CN109451556B - Method for charging wireless sensor network based on UAV - Google Patents

Abstract

The invention relates to a wireless sensor network charging technology, in particular to a method for charging a wireless sensor network based on a UAV (unmanned aerial vehicle), which comprises the following steps: collecting ID, position information, residual energy and working state of each sensor network node; determining member nodes and a clustering scheme, and solving the communication energy consumption of the member nodes and cluster head nodes; determining an energy transmission power model of the UAV; calculating the time required by charging each node i by the UAV in the mth random periodic clustering scheme through an optimization algorithm; searching the time required by the UAV to fully charge all nodes at any position by using an optimization algorithm from N possible clustering schemes, and selecting the maximum time to be regarded as the time required by the UAV to fully charge the whole wireless sensing network; and finding out the minimum time cost value and the corresponding optimal UAV charging position or path from the maximum time corresponding to the N possible clustering schemes. The invention can fully charge the wireless sensor network at the minimum time cost after multiple rounds of energy consumption, so that the wireless sensor network can stably run for a long time.

Description

A method of charging wireless sensor network based on UAV

technical field

The invention relates to a wireless energy transmission technology of a wireless sensor network, in particular to a method for charging a wireless sensor network based on UAV.

Background technique

With the increasing development and maturity of embedded microcomputer system, wireless communication technology and sensor technology, wireless sensor network enables people to continuously enhance the ability of non-contact interaction with various situations in the real world. An emerging networking application model comparable to the Internet.

Wireless sensor network is a distributed wireless sensor network composed of multiple wireless sensor nodes. Since its basic composition is a wireless sensor that can perceive and collect external environmental data, the relevant data of the monitored and detected objects can be collected by the wireless sensor network terminal - wireless sensor, and form a unique multi-hop through wireless communication. Self-organizing network.

Since the birth of wireless sensor network technology, how to prolong the life cycle of wireless sensor network and ensure the stability of wireless sensor network has always been an important problem to be solved by scholars at home and abroad. Since wireless sensor network nodes are generally powered by batteries and carry very limited energy, how can the wireless sensor network still maintain a long network operation life cycle when data collection, fusion and transmission are performed under limited energy, so that the It has become one of the current research hotspots to run the wireless sensor network as stably as possible. From the perspective of wireless sensor network structure and operation mode, the energy consumption in wireless sensor network is divided into sensor computing energy consumption, sensor communication energy consumption, sensor data acquisition energy consumption, etc. Among them, sensor communication energy consumption is the most important, and in The way of clustering in wireless sensor network has a great influence on the energy consumption of sensor communication. Therefore, choosing the optimal clustering principle in the wireless sensor network area is an important strategy to prolong the life cycle of wireless sensor network.

With the development of emerging UAV technology, the UAV path planning strategy and design have attracted extensive attention from scholars at home and abroad. Extending the life cycle of wireless sensor networks by combining UAV path planning strategies with wireless energy transmission technology has also become a research hotspot.

SUMMARY OF THE INVENTION

The purpose of the present invention is to solve the problems of poor stability and too short life of the existing wireless sensor network, and propose a method for charging the wireless sensor network based on UAV. The location of the charging UAV and the optimal cluster head selection scheme can fully charge the wireless sensor network with the least time cost after multiple rounds of energy consumption, so that the wireless sensor network can run stably for a long time.

The present invention is achieved through the following technical solutions: a method for charging a wireless sensor network based on UAV, comprising the following steps:

S1. Collect the ID, location information, remaining energy and working status of each sensor network node in the wireless sensor network; the sensor network nodes are divided into cluster head nodes and member nodes;

S2. Determine the member nodes and the clustering scheme in the wireless sensor network area, and obtain the communication energy consumption of the corresponding member nodes and the communication energy consumption of the cluster head node;

S3, determine the energy transmission power model of the energy transmission module UAV;

S4. Calculate the time required for the energy transmission module UAV in the mth random periodic clustering scheme to fully charge each sensor network node i in the wireless sensor network through an optimization algorithm;

S5. Use the optimization algorithm to search for the time required for the energy transmission module UAV to fully charge all sensor network nodes at any position in the entire wireless sensor network area from the N possible clustering schemes, select the maximum time among them, and determine it as The time required for the energy transfer module UAV to fully charge the entire wireless sensor network;

S6. From the maximum time corresponding to the N possible clustering schemes, find the minimum time cost value and the corresponding optimal UAV charging position or path.

