Research On Unmanned Aerial Vehicle Deployment As Communication Relays

Communication relay can improve the communication performance between nodes,so as to extend mission coverage,enhance their cooperation ability,and optimize transmit power.Compared to fixed infrastructures,satellites and ground mobile platforms,unmanned aerial vehicle(UAV)has shown excellent maneuvering properties,dynamic networking and environment adaptabilities,using UAV as communication relay in wireless sensor networks(WSNs),disaster rescue and cooperative combat becomes research hotspots in relay communication area,therefore how to deploy UAV platforms turns to be a challenge we have to face.This thesis studies the relay UAV deployment problem from pre-deployment,redeployment and dynamic deployment three aspects based on the uncertainty of user environment,where signal strength is used as communication performance metric and optimized by taking advantage of UAV maneuvering ability,as well as considering minimum UAV number and time cost.The practical work is as follows:(1)To address the problem that current researches on relay UAV pre-deployment cannot realize the tradeoff between UAV number and global signal strength objective,an improved clustering algorithm BPN is proposed to avoid the over inclination to any of them,and simultaneously optimize to generate a reasonable deployment result with the desired tradeoff.Compared to the original K-means++,and fast search and find of density peaks based clustering algorithms,the proposed algorithm reduces the cluster number by extending potential deployment positions and building a local density function with reflected channel properties,the proposed algorithm also avoids unreasonable clustering caused by their similarity by using centroid-based iteration update strategy.Simulation results show that the proposed algorithm can achieve near optimal result with faster calculation speed than K-means++ based algorithm,and uses less UAV number than fast search and find of density peaks based algorithms,these demonstrates the pre-deployment is reasonable.(2)To address the problem that current researches on relay UAV re-deployment have not synchronously considered communication performance,UAV number and deployment time cost under pop-up cases,a polynomial time algorithm is proposed to reduce deployment time cost and generate a near optimal solution with fast calculation under the condition that minimum received signal strength satisfies application requirements.The algorithm includes two stages:the first stage generates the minimum UAV number and the set of all possible re-deployment positions for each hop,the second stage reduces re-deployment time cost by using a greedy strategy so that the hop hardest to arrive can be handled first.Simulation results show that the proposed re-deployment algorithm costs lower deployment time compared to take precedence over optimal communication performance.Simulation results also show that the proposed redeployment algorithm generates better deployment result than stochastic sampling algorithm,as well as costs less time than genetic algorithm.(3)To address the problem that current researches on relay UAV dynamic deployment cannot operate online and autonomous with only use received signal strength and positon information in unknown environment for serving mobile users,a gradient based algorithm is proposed to drive the UAV towards optimal relay position serving mobile users with unknown channel parameters,unknown angle of arrival signal.The algorithm includes four parts: user state and position prediction,online wireless channel parameters estimation,optimal relay position generation,and guidance law.Simulations results with using airborne user,ground user,end-to-end communication and multi-user-inter communication show that the proposed algorithm could drive the UAV to and track the optimal relay position,as well as achieve near optimal communication performance.(4)To address the present application researches situations on deploying UAV as communication relay supporting multiple combat node cooperatively search unkown battlefields,this thesis considers the scenarios of using relay UAV supporting multiple unmanned reconnaissance vehicles cooperatively search unknown battlefields.A cooperative search model is built,different relay UAV deployment schemes in unkown environment with limited information are built,and their affections on environment uncertainty reduction are studied.Simulation results validates the effectivenesses of the former proposed relay UAV deployment methods,and they also show that deploying UAV as communication relay improves the information exchange capabilities among these unmanned reconnaissance vehicles,achieves lower environment uncertainty than none relay UAV deployment,and therefore mission efficiency of multiple unmanned reconnaissance vehicles cooperative search are improved.