Wireless sensor network charging method with cooperation of multiple charging vehicles

Abstract

The invention belongs to a wireless sensor network charging method, and particularly relates to a wireless sensor network charging method with cooperation of multiple charging vehicles. The method comprises the following steps: step 1, a sensor uniform distribution stage; the sensors in the monitoring area are uniformly positioned to balance the space monitoring quality; 2, determining the number and the positions of the charging stations; the monitoring area is divided into m sub-areas, and a charging vehicle is arranged in each sub-area; step 2, dividing each sub-region into k grids, and deploying a charging station in each grid; and step 3, a batch charging stage. According to the invention, the monitoring area is divided into a plurality of sub-areas, each sub-area is provided with a charging vehicle, each sub-area is divided into a plurality of grids, each grid is internally provided with a charging station, and the charging vehicle moves in the corresponding sub-area to charge the sensor in each grid, so that the charging efficiency is improved on the premise of continuing the life cycle of the sensor, and the charging efficiency is improved. The monitoring quality is ensured.

Description

Technical Field

[0001] The invention relates to a wireless sensor network charging method, in particular to a wireless sensor network charging method in which multiple charging vehicles cooperate. Background Art

[0002] Wireless sensor networks (WSNs) are composed of multiple sensor nodes and have a wide range of applications, such as smart homes and environmental monitoring. However, since sensors are usually powered by batteries, their life cycle is limited, which directly affects the continuous operation of wireless sensor networks. This limitation has become a bottleneck for the further development of wireless sensor networks. Therefore, in recent years, research on charging issues in wireless sensor networks has received increasing attention, among which charging sensor nodes through mobile charging vehicles (MCVs) has become a research hotspot.

[0003] At present, the research on charging vehicles is roughly divided into two categories: single-charging vehicle charging and multi-charging vehicle charging. Single-charging vehicles are further divided into periodic charging and demand-based charging.

[0004] In periodic charging, the charging vehicle will charge the sensors on the path according to the planned charging path at certain time intervals. The periodic charging method may cause an imbalance in the number of sensors and charging opportunities passed by the charging vehicle because the charging path is fixed. Some sensors may be charged frequently, while others are rarely charged. This imbalance may reduce the overall charging efficiency. At the same time, the charging interval is also fixed, which may cause some sensors to have sufficient power when charging, while other sensors are low in power but fail to be charged in time, and thus run out of power, affecting the monitoring quality. For this reason, some scholars have proposed charging on demand.

[0005] On-demand charging means that the charging vehicle plans the charging route according to the charging needs of the sensor. The charging order is determined based on factors such as the remaining power of the sensor, its importance, and its distance from the charging vehicle. The charging vehicle usually plans the shortest route to minimize the time and energy consumption during the charging process. However, when the sensor nodes are densely distributed, a single charging vehicle still cannot meet the charging needs of all nodes, so research on multi-charging vehicle charging technology has emerged.

[0006] In the multi-charging vehicle charging mode, the monitoring area is usually divided into several sub-areas, and each charging vehicle is responsible for the charging task of one sub-area. However, due to the random deployment of sensor nodes, sensor nodes in some areas are often dense, while sensor nodes in other areas are sparse. This uneven node distribution may lead to an imbalance in the workload of charging vehicles: in areas with dense sensor nodes, the charging capacity of charging vehicles is insufficient; while in areas with sparse sensor nodes, the charging capacity of charging vehicles may be excessive. In addition, the lack of cooperation between charging vehicles will also lead to low overall charging efficiency. Some sensor nodes may even stop working due to the inability to charge in time, which in turn affects the execution of monitoring tasks. Summary of the invention

[0007] The purpose of the present invention is to solve the problem that in the existing charging mode of multiple charging vehicles, due to the random deployment of sensor nodes, the sensor nodes in some areas are often dense, while the sensor nodes in other areas are sparse, which may lead to the problem of unbalanced workload of the charging vehicles. At the same time, the lack of cooperation between charging vehicles will also lead to low overall charging efficiency. Some sensor nodes may even stop working due to the inability to charge in time, thereby affecting the execution of monitoring tasks. A wireless sensor network charging method with cooperation of multiple charging vehicles is provided.

