CN116634391A - Unmanned aerial vehicle-assisted data collection method - Google Patents
Unmanned aerial vehicle-assisted data collection method Download PDFInfo
- Publication number
- CN116634391A CN116634391A CN202310591381.4A CN202310591381A CN116634391A CN 116634391 A CN116634391 A CN 116634391A CN 202310591381 A CN202310591381 A CN 202310591381A CN 116634391 A CN116634391 A CN 116634391A
- Authority
- CN
- China
- Prior art keywords
- uav
- unmanned aerial
- data
- aerial vehicle
- chs
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 35
- 238000013480 data collection Methods 0.000 title claims abstract description 18
- 230000007613 environmental effect Effects 0.000 claims abstract description 7
- 241000854291 Dianthus carthusianorum Species 0.000 claims abstract description 5
- 238000012545 processing Methods 0.000 claims abstract description 5
- 230000005540 biological transmission Effects 0.000 claims description 19
- 238000012549 training Methods 0.000 claims description 18
- 230000009471 action Effects 0.000 claims description 17
- 238000004891 communication Methods 0.000 claims description 17
- 238000005457 optimization Methods 0.000 claims description 11
- 239000011159 matrix material Substances 0.000 claims description 10
- 230000000875 corresponding effect Effects 0.000 claims description 9
- 230000008569 process Effects 0.000 claims description 8
- 238000012544 monitoring process Methods 0.000 claims description 7
- 230000002787 reinforcement Effects 0.000 claims description 6
- 230000006870 function Effects 0.000 claims description 5
- 230000000694 effects Effects 0.000 claims description 3
- 239000000126 substance Substances 0.000 claims description 3
- 230000007423 decrease Effects 0.000 claims description 2
- 238000007405 data analysis Methods 0.000 abstract description 2
- 238000010295 mobile communication Methods 0.000 abstract description 2
- 238000005516 engineering process Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 238000005265 energy consumption Methods 0.000 description 2
- 230000009916 joint effect Effects 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000001413 cellular effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 238000004904 shortening Methods 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/30—Services specially adapted for particular environments, situations or purposes
- H04W4/38—Services specially adapted for particular environments, situations or purposes for collecting sensor information
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B7/00—Radio transmission systems, i.e. using radiation field
- H04B7/14—Relay systems
- H04B7/15—Active relay systems
- H04B7/185—Space-based or airborne stations; Stations for satellite systems
- H04B7/18502—Airborne stations
- H04B7/18506—Communications with or from aircraft, i.e. aeronautical mobile service
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W16/00—Network planning, e.g. coverage or traffic planning tools; Network deployment, e.g. resource partitioning or cells structures
- H04W16/18—Network planning tools
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W16/00—Network planning, e.g. coverage or traffic planning tools; Network deployment, e.g. resource partitioning or cells structures
- H04W16/22—Traffic simulation tools or models
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W40/00—Communication routing or communication path finding
- H04W40/24—Connectivity information management, e.g. connectivity discovery or connectivity update
- H04W40/32—Connectivity information management, e.g. connectivity discovery or connectivity update for defining a routing cluster membership
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Computing Systems (AREA)
- Molecular Biology (AREA)
- General Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Probability & Statistics with Applications (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Aviation & Aerospace Engineering (AREA)
- Astronomy & Astrophysics (AREA)
- Mobile Radio Communication Systems (AREA)
Abstract
The invention relates to a data collection method based on unmanned aerial vehicle assistance, and belongs to the technical field of mobile communication. Aiming at minimizing the total power of the wireless sensor network, an efficient data collection method based on unmanned aerial vehicle assistance is provided. First, a UAVs-assisted data acquisition model is constructed. The model is divided into two layers, wherein the lower layer is a ground layer, and the ground layer divides the sensor into cluster members CM and cluster heads CH. CMs are responsible for sensing environmental data and transmitting the data to the CH. The CH is responsible for collecting and processing the data sent by the intra-cluster CMs and uploading the final data to the UAV. The upper layer is an air network layer and consists of a plurality of UAVs, is responsible for collecting, storing and forwarding CHs uploading data, and finally transmits the data to a ground data center DC for data analysis. By optimizing the number and location of CH, UAV location, and UAV association with CHs, the total power of the system is minimized.
Description
Technical Field
The invention belongs to the technical field of mobile communication, and relates to a data collection method based on unmanned aerial vehicle assistance.
Background
With the continuous development of internet of things (Internet of Things, ioT) technology, more and more devices and articles are connected to the internet, forming a huge, highly intelligent network. The network not only can realize mutual communication and data transmission among the devices, but also can realize intelligent control and management and more efficient resource utilization through technologies such as cloud computing, big data analysis and the like. IoT technology architecture can be divided into three layers, an application layer, a network layer, and a perception layer, respectively. In the IoT aware layer, the wireless sensor network is an integral part.
