Research On Data Collection Algorithms In UAV-Assisted Wireless Sensor Network

Benefited from the characteristics of high flexibility,easy deployment and low cost,the application of unmanned aerial vehicles(UAVs)in wireless communication systems has attracted extensive attention.With the rapid development of wireless sensor and microelectronic technologies,the application fileds of wireless sensor network(WSN)are gradually expanding,such as environmental monitoring and smart home.In WSNs,sensor nodes(SNs)perceive data from the environment,and then transmit their data to the sink node.By leveraging the mobility,flexibility and line of sight(Lo S)characteristics of UAV,efficient UAV-assisted data collection can be achieved in WSN.In the UAV-assisted WSN data collection system,how to design reasonable and efficient data collection algorithms by jointly considering data transmission requirements and the characteristics of air-ground channel and UAV is an important issue to be solved.This thesis studies the data collection algorithms for UAV-assisted WSN.The main contents are as follows.For the network scenario with one UAV,one base station and multiple SNs,the clustering and UAV flight trajectory optimization problem is formulated as a data collection time minimization problem by considering the constraints of data transmission requirements,available network resources and the characteristics of UAV.Since the optimization problem is a mixed integer nonlinear programming(MINLP)problem,which is difficult to solve directly,this thesis decomposes the original optimization problem into subproblems,i.e.,clustering subproblem,cluster head(CH)transmission mode selection subproblem,UAV trajectory design subproblem and UAV velocity optimization subproblem.For the clustering subproblem,a modified K-means algorithm-based clustering strategy are proposed.Based on the obtained clustering strategy,a heuristic algorithm is proposed to determine the CH transmission mode selection,and a simulated annealing algorithm-based UAV trajectory design algorithm is then presented.Then,based on the optimal UAV flight trajectory,the UAV velocity optimization subproblem is modeled as a linear programming problem,and the convex optimization theory is adopted to solve it.Finally,the effectiveness of the proposed algorithm is verified by MATLAB simulation.For the UAV-assisted WSN data collection scenario that supports the dynamic arrival of data packets,under the considerations on data freshness,the limitation of data packet queues and the characteristics of UAV,the network cost is defined as the weighted sum of the average age of information(Ao I)of SN,the average number of dropped packets and the energy consumption of UAV.Then,joint data packets scheduling,UAV flight trajectory and velocity optimization problem is modeled as network cost minimization problem by considering the constraints of data transmission requirements and the characteristics of UAV.Since the formulated problem is an MINLP problem,and is not easy to solve directly,an alternating iteration algorithm-based data collection algorithm is proposed to determine the joint optimization strategy.Firstly,given the UAV flight trajectory and velocity,a packet deadline and data volume-based packet scheduling strategy is proposed.Given the packet scheduling strategy and UAV flight velocity,the UAV flight trajectory optimizaiton subproblem is modeled and solved.Specifically,the SN scheduling strategy is firstly determined,and then a dynamic programming(DP)-based UAV flight trajectory design algorithm is proposed.Then,based on the packet scheduling strategy and UAV flight trajectory,the binary search algorithm is used in this thesis to determine the optimal velocity of UAV.The above steps repeat until the algorithm achieves convergence.Simulation results verify the effectiveness of the proposed algorithm.