[0008] To achieve the above purpose, the technical solution adopted by the present invention is:

[0009] A wireless sensor network charging method for multiple charging vehicles is characterized in that it includes the following steps:

[0010] Step 1: Sensor uniform distribution stage

[0011] Evenly distribute sensors within the monitoring area to balance the spatial monitoring quality;

[0012] Step 2: Determine the number and location of charging stations

[0013] Step 2.1, divide the monitoring area into m sub-areas, and set a charging vehicle in each sub-area; m is an integer greater than or equal to 1;

[0014] Step 2.2: Divide each sub-area into k grids, deploy a charging station in each grid, and use the following formula to determine the number of charging stations k:

[0015] And k is an integer;

[0016] Where: ε is a constant greater than 10; is the power consumption rate of the sensor per unit distance; L and M are the length and width of the monitoring area respectively; is the total energy of the sensor; l is the charging efficiency of the charging vehicle; is the total energy of the charging vehicle; D is the average distance between two adjacent charging stations; The power consumption rate per unit distance of the charging vehicle;

[0017] Step 2.3, the mobile sensors in the k grids in each sub-area are grouped as k clusters, the center of gravity of each cluster is determined, and the center of gravity of each cluster is set as the final setting location of the charging station, and the center of gravity of the cluster close to the center of the monitoring area is set as the final setting location of the sensor network base station;

[0018] Step 3: Bulk charging stage

[0019] According to the location of the charging station in each sub-area, the shortest charging path for the charging vehicle in each sub-area is constructed; first, the charging time of a single sensor is calculated, and then the starting charging time of each charging station is calculated based on this, and it is broadcast to all sensors in the corresponding grid; secondly, the average working time and charging interval of the sensor is calculated, and the number of charging batches of the sensors in the corresponding grid is calculated based on the average working time and charging interval; finally, the working frequency of the sensor is calculated and adjusted to ensure that the sensor can arrive at the charging station on time for charging before the battery is exhausted; the sensor starts working, and the charging vehicle arrives at each charging station in turn according to the constructed shortest charging path. After the charging vehicle arrives, the sensors in the corresponding grid of each charging station move to the corresponding charging station according to the calculated batch number to be charged by the charging vehicle, and after charging is completed, they return to their original position to re-execute the monitoring task.

[0020] Furthermore, step 1 is specifically as follows:

[0021] By using the method of repulsion and attraction, two sensors that are close to each other are pulled apart, and two sensors that are far away are pulled together, so that the distance between each sensor and its neighboring nodes meets the set conditions, thereby achieving uniform positioning of sensors in the monitoring area to balance the quality of spatial monitoring;

[0022] The neighboring nodes refer to all other sensors within the communication radius of each sensor.

[0023] Furthermore, we define sensor s x and sensors y For any two adjacent sensors in the monitoring area, V x Indicates sensor x The Voronoi cell, V y Indicates sensor y The Voronoi cells, and Respectively represent V x and V yThe area of ​​​​the sensor s is represented by σ, and σ represents the set threshold; the Voronoi unit is x A closed geometric figure enclosed by the perpendicular bisectors of the lines connecting its adjacent nodes;

[0024] Then, sensor s x and sensors y The distance between them meets the following setting conditions:

[0025]

[0026] Where: N x For sensors x A collection of neighboring nodes.

[0027] Furthermore, in step 2.1, the areas of the m sub-regions are equal; in step 2.2, the areas of the k grids are equal; in step 2.3, a K-means clustering algorithm is applied to the sensors in each sub-region to determine the center of gravity of each cluster.

[0028] Furthermore, in step 3, the shortest charging path for the charging vehicle in each sub-area is constructed based on the TSP algorithm.

[0029] Further, define, Indicates the i-th charging vehicle in the i-th sub-area Charging stations, at charging stations There are u sensors in the corresponding grid, which are recorded as

[0030] Then in step 3, the sensor is calculated using the following formula Charging time

[0031]

[0032] Where: Indicates sensor The remaining energy used to perform monitoring tasks; Indicates sensor Monitoring power consumption rate; Indicates sensor from Move to Energy consumed; Indicates sensor location; Indicates charging station The position of; d(·,·) represents the Euclidean distance between two positions.