WSNs are a distributed sensing network consisting of small embedded devices, called sensors, that can communicate by wired or wireless means. The common WSN forms a multi-hop self-organizing network through wireless communication, so that the sensing, the acquisition and the processing of the measurement objects in the coverage area are realized, and the final result is transmitted to a data center.
In recent years, due to technical advantages of rapid deployment, high fault tolerance, low cost and the like of WSNs, WSNs are often used in large-scale distributed monitoring scenarios, such as disaster monitoring, environmental monitoring, infrastructure monitoring and the like. However, geologic hazards or other natural hazards often occur in remote mountainous areas. Therefore, two important bottlenecks exist in WSN data acquisition systems: (1) The Sensor Node (SN) is mostly powered by a battery, and energy consumption needs to be considered; (2) And the partial area is not covered by the cellular network, the data cannot be uploaded, and the data acquisition problem needs to be considered.
Disclosure of Invention
In view of the above, the present invention aims to provide a data collection method based on unmanned aerial vehicle assistance.
In order to achieve the above purpose, the present invention provides the following technical solutions:
a data collection method based on unmanned aerial vehicle assistance, the method comprising the steps of:
s1: constructing a UAVs auxiliary WSN data acquisition model;
s2: designing a total emission power optimization scheme of the sensor;
s3: clustering the sensors, determining the number and the positions of the cluster head sensors, and determining corresponding cluster members;
s4: optimizing the deployment position of the unmanned aerial vehicle;
s5: and optimizing the association scheme of the cluster head and the unmanned aerial vehicle.
Optionally, in the step S1, a data acquisition model of the UAVs auxiliary WSN is constructed;
the model consists of an upper layer and a lower layer:
1) The upper layer is an air network and comprises a plurality of unmanned aerial vehicles, wherein the unmanned aerial vehicles fly to a designated position under the dispatching of DC and hover, and collect and store data uploaded by CHs associated with the unmanned aerial vehicles; after the acquisition task is completed, the unmanned aerial vehicle carries data and returns to a DC designated position for charging, and meanwhile, the data is forwarded to the DC;
2) The lower layer is a ground layer and consists of a WSN, a charging station and DC;
in order to ensure that the distance of SNs in WSN for transmitting data is within the maximum communication distance, clustering SNs is carried out by adopting a clustering algorithm, one SN is selected from each cluster to be used as CH, and the rest SNs are used as CMs; the CH is responsible for collecting and processing data sent by CMs in the cluster and uploading the final data to the UAV; CMs are responsible for sensing environmental data and transmitting the data to the CH; the data transmission is carried out in a cluster in a single-hop mode; the charging station is responsible for charging the UAV, which is deployed where the UAV can successfully establish communication with the DC; the DC is responsible for reconstructing data sent by the UAVs so as to realize environmental monitoring and disaster early warning of the area.
Optionally, in the step S2, a total emission power optimization scheme of the sensor is designed; the UAVs set is M, the CHs set is k, and the CMs set is N; considering two links of line-of-sight LoS and non-line-of-sight NLoS between the UAV and the CH, and adopting orthogonal multiple access between the UAV and the CH to avoid interference; the average path loss between UAV m and CH k is expressed as:
wherein eta LoS And eta NLoS The average additional loss of the LoS link and the NLoS link is different from scene to scene; a and b are S curve parameters; h is UAV flying height; f is the carrier frequency; d (D) m,k Representing the distance between CH k and the vertical projection of UAV m on the ground; c is the speed of light; d, d m,k A linear distance from CH k to UAV m;
the required transmit power for CH k to UAV m is as follows:
wherein delta m,k The correlation coefficient from CH k to UAV m is 0 and 1, delta m,k CH k is associated with UAV m when=1,for received power of UAV m, PL m,k Is the average path loss; the association relation between all CH and UAV is expressed as a KxM matrix omega, and the M-th column expresses the association condition of UAV M to all CH;
the communication model in the WSN adopts a wireless communication channel model, and the transmission power of CM n to CH k is expressed as:
wherein,,the correlation coefficient between CM n and CH k is 0 and 1, and the values of the correlation coefficient are +.>When CM n is associated with CH k and vice versa; />The received power for CH k; d, d n,k Represents the distance between CM n and CH k; l is a transmitted system loss factor, and L is more than or equal to 1; g t And G r Gain of CM transmit and receive antennas, respectively; lambda is the wavelength; the association relation between all the CM and the CH is expressed as a matrix gamma of N multiplied by K, and the K column expresses the association condition of CH K to all the CM;
the method comprises the steps of providing a sensor total emission power optimization scheme which jointly optimizes the number and the positions of CH, the deployment position of UAVs and the association of the UAVs and CHs, minimizing the total emission power of SNs in a WSN, wherein the optimization targets are as follows:
one CH can only be associated with one UAV, and one CM can only be associated with one CH; meanwhile, the transmitting power of CH cannot exceed the maximum transmitting power of CH, and the transmitting power of CM cannot exceed the maximum transmitting power of CM; the deployment location of each UAV and the location of CH are ensured to be within the target area Λ.