It can be seen from the above technical solutions that in the present invention, in the entire wireless sensor network area, each block wireless sensor network area randomly selects a cluster head and a member configuration scheme, and each block wireless sensor network area clustering scheme Without affecting each other, each regional clustering scheme is combined with each other to form the cluster head and member configuration scheme of the entire wireless sensor network (that is, the clustering scheme). Compared with the existing technology, the beneficial effects obtained include:

By comparing the optimal time required to fill the wireless sensor network corresponding to the cluster head and member configuration scheme of the entire wireless sensor network, the clustering scheme corresponding to the least time and the corresponding optimal charging position of the UAV are selected. ; that is, in the case of the energy consumption corresponding to the cluster head and member configuration scheme of the wireless sensor network, the position or path corresponding to the optimal cost of UAV time is obtained. The invention searches for the optimal time cost position in each cluster head and member configuration scheme, and finally compares each optimal time cost, and selects the UAV position and the cluster head member configuration scheme corresponding to the least optimal time cost. The wireless sensor network can be fully charged with the least time cost after multiple rounds of energy consumption. The invention patent researches the relevant optimization algorithm to find the optimal UAV charging position and path under the condition of the least time cost, and selects the optimal cluster head selection scheme at the same time.

Description of drawings

Fig. 1 is the flow chart of the inventive method;

Fig. 2 is the wireless sensor network applied in one embodiment;

Fig. 3 is the network clustering situation of the wireless sensor network in one embodiment;

FIG. 4 is a schematic diagram of the result of simulating the entire wireless sensor network in one embodiment;

Fig. 5 is the relation diagram of objective function value and algorithm algebra in one embodiment;

6 is a simulation diagram of the optimal UAV position when the position of the communication base station is fixed in one embodiment;

FIG. 7 is an objective function convergence diagram when the position of the communication base station is fixed in one embodiment.

Detailed ways

The present invention will be described in further detail below with reference to the embodiments and the accompanying drawings, but the embodiments of the present invention are not limited thereto.

Example

In the present invention, the sensor network node selects the rechargeable wireless sensor network node; the UAV is an energy transmission module for wireless energy transmission, and the UAV flies at a fixed altitude and transmits electric energy to the wireless sensor nodes within the energy transmission range.

The scene of the wireless rechargeable sensor network is divided into four areas, and the sensor network nodes in each area are divided into cluster head nodes and member nodes. Periodically select and generate cluster head nodes. In the wireless rechargeable sensor network, each working node transmits its own location information and ID information to the periodically randomly generated cluster head node through a single hop, and the cluster head node integrates and fuses the received information into data The packet is then transmitted to the communication base station in the form of a single hop. Among them, the wireless sensor network node clustering principle is the random periodic clustering principle, and the type of each cluster head member configuration scheme (that is, the clustering scheme) is multiplied by the number of nodes in each area, namely:

N=N ₁ *N ₂ *N ₃ *N ₄ ......N _n .

As shown in Figure 1, the realization flow and the steps of the method of the present invention are as follows:

1. After the L-round cluster head rotation of the wireless sensor network, collect the ID, location information, remaining energy and working status of each sensor network node at this time.

Here, it is assumed that the remaining energy of each working node is E_min(i) (the remaining energy of the cluster head node is small due to the need to integrate the received data for forwarding and other operations, but the lower limit of the minimum remaining energy is e_min, and the remaining energy of the working node is e_min. When the energy reaches e_min, the working node turns into a dormant node), and the communication energy consumption is ei=E-E_min (E is the initial energy of each node). energy, and the remaining energy is calculated specifically through the energy consumption model.