[0033] Furthermore, in step 3, the charging start time of the first charging station in the i-th sub-area is taken as the starting point, and the charging start time is calculated using the following formula:

[0034]

[0035] Start charging time at the first charging station Based on this, the following formula is used to calculate the start charging time of the remaining charging stations:

[0036]

[0037] Where: Indicates charging station The start time of charging; Indicates the start time of charging at other charging stations; Indicates the charging rate of the charging vehicle; v M Indicates the moving speed of the charging vehicle.

[0038] Furthermore, in step 3, the batch number is calculated using the following formula:

[0039]

[0040] Where: Indicates charging station The next charging start time; average() represents the average value function; Indicates charging station Corresponding sensor working hours.

[0041] Furthermore, define s i Represents a sensor;

[0042] Then in step 3, the sensor s i Working hours and operating frequency The calculation is done using the following formula:

[0043]

[0044] Where: Indicates sensor i The number of monitoring tasks; Indicates sensor i Energy consumption of executing a monitoring task; v s Indicates sensor i Movement speed; Indicates sensor i The remaining power; iIndicates sensor i The operating frequency; Indicates sensor i Monitoring power consumption rate; Indicates sensor i location.

[0045] The beneficial effects of the present invention are:

[0046] The present invention divides the monitoring area into several sub-areas, sets a charging vehicle in each sub-area, and then divides each sub-area into several grids. A charging station is set in each grid. The charging vehicle moves in the corresponding sub-area to charge the sensors in each grid, thereby improving the charging efficiency and ensuring the monitoring quality under the premise of perpetuating the life cycle of the sensor. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 is a schematic diagram of the principle of a monitoring network in an embodiment of the present invention;

[0048] Figure 2 Schematic diagram of the structure principle of the Voronoi unit in an embodiment of the present invention;

[0049] Figure 3 is a schematic diagram of the principle of using attraction to pull two sensors closer together in an embodiment of the present invention;

[0050] Figure 4 is a schematic diagram of the principle of using repulsive force to pull two sensors apart in an embodiment of the present invention;

[0051] Figure 5 The sensor s in the embodiment of the present invention x Force analysis diagram of

[0052] Figure 6 The sensor s in the embodiment of the present invention x Schematic diagram of the resultant force. DETAILED DESCRIPTION

[0053] To make the purpose, advantages and features of the present invention more clear, the wireless sensor network charging method for multiple charging vehicles in cooperation with the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. The advantages and features of the present invention will become clearer according to the following specific implementation methods.

[0054] A wireless sensor network charging solution based on multi-charging vehicle collaboration in this embodiment dynamically adjusts the distribution of charging tasks through cooperation between charging vehicles to cope with the random distribution of sensor nodes and the imbalance of workload, thereby ensuring the continuation of the life cycle of sensor nodes and maintaining a high network monitoring quality, including spatial monitoring quality and temporal monitoring quality.

[0055] Network Model

[0056] This embodiment sets a two-dimensional monitoring area R with a size of L×M, where L and M are the length and width of the monitoring area R respectively.

[0057] In the monitoring area, the base station has two roles: 1) charging the charging vehicle; 2) storing the data collected by the sensor.

[0058] Assume that n mobile sensors are randomly deployed in the monitoring area R, denoted by S = {s1, s2, …, s n}, and assumes that all sensors are powered by batteries. This embodiment uses a charging vehicle to charge the sensors, aiming to maintain the continuous service life of the sensors and improve the monitoring quality. This embodiment will calculate the number and location of charging stations, referred to as RS. According to the number of charging stations RS, the entire area will be divided into several grids, and a charging station RS will be set up in each grid. In each grid, the sensor and a charging vehicle will move to the charging station RS at the agreed time, and the charging vehicle will charge the sensor and collect its perception data.

[0059] The network scenario considered is as follows Figure 1 As shown. The monitoring area R is divided into six grids. The base station charges the sensors in the central grid, while the other five charging vehicles are responsible for charging the sensors in the other five grids. Figure 1 In the figure, the blue line represents the charging path of the charging vehicle. Two charging vehicles are charging the sensors in charging stations RS1 and RS2, while the other three charging vehicles are heading to the next charging station RS to provide charging services.