Optionally, in the step S3, firstly, the error square sum SSE of the clustering result of the K-means algorithm is evaluated by an elbow method to find the optimal cluster number K; as the clustering number k increases, the SSE index gradually decreases as the clustering effect is better; in the process of increasing k, when the k value is smaller than the optimal cluster number, the descending speed of SSE is fast, and when the k value exceeds the optimal cluster number, the descending speed of SSE is slow; an elbow is imaged by the relation between SSE and k, and the k value corresponding to the elbow is the optimal cluster number;
after finding the optimal cluster number, using K-mean++ algorithm to find the corresponding K valueOptimal clustering, judging whether the transmission power from all the CMs to the associated CH is smaller than the maximum transmission power, if yes, outputting an optimal K value, a CH position and an association relation between CH and the CM, if not, increasing the K value to execute K-mean++ algorithm clustering again until the transmission power from all the CMs to the associated CH is smaller than the maximum transmission power, and outputting the optimal K value and the CH position l k And the association of CH with CM.
Optionally, in S4, when determining the number k of CH and the position l k After the correlation y between CH and CMs, the chemical problem is only related to the deployment position l of UAV m Related to the relationship Ω of UAV and CHs; the optimization problem is updated as follows:
based on the DRL algorithm, an UAV deployment algorithm based on AC is provided, a deployment process of the UAV is modeled as an MDP process, and a deployment position of the UAV is determined through centralized training.
Optionally, in the step S5, during the training, a differential power correlation algorithm is provided to determine rewards in each training state, where the algorithm minimizes the total emission power of CHs under the condition of considering the unmanned aerial vehicle load balancing; the CHs selects unmanned aerial vehicle to correlate according to the shortest distance, and when the correlation number of unmanned aerial vehicles exceeds the maximum load numberWhen the unmanned aerial vehicle calculates the required transmitting power related to other unmanned aerial vehicles, the unmanned aerial vehicle selects the +.>Each CHs establishes an association and the remaining CHs re-match the remaining UAVs.
Optionally, in the step S4, an AC deep reinforcement learning algorithm is used to deploy the unmanned aerial vehicle; the critic network evaluates the action selected by the actor network by means of a state-action-value function, which, in each iteration,the system will be based on the current state s i And policy pi i Selecting a corresponding action a i Training by adopting a Monte Carlo strategy gradient reinforcement learning method; wherein, when calculating action rewards, adopting the correlation algorithm based on the difference power proposed in the step S5 to determine the correlation relationship; and after the algorithm converges, outputting the deployment position of the UAV and the association relation between the UAV and the CHs.
The invention has the beneficial effects that: the invention designs a data collection method based on unmanned aerial vehicle assistance, and constructs a UAVs assisted WSN data collection model. And clustering the sensors by adopting a clustering algorithm, and shortening the data transmission distance to reduce the wireless transmission energy consumption. And selecting one sensor from each cluster as a CH, sending the acquired data to the CH by other SNs in the cluster, and gathering all the received information by the CH and sending the information to the unmanned aerial vehicle. The unmanned aerial vehicle hovers after flying to a specified position under the dispatching of DC, and acquires and stores data uploaded by CHs associated with the unmanned aerial vehicle. After the acquisition task is completed, the unmanned aerial vehicle carries data and returns to the position appointed by the DC for charging, and meanwhile, the data is forwarded to the DC. The invention aims at minimizing the total transmitting power of a wireless sensor network, and proposes to optimize the quantity and the positions of CHs based on an improved K-means++ clustering algorithm. Meanwhile, UAV deployment and association algorithm based on AC deep reinforcement learning is proposed to determine UAV and CHs association and unmanned aerial vehicle deployment position.
Additional advantages, objects, and features of the invention will be set forth in part in the description which follows and in part will become apparent to those having ordinary skill in the art upon examination of the following or may be learned from practice of the invention. The objects and other advantages of the invention may be realized and obtained by means of the instrumentalities and combinations particularly pointed out in the specification.
Drawings
For the purpose of making the objects, technical solutions and advantages of the present invention more apparent, the present invention will be described in the following preferred detail with reference to the accompanying drawings, in which:
FIG. 1 is a model diagram of a UAVs-assisted WSN data acquisition system;
FIG. 2 is a flowchart of an AC-based UAVs deployment algorithm;
fig. 3 is a flow chart of wireless sensor network data collection based on unmanned aerial vehicle assistance.