In the application scenario of the present invention, the communication energy consumption model (ie, the energy consumption model) of the working nodes of the wireless sensor network adopts the first-order radio model, that is, when the communication distance between nodes is _relatively close, the energy consumption model adopts the free space channel model. ; When the communication distance d _ii between nodes is long, the energy consumption model adopts the multi-path attenuation model. Due to the different types of nodes, the energy consumption model of the algorithm scene can be divided into the energy consumption model of the member node and the energy consumption model of the cluster head node. Assuming that the distance between nodes that communicate with each other is known to be d _ii , when sending 1-bit data, calculate the relevant energy consumption model:

1. When the sending node is a member node, the energy consumed by the sending node is:

2. When the sending node is a member node, and the transmission distance comparison is not considered, the energy consumption of the sending node is:

3. When the sending node is the cluster head node, the energy consumed by sending information data is:

4. The energy consumed by receiving information data (cluster head nodes and member nodes) is:

E _RX (K)=KE _elec

The above formulas are the energy consumed by the node when sending and receiving 1 bit data, E _DA is the energy consumption of data integration, that is, the energy consumed by integrating 1 bit data; d ₀ is the distance threshold of the node energy consumption, which is a fixed value; K is the receiving and The amount of data sent; E _elec is the energy consumption of a node to receive or send 1 bit data, the present invention only considers the cluster head node to receive information; ε _fs and ε _mp are the attribute parameters of the sensor network node; symbol b is a Brun Variable, when the data transmission mode between two nodes is single-hop, the value of b is zero, and when the data transmission mode between nodes is multi-hop, the value of b is 1.

In this algorithm model, the mathematical model of energy transmission of mobile UAV to wireless sensor network nodes is a linear energy transmission model, namely:

Among them, Q _i (x(t), y(t)) represents the energy received power of the i-th sensor network node at point (x(t), y(t)) in the wireless sensor network area, h _k (t ) is the distance-dependent power magnitude, and β ₀ represents the energy received by a node with a unit distance of 1 m on the transmission channel. P represents a constant in transmission power. (x _k , y _k ) represents the coordinates of the UAV flying in the two-dimensional plane, and H represents the flying height in the three-dimensional coordinates.

It can be seen from the above that the energy consumption of the cluster head node is much larger than that of the member nodes because the cluster head node has more energy consumption for reception and transmission.

2. Determine the member nodes in the wireless sensor network area and the clustering scheme, and obtain the corresponding communication energy consumption e _i of member nodes and communication energy consumption E _i of the cluster head node, where the communication energy consumption of the cluster head node is the energy consumption of receiving information and the sum of energy consumption for sending information.

3. Determine the energy transmission power model of the energy transmission module UAV.

(包括对簇头节点充满电时所需时间)，

4. Find the time required for the energy transmission module UAV to fully charge each sensor network node i in the wireless sensor network in the mth random periodic clustering scheme by optimizing the search algorithm

(including the time required to fully charge the cluster head node),

其中

选取其中的最大时间T_{i_max_UAV_loc(x(t),y(t))} ^m，认定其为能量传输模块UAV给整个无线传感网络充满电所需的时间。5. From N possible clustering schemes, use the optimization algorithm to search for the time required for UAV_loc(x, y) to fully charge all sensor network nodes at any location in the entire wireless sensor network area

in

The maximum time T _{i_max_UAV_loc(x(t),y(t))} ^m is selected, and it is determined as the time required for the energy transmission module UAV to fully charge the entire wireless sensor network.

以及对应最优UAV充电位置或路径。6. Find the minimum time cost value from the maximum time T _{i_max_UAV_loc(x(t), y(t))} ^m corresponding to the N possible clustering schemes

And the corresponding optimal UAV charging location or path.

Different optimization algorithms have different convergence degree, convergence speed and accuracy to the problem objective function in the patent application scenario of the present invention. In this embodiment, the basic differential evolution algorithm is used to optimize the search for the UAV position in the sensor network area. Because of its simple structure, easy implementation, fast convergence and strong robustness, it is widely used in data mining, pattern recognition, digital filter design, artificial neural network, electromagnetics and other fields. In this embodiment, the differential evolution algorithm is used to perform mutation, crossover, competition operation, etc. on the communication base station and UAV position data in the time cost function of the unmanned aerial vehicle (UAV) assisting the wireless sensor network, and then the mutation, crossover, competition, etc. After the operation, the data is substituted into the target time cost function, and the shortest and optimal time cost function value is obtained.