[0060] Monitoring quality model

[0061] The quality of monitoring is determined by the accuracy of the monitoring data collected in the monitoring area R. Data accuracy refers to the difference between the reported data and the actual data. Data accuracy can be measured in both spatial and temporal terms. For example, multiple sensors are deployed in a monitoring area R to detect the concentration of PM2.5. If a user wants to query the PM2.5 at location a, but there is no sensor at a. Assume that sensor s is deployed at location b and is the closest sensor to the query location a. The monitoring system will report the data detected by sensor s. If the distance between locations a and b is short, the spatial accuracy is high. This is because the reported data and the actual data are very close due to the short distance between locations a and b. However, if the distance between locations a and b is long, the detected data and the actual data may differ greatly. In this case, the spatial accuracy is low. Therefore, the spatial accuracy mainly depends on the distance between the user's query location and the nearest sensor location.

[0062] Similarly, consider location b. If the user wants to query the PM2.5 at location b at time t, but sensor s is not monitoring at time t. Assume that sensor s works at time t1, which is closest to t. The monitoring system will report the sensed data at time t1. If the time difference between t and t1 is small, the time accuracy is high. Otherwise, the time accuracy is low. This is because the closeness of the reported data to the actual data is proportional to the time difference between t and t1.

[0063] Next, we propose an accuracy measurement model to measure the monitoring quality from both spatial and temporal aspects. s Indicates sensor i The monitoring radius is is a Boolean variable indicating whether the point α∈R falls on the sensor s i If α falls within the range of s i Within the perception range, is 1; otherwise, is 0.

[0064]

[0065] Let d(α,s i ) represents the point α and the sensor s i The Euclidean distance between is a Boolean variable, indicating sensor s i Whether to perform the monitoring task at time t. If s i If monitoring is performed at time t, is 1; otherwise, is 0.

[0066] Let θ α，t represents the spatial and temporal monitoring quality of α at time t. If α is at sensor s i within the range of perception And i Monitoring at time t Then θ α，t is 1; if point α is not on sensor s i within the range of perception, but s i Monitoring at time t, θ α，t Then

[0067]

[0068] Assume κ spa represents the spatial monitoring mass loss rate per unit distance. Assume that point α is at sensor s i within the range of perception, and s i It does not monitor at time t. Instead, it performs sensing at time t1, which is closest to time t.α，t The value of can be obtained by:

[0069]

[0070] Assume κ tem represents the time monitoring mass loss rate per unit time, and d(t,t1) represents the time difference between t and t1. Assume that point α is at sensor s i If θ is outside the range of perception, α，t The values ​​are:

[0071]

[0072] Let Q represent the data monitoring quality of all monitoring areas:

[0073]

[0074] The goal of this invention is to find a charging method to maximize the quality of data monitoring. Indicates the best charging method, represents a set of charging methods, and expression (6) represents the objective function of this embodiment.

[0075]

[0076] Method design

[0077] To achieve the above objectives, this embodiment proposes a wireless sensor network charging method for multi-charging vehicle collaboration, aiming to maximize the monitoring quality of the monitoring area. The method mainly includes three stages: sensor uniform distribution (SR) stage, charging station number and location determination (RSNL) stage, and batch charging (BR) stage.

[0078] In the SR phase: sensors are evenly redistributed, aiming to balance the spatial monitoring quality.

[0079] In the RSNL stage: the monitoring area is divided into sub-areas of equal size to determine the number and location of charging stations.

[0080] In the BR stage: according to the location of the charging station in each sub-area, the shortest charging path for the charging vehicle in each sub-area is constructed; first, the charging time of a single sensor is calculated, and then the start charging time of each charging station is calculated based on this, and it is broadcast to all sensors in the corresponding grid; secondly, the average working time and charging interval of the sensor are calculated, and the number of charging batches of the sensors in the corresponding grid is calculated based on the average working time and charging interval; finally, the working frequency of the sensor is calculated and adjusted to ensure that the sensor can reach the charging station on time for charging before the battery is exhausted.