Detailed Description
Other advantages and effects of the present invention will become apparent to those skilled in the art from the following disclosure, which describes the embodiments of the present invention with reference to specific examples. The invention may be practiced or carried out in other embodiments that depart from the specific details, and the details of the present description may be modified or varied from the spirit and scope of the present invention. It should be noted that the illustrations provided in the following embodiments merely illustrate the basic idea of the present invention by way of illustration, and the following embodiments and features in the embodiments may be combined with each other without conflict.
Wherein the drawings are for illustrative purposes only and are shown in schematic, non-physical, and not intended to limit the invention; for the purpose of better illustrating embodiments of the invention, certain elements of the drawings may be omitted, enlarged or reduced and do not represent the size of the actual product; it will be appreciated by those skilled in the art that certain well-known structures in the drawings and descriptions thereof may be omitted.
The same or similar reference numbers in the drawings of embodiments of the invention correspond to the same or similar components; in the description of the present invention, it should be understood that, if there are terms such as "upper", "lower", "left", "right", "front", "rear", etc., that indicate an azimuth or a positional relationship based on the azimuth or the positional relationship shown in the drawings, it is only for convenience of describing the present invention and simplifying the description, but not for indicating or suggesting that the referred device or element must have a specific azimuth, be constructed and operated in a specific azimuth, so that the terms describing the positional relationship in the drawings are merely for exemplary illustration and should not be construed as limiting the present invention, and that the specific meaning of the above terms may be understood by those of ordinary skill in the art according to the specific circumstances.
Fig. 1 shows a schematic diagram of a possible structure of a data acquisition system of an unmanned aerial vehicle-assisted wireless sensor network according to an embodiment of the present invention. As shown in fig. 1, the system considers a two-layer network, including an air layer and a ground layer. The aerial network layer is composed of a plurality of unmanned aerial vehicles, and the unmanned aerial vehicles hover after flying to a designated position under the dispatching of DC, and collect and store data uploaded by CHs associated with the unmanned aerial vehicles. After the acquisition task is completed, the unmanned aerial vehicle carries data and returns to the position appointed by the DC for charging, and meanwhile, the data is forwarded to the DC. The ground layer is composed of WSNs, charging stations, and DC. In order to ensure that the distance of SNs in the WSN for transmitting data is within the maximum communication distance, clustering is carried out on SNs by adopting a clustering algorithm, one SN is selected from each cluster to be used as a CH, and the rest SNs are used as CMs. The CH is responsible for collecting and processing the data sent by the intra-cluster CMs and uploading the final data to the UAV. CMs are responsible for sensing environmental data and transmitting the data to the CH. And carrying out data transmission in a cluster in a single-hop mode. The charging station is responsible for charging the UAV, which is deployed where it can successfully establish communication between the UAV and the DC. The DC is responsible for reconstructing data sent by the UAVs so as to realize the functions of environment monitoring, disaster early warning and the like of the area.
1. Communication model
The ground in fig. 1 is taken as an x-axis and a y-axis, and the air is taken as a z-axis to establish a three-dimensional cartesian coordinate system. Assuming that the clustering algorithm divides the area to be covered Λ into K clusters, i.e. K CH, denoted k= {1,2, …, K, … K }, the three-dimensional cartesian coordinates of CH h are denoted as l k ＝(x k ,y k 0); the total number of CMs is N, noted as n= {1,2, N, N, three-dimensional cartesian coordinates of CM N are denoted as l n ＝(x n ,y n 0); m UAVs are deployed in the area for data acquisition, and are marked as M= {1,2, …, M, … M }, the flying heights of all UAVs are consistent, and the three-dimensional Cartesian coordinate of the UAV M is expressed as l m ＝(x m ,y m H). In this system model, only the communication links between CM and CH and between CH and UAV are considered.
In Air to Ground (A2G) communication, consider both LoS and NLoS links between UAV and CH, the average path loss between CH k and UAV m is denoted as PL m,k The required transmit power for CH k to UAV m isThe association relation between all CH and UAV is expressed as a KXM matrix omega, and the M-th column expresses the association condition of UAV M to all CH. The communication model in WSN adopts a wireless communication channel model, and the transmission power of CM n to CH k is expressed as +.>The association relation between all the CMs and the CH is expressed as a matrix y of n×k, and the kth column indicates the association of CH K to all the CMs.
2. Unmanned aerial vehicle data acquisition method based on actors-critics
FIG. 2 is a flowchart of an AC-based UAVs deployment algorithm. Improved K-mean++ based clustering algorithm for determining number K and position l of CH k After the association relation gamma between CH and CMs, the chemical problem is only related to the deployment position l of the UAV m And the association omega of UAVs with CHs. In the updated optimization problem, the deployment variable and the association variable are coupled to each other. On the other hand, UAVs are deployed in a continuous space, which results in a myriad of deployment scenarios. Therefore, the invention provides an UAV deployment algorithm based on an AC based on a DRL algorithm.
