When the mobile charger needs to replenish energy on a certain edge of the path p, insert the nearest energy supply station of this edge into the path p.

[0050] Step 5: When the mobile charger arrives at any energy supply station on the path P, the energy supply station will charge it fully.

[0051] Step 6: When the mobile charger charges the sensor nodes, all the sensor nodes between the two nearest energy supply stations are allocated energy according to the water filling algorithm. First, sort the sensor sequence in non-decreasing order of the remaining energy, then allocate the energy to the first sensor in the non-decreasing sequence until its remaining energy is equal to that of the second sensor in the non-decreasing sequence. Finally, regard the first and second sensors with equal remaining energy as a whole in the next energy allocation, and repeat the above steps until all nodes are allocated or the available energy is exhausted. Finally, output a closed-loop path P′ that includes all sensors and some energy supply stations.

[0052] In order to maximize the compression time with minimal reduction in charging utility, sensor removal and sensor compression strategies are adopted.

[0053] Sensor removal reduces the charging energy of certain sensors to save charging time. The goal of sensor compression is also to minimize the loss of charging benefits while maximizing time savings.

[0054] Remove some sensors from the charging path to reduce the charging time. In order to minimize the reduction in system charging efficiency and maximize time savings after removal, define the first charging ratio. Sensor v i Rewards for charging i , consumption time Δt i , then:

[0055] Sort the sensors on path P′ in non-decreasing order by their remaining energy, and divide these sensors into multiple groups, with the remaining energy of the sensors in each group being equal. Then, sort these sensor groups in non-decreasing order by their remaining energy. Define the second charging ratio Sensor Group g m Rewards for charging gm , consumption time Δt m , then:

[0056] Given a time threshold t s ,like Figure 2 As shown in Figure 2, the specific steps of the charging benefit maximization path design method under time constraints are as follows:

[0057] Step 1: For each sensor in the graph G(E1, E2, V, U), establish a set of energy refueling station nodes within a given radius.

[0058] Step 2: Move the mobile charger from a given starting point to traverse all sensors in the graph G(E1, E2, V, U) and find an optimal charging benefit closed-loop path P′ that includes all sensors and some energy refueling stations.

[0059] Step 3: Calculate the time t(P′) consumed by the path P′.

[0060] Step 4: Charge the sensors on path P' at the first charging ratio. The non-decreasing sorting generates a sequence S1. Each element of the S1 sequence contains two members: sensor node v i and its first charging efficiency ratio

[0061] Step 5: Assign the sensors on path P′ to the same sensor group g according to the same residual energy. m, and then these sensor groups are charged according to their second efficiency ratio Non-decreasing sorting generates sequence S2, each element of S2 contains two members: sensor group g m and

[0062] Step 6. If t(P′) <t s , t s For a given time threshold, we directly return to path P′ and obtain a path that includes all sensors and some energy supply stations and can reach the target at time t s A closed-loop path to achieve maximum charging benefit under constraints.

[0063] Otherwise, compare the first charging ratio of the first element of sequence S1 and sequence S2 in turn and the second charge ratio if Greater than or equal to Then calculate the time it takes in S2 to charge the first sensor group to the same energy as the second sensor group And order n is the number of sensors in the first sensor group, then delete the first element in the S2 sequence; if Less than Remove the first sensor S1 on the path P′ to obtain a new path P″, calculate t(P″), update t(P′)=t(P″) and P′=P″. Repeat step 6 until it returns to the path P′.

Claims

1. A hybrid wireless network path design method for maximizing charging efficiency, characterized by: The specific steps include: Step 1: Construct a graph G based on a set of rechargeable sensors V and a set of energy recharging stations U. Given a node v0 in the graph G representing the starting point of the mobile charger, the mobile charger starts from v0 and uses the TSP algorithm to find a shortest closed-loop path P that traverses all sensor nodes and eventually returns to the starting point v0. Fixed energy recharging stations are placed around the path. Step 2: Take each sensor node of path P as the center of the circle and set the given radius r i The set of connecting edges between any energy supply station and sensor nodes is E1, and the set of connecting edges between any sensor nodes is E2, and the graph G(E1, E2, V, U) is obtained; Step 3: Set sensor v i Radius r i The number of energy supply stations within the range is N i , that is, for each sensor, a radius r is established i The set of energy supply station nodes within range; Step 4. For any two sensors \(v\) i and \(v\) j in the path \(P\), for the connecting edge \(e\) ij between them (\(i < j\)), find the nearest energy replenishment station \(u\) k . \(u\) k is called the nearest energy replenishment station. The real-time energy consumption of the mobile charger is calculated using the energy loss formula. When the mobile charger needs to replenish energy on a certain edge of path P, the nearest energy supply station on this edge is inserted into path P. Step 5: When the mobile charger reaches any energy refueling station on path P, the energy refueling station will recharge it with energy; Step 6. When the mobile charger charges the sensor node, all sensor nodes between the two closest energy recharge stations are allocated energy according to the water injection algorithm. First, the sensor sequence is sorted in non-decreasing order according to the remaining energy. Then, energy is allocated to the first sensor in the non-decreasing order until its remaining energy equals the second sensor in the non-decreasing order. Finally, the first and second sensors with equal remaining energy are treated as a whole in the subsequent energy allocation. The above steps are repeated until all nodes are allocated or the allocable energy is exhausted. Finally, an optimal charging efficiency closed-loop path P' that includes all sensors and some energy recharge stations is output.

2. The hybrid wireless network path design method for maximizing charging efficiency according to claim 1 is characterized by: There are the following time constraints, including the following steps: Step 1: For each sensor in the graph G(E1, E2, V, U), establish a set of energy refueling station nodes within a given radius. Step 2: Move the mobile charger from a given starting point to all sensors in the graph G(E1, E2, V, U) and find an optimal charging efficiency closed-loop path P' that includes all sensors and some energy refueling stations. Step 3: Calculate the time t(P') consumed by the path P'; Step 4: Charge the sensors on path P' at the first charging ratio. The non-decreasing sorting generates a sequence S1. Each element of the S1 sequence contains two members: sensor node v i and its first charge ratio Define the first charge ratio Sensor v i Rewards for charging i , consumption time Δt i , then: Step 5: Assign the sensors on path P' to the same sensor group g according to the same residual energy. m , and then these sensor groups are charged according to their second charging ratio Non-decreasing sorting generates sequence S2, each element of S2 contains two members: sensor group g m and Define the second charge ratio Sensor Group g m Rewards for charging Consumption time Δt m , then: Step 6. If t(P') <t s , t s For a given time threshold, path P' is directly returned to obtain a closed-loop path that includes all sensors and some energy supply stations and can achieve the maximum charging benefit under the time constraint of t5; Otherwise, compare the first charging ratio of the first element of sequence S1 and sequence S2 in turn and the second charge ratio if Greater than or equal to Then calculate the time it takes in S2 to charge the first sensor group to the same energy as the second sensor group And order n is the number of sensors in the first sensor group, then delete the first element in the S2 sequence; if Less than Remove the first sensor of S1 on path P' to obtain a new path P", calculate t(P"), and update t(P') = t(P") and P' = P". Repeat step 6 until it returns to path P'.

Images