[0081] 1. SR stage

[0082] The sensor nodes are randomly deployed in the monitoring area. Some areas have dense nodes, while other areas may have fewer nodes. In order to balance the distance between sensors, this embodiment adopts the method of repulsion and attraction to achieve a more balanced distribution of sensors. When two nodes are too close, a repulsive force should be generated to push them apart; conversely, if the two nodes are too far apart, an attractive force should be generated to pull them closer. However, a sensor may be adjacent to multiple sensors, and the resultant force on the node is the vector sum of all repulsive and attractive forces from neighboring nodes.

[0083] The present invention assumes that the sensor can obtain the location information of itself and neighboring nodes (all other sensors within the communication radius of the sensor), which can be achieved by installing GPS on the sensor. x Indicates sensor x Voronoi cells. Figure 2 As shown, based on the neighboring sensors s1, s2, s3, s4 and s5, sensor s x Its Voronoi cell V can be constructed i .s x The perpendicular bisector between it and its neighbors forms a V x The real edge of V x In the example, each point is s away from the sensor x Closer than other sensors.

[0084] Let d(s x ,s y ) indicates sensor s x With s y The distance between ave represents the average distance between all sensor pairs in the monitoring area. Since the size of the monitoring area and the number of sensors are known, d can be calculated in advance. ave .

[0085] If the condition d(s x ,s y ) <d ave If it holds, then node s x and y The distance between ave -d(s x ,s y )) / 2. On the other hand, if the condition d(s x ,s y )>d ave If it holds, then node s x and y The distance between them should be reduced to (d(s x ,s y )-dave ) / 2. Since each sensor can calculate the distance and direction to the neighboring sensors, the distances between sensors will be balanced in a distributed manner. Indicates the slave sensor s y To sensors x The virtual force can be calculated by the following formula Values:

[0086]

[0087] Assume N x Indicates sensor x The total force of the neighboring sensor set It can be expressed as:

[0088]

[0089] Figure 3 and Figure 4 An example of attractive and repulsive forces between two nodes is provided in Figure 3 In the above example, we assume that d ave =3, d(s x ,s y )=5, it can be clearly seen that s x and y The distance between them is greater than the average distance. Therefore, the two nodes experience an attraction with strength -1, indicating that the nodes are 1 unit closer to each other. Figure 4 In, d(s x ,s y )=2, which is lower than the average distance. Therefore, s x and y A repulsive force of strength 0.5 is applied, causing each node to move 0.5 units away from each other.

[0090] Figure 5 Given the sensor s x Example of resultant force calculation. Figure 5 As shown, the sensor s x There are three neighbor nodes s1, s2, and s3, because:

[0091] d(s x ,s1)>d ave

[0092] d(s x ,s3)>d ave

[0093] d(s x ,s2) <d ave

[0094] To balancex The distance between sensor s1, s2, and s3 x It should move closer to s1 and s3 and away from s2. x exerts an attractive force, and s2 exerts an attractive force on s x Apply repulsive forces, which are expressed as Figure 5 The orange, purple, and blue lines represent the Figure 6 In the figure, the red line indicates the effect on s x The combined force.

[0095] Assume y For x Neighbors of . Let the symbol and Respectively represent V x and V y The area of ​​. Let σ be the set threshold. The calculation of virtual force and the movement of the sensor can be performed repeatedly until condition (9) is satisfied.

[0096]

[0097] 2. RSNL stage

[0098] In the present invention, all sensors will be moved to the nearest charging station. In the conceptual design, sensors, charging vehicles, and charging stations are similar to people, buses, and bus stops, respectively. People who want to take the bus must go to the nearest bus stop. There may be multiple buses traveling along the same path, passing all bus stops. Similarly, all sensors that wish to be charged should also be moved to the charging station for charging. The goal of this stage is to determine the number and location of charging stations.

[0099] To achieve this goal, the present invention initially divides the entire monitoring area into m sub-areas, where m is an integer greater than or equal to 1. Each sub-area will dispatch a charging vehicle to charge the sensors in that specific area. In order to minimize the cost of moving the sensor from its sensing location to the charging station, each sub-area may have multiple charging stations. Therefore, each sub-area will be further divided into k grids, and a charging station will be deployed in each grid.