And setting rewards obtained by the intelligent agent after the state of the first step, the action is executed and the action of the first step is executed based on a standard reinforcement learning process. Specific state sets, action sets and bonus function descriptions are shown in fig. 2:
state set: definition s i For the state set of the agent in the i step, the coordinates of all UAVs are used as the combined state set of the agent and expressed as
Action set: the action set of each UAV is defined to consist of five motion directions, namely forward, backward, left turn, right turn and keep stationary, and is marked as a set A= { A 1 ,A 2 ,…,A 5 And the moving step of each action is fixed as a constant d step Thereby moving each UAVDiscretizing, in step i of training, UAV m selects an action from set A, recorded asThus, the joint action performed by all UAVs at step i is +.>
Rewarding: introduction of r(s) i ,a i ) Indicating that joint action a was performed at step i i After that, UAVs are in state s i And rewards at that time. Rewards r(s) i ,a i ) Can accurately evaluate the current state s i To the next state s i+1 Is a conversion quality of (a). It is in state s i+1 As a function of the minimum total transmit power required. Specifically, the lower the total transmit power, the more rewards are obtained. Will be in state s i+1 At the time of the optimal association schemeIn the following, the total emission power of CHs is defined as +.>The expression is as follows:
thus, the rewards may be expressed as:
from equation (6), it can be seen that the transmit power of CHs is mainly the distance d between CHs and the UAV m,k In relation, the shorter the distance, the lower the required transmit power. Thus, each CH need only find the UAV nearest to itself to associate with. But this approach may lead to UAV load imbalance, with the individual UAVs running out of charge in advance. The invention therefore proposes a differential power basedAnd (3) an association algorithm, wherein the algorithm minimizes the total transmitting power of the CHs under the condition of considering the load balancing of the unmanned aerial vehicle. The CHs firstly selects unmanned aerial vehicles to associate according to the shortest distance, and when the association number of the unmanned aerial vehicles exceeds the maximum load numberAt this time, the unmanned opportunity is associated with +.>And (3) the CHs and the rest of the CHs are matched with the rest of the UAVs again.
Furthermore, an AC deep learning algorithm is used for UAV deployment. As shown in fig. 2, the AC algorithm includes two neural networks of an actor network and a critique network, the actor network uses a policy gradient algorithm, an action is selected based on probability, the critique network uses a depth Q network algorithm, the action of the actor network is scored, and the actor network modifies the probability of the action according to the score. The algorithm is divided into three steps:
1) Initializing weight parameters of two networks, and setting the number of training rounds and the number of iteration steps in each round;
2) And (5) updating the weight parameters of the two networks in a loop iteration mode until the algorithm converges. Wherein, when calculating action rewards, adopting an association algorithm based on difference power to determine the association relation between the UAV and the CHs;
3) And after the algorithm converges, outputting the deployment position of the UAV and the association relation between the UAV and the CHs.
3. System flow
Fig. 3 is a flowchart of unmanned aerial vehicle assisted wireless sensor network data collection, which specifically includes the following steps:
s301: initializing a system;
s302: performing an improved K-mean++ based clustering algorithm, and inputting the geographic positions of all SNs and the SNs maximum transmitting power
S303: and executing an elbow method, initializing an upper limit k_max of K values of the elbow method, executing a K-means algorithm when the number of clusters is 1 to k_max, calculating SSE, recording, and finding the value of the optimal number of clusters K according to the SSE. If the optimal value is not found when the upper limit of the k value is reached, continuing to increase the upper limit value, and calculating SSE until the optimal cluster number k is found;
s304: executing a K-mean++ algorithm to obtain an optimal cluster corresponding to the K value;
s305: judging whether the transmission power from all the CM to the associated CH is smaller than the maximum transmission power, if yes, executing S307, if not, executing S306;
s306: increasing the value of k, let k=k+1;
s307: outputting an optimal k value, a CH position and an association relation of CH and CM;
s308: executing unmanned aerial vehicle deployment algorithm based on actors-critics, and inputting position l of CH k k ；
S309: initializing weights θ for actor networks π Weight θ of commentator network Q Training round number L and iteration step number I;
s310: randomly initializing a state at the beginning of each training round;
s311: action a is selected according to actor network at ith iteration i ；
S312: executing action a i Obtaining the observed state s i+1 The execution of the correlation algorithm based on the difference power is started to reward r(s) i ,a i ) And CH and UAV association;
s313: input state s i+1 Position l of lower UAV m m Position l of CH k k ；
S314: initializing: Ω=0 and is set to be equal to,all UAVs and CHs are UAVs with unconfirmed association and unassociated CH;
s315: calculating the transmitting power of each CH to all UAVs according to the formula (2), and forming a matrix pi;
s316: the unassociated CH selects UAVs with minimum transmitting power from UAVs with unacknowledged association relation according to the matrix pi to associate, and updates omega;
s317: the UAV which does not confirm the association relation obtains the associated CH number according to omega;
s318: selecting UAVs with the largest number of associated CH, if the number of associated CH is larger thanS319 is performed; otherwise, executing S322;
s319: the CH associated with the UAV calculates the difference between itself and the required minimum transmit power in the UAV and other unacknowledged associated UAVs;
s320: sorting differences from large to small, the UAV before selectionThe CH are associated;