In this implementation, the actual application and simulation scenario of the algorithm is the wireless sensor network as shown in Figure 2. The wireless sensor network nodes are divided into two sensor network node areas, in which the small dots are the UAV energy transmission corresponding to the optimal solution Location.

As shown in Figure 3, according to the periodic random clustering principle and information flow, the network clustering situation of one of the wireless sensor networks is described. The working nodes transmit their respective node ID information and location information through single-hop transmission. To the cluster head node, the cluster head node integrates and fuses the information received from each sensor node, and then sends the fused data packet to the communication base station in the form of single-hop transmission.

As shown in Figure 4, in the case of optimizing the UAV position, by comparing the time required for full charging of various clustering schemes in the entire wireless sensor network area, it is found that the time cost is the least (that is, the mobile flight with charging It takes the least time for the UAV to fully charge the entire wireless sensor network), and record the position of the UAV at this time. This position is the optimal position where the rechargeable UAV has the least energy transmission time cost. This clustering scheme is the optimal cluster scheme. At this time, under the condition that the position of the communication base station is fixed, the differential evolution algorithm is used to optimize the variable of UAV position, and the simulation is carried out in the entire wireless sensor network. The results are shown in Figure 4.

Figure 5 shows the optimal search for the flight position of the unmanned aerial vehicle (UAV) in the wireless sensor network area using the differential evolution algorithm. In the wireless sensor network clustering situation and the fixed communication base station, there is no time cost objective function in the constructed time cost objective function. The relationship between the objective function value and the algorithm algebra after the evolution, competition, mutation and other operations of the human-machine (UAV) position variable, where the ordinate shows the time cost objective function value, and the abscissa shows the running algebra of the algorithm. As the running algebra of the algorithm increases, the time cost objective function value decreases with the increase of running algebra and tends to converge.

Figure 6 shows the simulation of the optimal UAV charging position in the wireless sensor network with the expanded number of nodes. In the figure, the network scale is slightly enlarged and divided into three independent clustered areas. Theoretically speaking, the data that needs to be processed in the wireless sensor network shows an order of magnitude increase with the continuous expansion of the network scale. Figure 7 illustrates the convergence of the objective function when the location of the communication base station is fixed.

It can be seen from the above that with the continuous expansion of the scale of the wireless sensor network, the data that the algorithm needs to process in the wireless sensor network is increasing as the scale of the wireless sensor network increases, and the time for the algorithm to process the data is also increasing. bigger.

Therefore, for large-scale wireless sensor networks in practical applications, an efficient UAV energy compensation strategy is very necessary. According to the large-scale wireless sensor network in practical application, due to the limited range of energy transmission and communication for UAVs, the present invention divides the large-scale wireless sensor network into countless small areas, and there are theoretically UAV energy transmission in small areas. The optimal location for transmission (as the case studied above), that is, in large-scale wireless sensor networks in practical applications, there are many optimal UAV energy transmission locations in different regions. In order to better meet the requirements of practical applications, the key to the simulated scene in the present invention is: 1. The scale of the wireless sensor network is increased in the original scene; 2. The UAV is used as a mobile communication relay and The wireless energy transmission party; 3. The original scene is regarded as a small area divided into a large-scale area, and the optimal position of the UAV for energy transmission and information collection is recorded as the UAV anchor point.