[0100] Assume that the charging vehicle consumes 1 / m of its total energy for charging and moving in each sub-area. The number of charging stations affects the monitoring quality and the charging efficiency of the charging vehicle. When there are more charging stations, the average moving distance between the sensor and the charging station can be significantly reduced. However, this strategy consumes too much energy of the charging vehicle because it needs to move to each charging station. In addition, this strategy has the disadvantage of low charging efficiency. Too many deployments of charging stations will reduce the number of sensors that can be charged simultaneously at each station, resulting in a decrease in the number of sensors that can be charged simultaneously, thereby reducing the charging efficiency. In addition, each charging station also takes a fixed charging time during the charging cycle. Therefore, the time period for the charging vehicle to move between all charging stations will be extended, resulting in failure to meet the constraints of data freshness or insufficient time to charge some sensors. Therefore, the number of charging stations should be carefully planned.

[0101] Assume that each sub-area has k virtual charging stations and the density of charging stations is Therefore, the average distance between two adjacent charging stations is The two main opportunities for charging vehicles to consume energy are moving and charging. They represent the energy cost of charging and moving in the sub-area respectively. The energy consumption of the charging vehicle must satisfy expression (10):

[0102]

[0103] This embodiment assumes that when the sensors arrive at the charging station for charging, their energy is just exhausted. In this way, the sensors can reserve the most energy for performing the monitoring task to maximize the monitoring quality. represents the total energy of the sensor, and l represents the charging efficiency of the charging vehicle. The charging vehicle needs to consume energy, that is to say:

[0104]

[0105]

[0106] According to formulas (10)-(12), the upper limit of symbol k is calculated as:

[0107]

[0108] Next, we focus on the lower bound of k.

[0109] Fewer charging stations result in shorter charging paths for charging vehicles, enabling more sensors to be charged at each charging station simultaneously, thus improving charging efficiency. However, this strategy also results in a longer average distance between mobile sensors and charging stations. As a result, mobile sensors need to move long distances to reach charging stations for charging, causing them to allocate more energy for movement. Therefore, it is crucial to avoid having too few charging stations, and the following outlines the calculation method for determining the lower limit of the symbol k.

[0110] The work efficiency constraint, as described in formula (8), can be rewritten as:

[0111]

[0112] Where ε is a constant greater than 10. Formula (15) calculates

[0113]

[0114] Without loss of generality, the longest distance a sensor can move is the diagonal distance of the grid. Therefore,

[0115]

[0116] According to formulas (14)-(16), the lower limit of symbol k is calculated:

[0117]

[0118] The symbol k is an integer between the upper and lower limits:

[0119]

[0120] Therefore, a feasible solution for the variable k is obtained, which indicates that the number of charging stations can be determined. The next problem is to determine the location of each charging station.

[0121] The key concept to determine the locations of k charging stations is to apply the K-means algorithm to the sensors in each sub-area. This means that the sensors in each sub-area should be divided into k clusters, and a charging station is deployed at the center of each cluster. Initially, k cluster points are randomly selected in each sub-area. Then, clustering is performed based on the distance between the cluster points and the sensors. Next, for each cluster, its centroid is determined, and the closest point will serve as the cluster point. These steps are repeated until the membership of each cluster remains unchanged. Finally, the centroid of each cluster is set to a charging station, and the centroid close to the center of the monitoring area is set to the base station.

[0122] 3. BR Stage

[0123] According to the charging station RS jThe shortest charging path for the charging vehicle is constructed based on the TSP algorithm for the monitoring area. represents the charging station of the i-th charging vehicle in the i-th sub-area. For example, Indicates the first charging station of the second charging vehicle in the second sub-area. Next, this embodiment calculates the start charging time of the first charging station in the ith sub-area, and on this basis, further calculates the start charging time of each charging station.