s321: judging whether all UAVs are traversed, if so, executing S322, otherwise, checking the next UAV, and executing S317;
s322: setting the optimal incidence matrix as the current incidence matrix, namely
S323: according toCalculate->Further calculate the prize r(s) i ,a i )；
S324: calculating a time sequence difference error phi;
s325: updating actor network parameters theta π Critics network parameters θ Q ；
S326: update state s i ＝s i+1 ；
S327: judging whether i is greater than or equal to the maximum iteration step number, if so, ending the training round, executing S328, otherwise, executing the next iteration, and executing S311;
s328: judging whether the training round is greater than or equal to the maximum training round number, if so, ending the training, executing S329, otherwise, executing the next training round, and executing S310;
s329: ending training, and outputting the converged UAV position and the association relation between the UAV and the CHs;
s330: ending the operation;
finally, it is noted that the above embodiments are only for illustrating the technical solution of the present invention and not for limiting the same, and although the present invention has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that modifications and equivalents may be made thereto without departing from the spirit and scope of the present invention, which is intended to be covered by the claims of the present invention.
Claims (7)
1. The data collection method based on unmanned aerial vehicle assistance is characterized by comprising the following steps of: the method comprises the following steps:
s1: constructing a UAVs auxiliary WSN data acquisition model;
s2: designing a total emission power optimization scheme of the sensor;
s3: clustering the sensors, determining the number and the positions of the cluster head sensors, and determining corresponding cluster members;
s4: optimizing the deployment position of the unmanned aerial vehicle;
s5: and optimizing the association scheme of the cluster head and the unmanned aerial vehicle.
2. The unmanned aerial vehicle-assisted data collection method of claim 1, wherein: in the S1, a UAVs assisted WSN data acquisition model is constructed;
the model consists of an upper layer and a lower layer:
1) The upper layer is an air network and comprises a plurality of unmanned aerial vehicles, wherein the unmanned aerial vehicles fly to a designated position under the dispatching of DC and hover, and collect and store data uploaded by CHs associated with the unmanned aerial vehicles; after the acquisition task is completed, the unmanned aerial vehicle carries data and returns to a DC designated position for charging, and meanwhile, the data is forwarded to the DC;
2) The lower layer is a ground layer and consists of a WSN, a charging station and DC;
in order to ensure that the distance of SNs in WSN for transmitting data is within the maximum communication distance, clustering SNs is carried out by adopting a clustering algorithm, one SN is selected from each cluster to be used as CH, and the rest SNs are used as CMs; the CH is responsible for collecting and processing data sent by CMs in the cluster and uploading the final data to the UAV; CMs are responsible for sensing environmental data and transmitting the data to the CH; the data transmission is carried out in a cluster in a single-hop mode; the charging station is responsible for charging the UAV, which is deployed where the UAV can successfully establish communication with the DC; the DC is responsible for reconstructing data sent by the UAVs so as to realize environmental monitoring and disaster early warning of the area.
3. The unmanned aerial vehicle-assisted data collection method according to claim 2, wherein: in the step S2, a total emission power optimization scheme of the sensor is designed; the UAVs set is M, the CHs set is K, and the CMs set is N; considering two links of line-of-sight LoS and non-line-of-sight NLoS between the UAV and the CH, and adopting orthogonal multiple access between the UAV and the CH to avoid interference; the average path loss between UAVm and CH k is expressed as:
wherein eta LoS And eta NLoS The average additional loss of the LoS link and the NLoS link is different from scene to scene; a and b are S curve parameters; h is UAV flying height; f is the carrier frequency; d (D) m,k Representing the distance between CHk and UAVm's perpendicular projection on the ground; c is the speed of light; d, d m,k A linear distance of CHk to UAVm;
the required transmit power for CH k to UAV m is as follows:
wherein delta m,k The correlation coefficient from CH k to UAV m is 0 and 1, delta m,k CH k is associated with UAV m when=1,for received power of UAV m, PL m,k Is the average path loss; the association relation between all CH and UAV is expressed as a KxM matrix omega, and the M-th column expresses the association condition of UAV M to all CH;
the communication model in the WSN employs a wireless communication channel model, and the transmission power of CM n to CHk is expressed as:
wherein θ n,k The correlation coefficient of CM n and CH k is 0 and 1, theta n,k When=1, CM n is associated with CH k, and vice versa;the received power for CH k; d, d n,k Represents the distance between CM n and CH k; l is a transmitted system loss factor, and L is more than or equal to 1; g t And G r Gain of CM transmit and receive antennas, respectively; lambda is the wavelength; the association relation between all the CM and the CH is expressed as a matrix gamma of N multiplied by K, and the K column expresses the association condition of CH K to all the CM;
the method comprises the steps of providing a sensor total emission power optimization scheme which jointly optimizes the number and the positions of CH, the deployment position of UAVs and the association of the UAVs and CHs, minimizing the total emission power of SNs in a WSN, wherein the optimization targets are as follows:
one CH can only be associated with one UAV, and one CM can only be associated with one CH; meanwhile, the transmitting power of CH cannot exceed the maximum transmitting power of CH, and the transmitting power of CM cannot exceed the maximum transmitting power of CM; the deployment location of each UAV and the location of CH are ensured to be within the target area Λ.