In actual large-scale wireless sensor network applications, due to the limited range of UAV energy transmission and communication, and because the wireless sensor network in practical applications is too large, UAV cannot perform large-scale energy supplementation within a certain period of time. . Based on the above reasons and the described scenarios, the present invention also proposes a priority energy replenishment algorithm based on large-scale wireless sensor network and unmanned aerial vehicle (UAV) technology, which is the optimal UAV position of the unmanned aerial vehicle in each small area The purpose of the movement trajectory strategy on the point is to improve the energy transmission efficiency of UAVs and optimize the energy consumption mode of large-scale wireless sensor networks through this optimized movement trajectory strategy. When the communication base station is fixed, each wireless sensor network node needs to transmit the collected data information to the cluster head node, and then forward it to the communication base station through the cluster head node. Wireless sensors have a huge energy consumption burden for this; UAV has less time cost for energy transmission to wireless sensor networks without fixed communication base stations, and the energy consumption of wireless sensor networks is less. Therefore, the present invention uses the UAV as a mobile communication relay for energy transmission to perform energy compensation on the large-scale wireless sensor network. Energy compensation and information gathering decisions. When the UAV is performing energy transmission and data collection tasks, the UAV takes the optimal information collection energy transmission position in each area as the theoretical optimal anchor point of the UAV, so as to carry out the optimal information collection and energy transmission tasks. The UAV is considering the optimal time cost t _e of UAV energy data in each area, the time cost t _f of flying to the theoretical optimal anchor point of UAV in small areas in all directions, and the sleep rate of sensor network nodes in small areas (ie energy After the ratio B _{i_empty} =n _empty /N _z ) of the number of exhausted nodes to the number of summary points, the next flight direction and trajectory of the UAV are judged and planned. When the UAV completes the energy transmission task in the previous cycle, it returns to the service station for energy replenishment (or waits for the next cycle task when the UAV has sufficient energy), and waits for the energy transmission and information collection tasks in the next cycle.

As described above, the present invention can be preferably implemented.

Claims (6)

1. A method for charging a wireless sensor network based on a UAV (unmanned aerial vehicle) is characterized by comprising the following steps:

s1, collecting IDs, position information, residual energy and working states of all sensor network nodes in the wireless sensor network after L rounds of cluster head rotation, wherein the sensor network nodes are divided into cluster head nodes and member nodes;

s2, determining member nodes and a clustering scheme in the wireless sensor network area, and solving the corresponding member node communication energy consumption and cluster head node communication energy consumption;

s3, determining an energy transmission power model of an energy transmission module UAV;

s4, calculating the time required by the energy transmission module UAV to fully charge each sensor network node i in the wireless sensor network in the mth random periodic clustering scheme through an optimized search algorithm;

s5, searching the time required by the energy transmission module UAV to fully charge all sensor network nodes at any position in the whole wireless sensor network area by using an optimization algorithm from N possible clustering schemes, selecting the maximum time, and determining the maximum time as the time required by the energy transmission module UAV to fully charge the whole wireless sensor network;

and S6, finding out the minimum time cost value and the corresponding optimal UAV charging position or path from the maximum time corresponding to the N possible clustering schemes.

2. The UAV-based method of charging a wireless sensor network of claim 1 wherein the cluster head node communication energy consumption is a sum of information receiving energy consumption and information transmitting energy consumption.

3. The UAV-based method of charging a wireless sensor network of claim 1 wherein the member nodes comprise a dormant node, a working node, and an inactive node, and wherein the cluster head nodes are generated based on an equiprobable, periodic selection.

4. The UAV-based method of charging a wireless sensor network of claim 1 wherein the optimization algorithm is a basic differential evolution algorithm.

5. The method of claim 3, wherein the energy consumption model of the working nodes is a first-order radio model, when the communication distance d between nodes is_iiWhen the energy consumption model is closer, a free space channel model is adopted; when the inter-node communication distance d_iiAnd when the distance is far, the energy consumption model adopts a multi-path attenuation model.

6. The UAV-based method of charging a wireless sensor network of claim 1, wherein: the energy transmission mathematical model of the energy transmission module UAV to the sensor network node is a linear energy transmission model:

wherein Q is_i(x (t), y (t)) represents the energy receiving power of the ith sensor network node at the point (x (t), y (t)) in the wireless sensor network area, h_k(t) is the power magnitude related to distance, β₀Representing the energy received by a node at a unit distance of 1m on the transmission channel, P represents a constant in the transmission power, (x)_k,y_k) The coordinates of the energy transmission module UAV flying in a two-dimensional plane are shown, and H represents the flying height in a three-dimensional coordinate.

Images