[0124] Assume that at the charging station There are u sensors in the cluster, denoted as Indicates sensor The position of; d(·,·) represents the Euclidean distance between two positions; Indicates sensor from Move to The energy consumed. The value of can be obtained by formula (19):

[0125]

[0126] set up Indicates sensor The remaining energy used to perform monitoring work. It can be calculated by formula (20):

[0127]

[0128] set up Indicates sensor The time required for the vehicle to be charged. It can be obtained by formula (21):

[0129]

[0130] In order to meet the charging requirements of each sensor and improve the monitoring quality, this embodiment assumes that The initial charging time is to minimize the sensor The moment of charging time. That is:

[0131]

[0132] according to The starting charging time of other charging stations is calculated by formula (23). Indicates the time to start charging. The value of can be calculated by applying formula (23):

[0133]

[0134] As mentioned above, this embodiment uses the K-means algorithm to divide the entire monitoring area into m×k clusters, denoted by C={c1,c2,…,c m×k At the start of charging, all sensors move to RS i This means that the sensor being charged cannot work, resulting in c i The monitoring quality is reduced. In order to improve the monitoring quality, the present invention assumes that the charging task of the sensor is performed in batches. i Indicates c i The number of sensor batches in the equation (24) is: i The value is equal to c i The quotient of the average working time of the sensor and the charging interval is rounded up, where Indicates charging station Next charging start time:

[0135]

[0136] Where: average() represents the average value function; Indicates charging station Corresponding sensor working hours.

[0137] After the sensor is charged in batches, the charging time is delayed by θ compared to non-batch charging. i To ensure that the sensor does not run out of battery before being charged by the charging vehicle, this paper adjusts the sensor's operating frequency according to the sensor's start charging time. Indicates sensor i The energy consumption when performing a perception task, The value of can be calculated by formula (25):

[0138]

[0139] set up Indicates sensor i Energy reserved for performing sensing tasks, sensor s i It is necessary to go to the corresponding charging station (RS) to be charged by the charging vehicle. After charging is completed, the sensor returns to its original working position to resume monitoring operation. The value of can be determined by formula (26):

[0140]

[0141] set up represents the number of sensor detections, where equal Divide by

[0142]

[0143] set up Indicates sensor i Working hours, The value of is calculated by formula (28):

[0144]

[0145] set up Indicates sensor i The adjusted operating frequency, The value of is calculated by formula (29):

[0146]

[0147] At this point, the required parameters have been obtained through the above calculations. After the sensor starts working according to the calculated working time and working frequency, the charging vehicle arrives at each charging station in turn according to the constructed shortest charging path. After the charging vehicle arrives, the sensors in the corresponding grid of each charging station are moved to the corresponding charging station according to the calculated batch number and charged by the charging vehicle. The charging vehicle completes charging of the sensor according to the calculated charging time. After charging is completed, the sensor returns to its original position to re-execute the monitoring task.

Claims

1. A wireless sensor network charging method with multiple charging vehicles cooperating, characterized in that: The following steps are involved: Step 1: Sensor uniform distribution stage Evenly distribute sensors within the monitoring area to balance the spatial monitoring quality; Step 2: Determine the number and location of charging stations Step 2.1, divide the monitoring area into m sub-areas, and set a charging vehicle in each sub-area; m is an integer greater than or equal to 1; Step 2.2: Divide each sub-area into k grids, deploy a charging station in each grid, and use the following formula to determine the number of charging stations k: And k is an integer; Where: ε is a constant greater than 10; is the power consumption rate of the sensor per unit distance; L and M are the length and width of the monitoring area respectively; is the total energy of the sensor; l is the charging efficiency of the charging vehicle; is the total energy of the charging vehicle; D is the average distance between two adjacent charging stations; The power consumption rate per unit distance of the charging vehicle; Step 2.3, the mobile sensors in the k grids in each sub-area are grouped as k clusters, the center of gravity of each cluster is determined, and the center of gravity of each cluster is set as the final setting location of the charging station, and the center of gravity of the cluster close to the center of the monitoring area is set as the final setting location of the sensor network base station; Step 3: Bulk charging stage According to the location of charging stations in each sub-area, the shortest charging path for charging vehicles in each sub-area is constructed; First, the charging time of a single sensor is calculated, and then the starting charging time of each charging station is calculated based on this, and the time is broadcast to all sensors in the corresponding grid; secondly, the average working time and charging interval of the sensor are calculated, and the number of charging batches of the sensors in the corresponding grid is calculated based on the average working time and charging interval; finally, the operating frequency of the sensor is calculated and adjusted to ensure that the sensor can arrive at the charging station on time for charging before the battery is exhausted; the sensor starts working, and the charging vehicle arrives at each charging station in turn according to the shortest charging path constructed. After the charging vehicle arrives, the sensors in the corresponding grid of each charging station move to the corresponding charging station according to the calculated number of batches for charging by the charging vehicle, and return to their original position after charging is completed to re-execute the monitoring task.