4. A data collection method based on unmanned aerial vehicle assistance according to claim 3, wherein: in the S3, firstly, evaluating error square sum SSE of K-means algorithm clustering results through an elbow method to find an optimal clustering number K; as the clustering number k increases, the SSE index gradually decreases as the clustering effect is better; in the process of increasing k, when the k value is smaller than the optimal cluster number, the descending speed of SSE is fast, and when the k value exceeds the optimal cluster number, the descending speed of SSE is slow; an elbow is imaged by the relation between SSE and k, and the k value corresponding to the elbow is the optimal cluster number;
after finding out the optimal cluster number, using K-mean++ algorithm to find out the optimal cluster when the corresponding K value, judging whether the transmission power from all CM to its associated CH is less than the maximum transmission power, if yes, outputting the optimal K value, CH position and the association relationship between CH and CM, if not, adding K value to execute K-mean++ algorithm cluster again until the transmission power from all CM to its associated CH is less than the maximum transmission power, outputting the optimal K value, CH position l k And the association of CH with CM.
5. The unmanned aerial vehicle-assisted data collection method of claim 4, wherein: in S4, when determining the number k of CH and the position l k After the correlation y between CH and CMs, the chemical problem is only related to the deployment position l of UAV m Related to the relationship Ω of UAV and CHs; the optimization problem is updated as follows:
based on the DRL algorithm, an UAV deployment algorithm based on AC is provided, a deployment process of the UAV is modeled as an MDP process, and a deployment position of the UAV is determined through centralized training.
6. The unmanned aerial vehicle-assisted data collection method of claim 5The method is characterized in that: in the step S5, during the training process, a balance power correlation algorithm is provided to determine rewards under each training state, and the algorithm makes the total transmitting power of CHs minimum under the condition of considering unmanned aerial vehicle load balancing; the CHs selects unmanned aerial vehicle to correlate according to the shortest distance, and when the correlation number of unmanned aerial vehicles exceeds the maximum load numberWhen the unmanned aerial vehicle calculates the required transmitting power related to other unmanned aerial vehicles, the unmanned aerial vehicle selects the +.>Each CHs establishes an association and the remaining CHs re-match the remaining UAVs.
7. The unmanned aerial vehicle-assisted data collection method of claim 6, wherein: in the step S4, an AC deep reinforcement learning algorithm is used for deployment of the unmanned aerial vehicle; the critic network evaluates the action selected by the actor network by means of a state-action-value function, the actor network, in each iteration, being based on the current state s i And policy pi i Selecting a corresponding action a i Training by adopting a Monte Carlo strategy gradient reinforcement learning method; wherein, when calculating action rewards, adopting the correlation algorithm based on the difference power proposed in the step S5 to determine the correlation relationship; and after the algorithm converges, outputting the deployment position of the UAV and the association relation between the UAV and the CHs.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310591381.4A CN116634391A (en) | 2023-05-24 | 2023-05-24 | Unmanned aerial vehicle-assisted data collection method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310591381.4A CN116634391A (en) | 2023-05-24 | 2023-05-24 | Unmanned aerial vehicle-assisted data collection method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116634391A true CN116634391A (en) | 2023-08-22 |
Family
ID=87636114
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310591381.4A Pending CN116634391A (en) | 2023-05-24 | 2023-05-24 | Unmanned aerial vehicle-assisted data collection method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116634391A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117241240A (en) * | 2023-09-22 | 2023-12-15 | 重庆邮电大学 | Data compression and scheduling method for unmanned aerial vehicle communication system |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109451556A (en) * | 2018-11-28 | 2019-03-08 | 广州大学 | The method to be charged based on UAV to wireless sense network |