2. The wireless sensor network charging method with multiple charging vehicles in cooperation according to claim 1 is characterized in that: Step 1 is as follows: By using the method of repulsion and attraction, two sensors that are close to each other are pulled apart, and two sensors that are far away are pulled together, so that the distance between each sensor and its neighboring nodes meets the set conditions, thereby achieving uniform positioning of sensors in the monitoring area to balance the quality of spatial monitoring; The neighboring nodes refer to all other sensors within the communication radius of each sensor.

3. The wireless sensor network charging method for multiple charging vehicles in cooperation according to claim 2 is characterized in that: In step 1, define sensor s x and sensors y For any two adjacent sensors in the monitoring area, V x Indicates sensor x The Voronoi cell, V y Indicates sensor y The Voronoi cells, and Respectively represent V x and V y The area of ​​​​the sensor s is represented by σ, and σ represents the set threshold; the Voronoi unit is x A closed geometric figure enclosed by the perpendicular bisectors of the lines connecting its adjacent nodes; Then, sensor s x and sensors y The distance between them meets the following setting conditions: Where: N x For sensors x A collection of neighboring nodes.

4. The wireless sensor network charging method with multiple charging vehicles cooperating according to any one of claims 1 to 3, characterized in that: In step 2.1, the areas of the m sub-regions are equal; In step 2.2, the areas of the k grids are equal; In step 2.3, the K-means clustering algorithm is applied to the sensors in each sub-area to determine the center of gravity of each cluster.

5. The wireless sensor network charging method for multiple charging vehicles in cooperation according to claim 4 is characterized in that: In step 3, the shortest charging path for the charging vehicle in each sub-area is constructed based on the TSP algorithm.

6. The wireless sensor network charging method with multiple charging vehicles cooperating according to claims 1-3 is characterized in that: definition, Indicates the i-th charging vehicle in the i-th sub-area Charging stations, at charging stations There are u sensors in the corresponding grid, which are recorded as Then in step 3, the sensor is calculated using the following formula Charging time Where: Indicates sensor The remaining energy used to perform monitoring tasks; Indicates sensor Monitoring power consumption rate; Indicates sensor from Move to Energy consumed; Indicates sensor location; Indicates charging station The position of; d(·,·) represents the Euclidean distance between two positions.

7. The wireless sensor network charging method for multiple charging vehicles in cooperation according to claim 6 is characterized in that: In step 3, the start charging time of the first charging station in the i-th sub-area is taken as the starting point, and the start charging time is calculated using the following formula: Start charging time at the first charging station Based on this, the following formula is used to calculate the start charging time of the remaining charging stations: Where: Indicates charging station The start time of charging; Indicates the start time of charging at other charging stations; Indicates the charging rate of the charging vehicle; v M Indicates the moving speed of the charging vehicle.

8. The wireless sensor network charging method for multiple charging vehicles in cooperation according to claim 7 is characterized in that: In step 3, the batch number is calculated using the following formula: Where: Indicates charging station The next charging start time; average() represents the average value function; Indicates charging station Corresponding sensor working hours.

9. The wireless sensor network charging method with multiple charging vehicles cooperating according to claim 8 is characterized in that: Definitions i Represents a sensor; Then in step 3, the sensor s i Working hours and operating frequency The calculation is done using the following formula: Where: Indicates sensor i The number of monitoring tasks; Indicates sensor i Energy consumption of executing a monitoring task; v s Indicates sensor i Movement speed; Indicates sensor i The remaining power; i Indicates sensor i The operating frequency; Indicates sensor i Monitoring power consumption rate; Indicates sensor i location.

Images