| KR20210007654A (en) * | 2019-07-12 | 2021-01-20 | 한국전자통신연구원 | Method for operating uav-assicted wireless sensor network and apparatus thereof |
| CN113131985A (en) * | 2019-12-31 | 2021-07-16 | 丽水青达科技合伙企业(有限合伙) | Multi-unmanned-aerial-vehicle data collection method based on information age optimal path planning |
| CN114980020A (en) * | 2022-05-17 | 2022-08-30 | 重庆邮电大学 | Unmanned aerial vehicle data collection method based on MADDPG algorithm |
-
2023
- 2023-05-24 CN CN202310591381.4A patent/CN116634391A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109451556A (en) * | 2018-11-28 | 2019-03-08 | 广州大学 | The method to be charged based on UAV to wireless sense network |
| KR20210007654A (en) * | 2019-07-12 | 2021-01-20 | 한국전자통신연구원 | Method for operating uav-assicted wireless sensor network and apparatus thereof |
| CN113131985A (en) * | 2019-12-31 | 2021-07-16 | 丽水青达科技合伙企业(有限合伙) | Multi-unmanned-aerial-vehicle data collection method based on information age optimal path planning |
| CN114980020A (en) * | 2022-05-17 | 2022-08-30 | 重庆邮电大学 | Unmanned aerial vehicle data collection method based on MADDPG algorithm |
Non-Patent Citations (3)
| Title |
|---|
| GICHEOL WANG: "UAV-Assisted Cluster Head Election for a UAV-Based Wireless Sensor Network", 《2018 IEEE 6TH INTERNATIONAL CONFERENCE ON FUTURE INTERNET OF THINGS AND CLOUD (FICLOUD)》, 11 September 2018 (2018-09-11) * |
| 兰婷婷: "基于无人机辅助的WSN数据采集机制研究", 《中国优秀硕士学位论文全文数据库》, 15 December 2022 (2022-12-15) * |
| 黄晓舸: "无人机群辅助的数据采集能耗优化方法", 《电子信息学报》, 5 September 2022 (2022-09-05) * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117241240A (en) * | 2023-09-22 | 2023-12-15 | 重庆邮电大学 | Data compression and scheduling method for unmanned aerial vehicle communication system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Wei et al. | UAV-assisted data collection for Internet of Things: A survey | |
| Masroor et al. | Resource management in UAV-assisted wireless networks: An optimization perspective | |
| CN111381499B (en) | Internet-connected aircraft self-adaptive control method based on three-dimensional space radio frequency map learning | |
| Sun et al. | Time and energy minimization communications based on collaborative beamforming for UAV networks: A multi-objective optimization method | |
| CN113433967B (en) | Chargeable unmanned aerial vehicle path planning method and system | |
| CN113543066B (en) | Integrated interaction and multi-target emergency networking method and system for sensing communication guide finger | |
| CN114690799A (en) | Air-space-ground integrated unmanned aerial vehicle Internet of things data acquisition method based on information age | |
| CN113131985B (en) | Multi-unmanned-aerial-vehicle data collection method based on information age optimal path planning | |
| Wu et al. | Load balance and trajectory design in multi-UAV aided large-scale wireless rechargeable networks | |
| Zhu et al. | Path planning of multi-UAVs based on deep Q-network for energy-efficient data collection in UAVs-assisted IoT | |
| CN114142908B (en) | Multi-unmanned aerial vehicle communication resource allocation method for coverage reconnaissance task | |
| CN116634391A (en) | Unmanned aerial vehicle-assisted data collection method | |
| CN112817327A (en) | Multi-unmanned aerial vehicle collaborative search method under communication constraint | |
| CN111163479B (en) | Node optimization deployment method suitable for wireless locatable sensor network | |
| CN113485409A (en) | Unmanned aerial vehicle path planning and distribution method and system for geographic fairness | |
| CN113625733A (en) | DDPG (distributed data processing) -based multi-target three-dimensional unmanned aerial vehicle path planning method | |
| Liu et al. | A distributed deployment algorithm for communication coverage in wireless robotic networks | |
| Akin et al. | Multiagent Q-learning based UAV trajectory planning for effective situationalawareness | |
| Pan et al. | Fresh data collection for UAV-assisted IoTs based on proximity-remote region collaboration | |
| Shin | Energy-efficient multidimensional trajectory of UAV-aided IoT networks with reinforcement learning | |
| Xu et al. | Aerial edge computing: Flying attitude-aware collaboration for multi-UAV | |
| Cicek et al. | 3-D dynamic UAV base station location problem | |
| CN117369485A (en) | Unmanned plane path collaborative planning method and device, electronic equipment and storage medium | |
| CN113885566B (en) | V-shaped track planning method oriented to minimization of data acquisition time of multiple unmanned aerial vehicles | |
| CN112867023B (en) | Method for minimizing perception data acquisition delay through dynamic scheduling of unmanned terminal |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |