Research On MAC Protocol And Path Planning Method For UAV In UAV-WSN Systems

UAVs (unmanned aerial vehicles) have some advantages in retrieving theinformation from the wireless sensor network (WSN). Meanwhile the UAV-WSNsystem has some differences with conventional WSN system. So it is important todesign an appropriate scheme to implement information collection in a WSN-UAVsystem.In this paper we proposed a protocol called A-OAloha (Adaptive-OpportunisticAloha) for WSN-UAV system to guarantee the efficiency of it. This protocol was across-layer design and was designed with the consideration of the distribution of thesuccessful sensors, energy consumption and the transmission efficiency. For thedistribution of the successful sensors, we design a mechanism based on the priority tomake the distribution more uniform and the priority could be adaptively changed. Forthe consideration of energy-efficiency we choose to use the method that make thesensors go into sleep when they do not need to send data until being waked up by thebeacon signal from the UAV. And for the transmission efficiency we add acommunication stage in every interval based on the O-Aloha scheme to enlarge thethroughput under an acceptable system bit error rate (BER). We use the methods ofcalculating and simulation in our work to show the advantages of our design.In this paper we also focus on the information collecting method for UAV in WSNsystems. In our study, we take both the kinematic constraints of UAV and the energyconsumption of sensors into account of the method. We proposed a path planning forKinematic-Constrained UAV for information collection based on life time of WSNs.Our purpose is to find a path for the UAV to collect information from some nodes in thetarget area while minimizing the cost function under kinematics constraints. In thispaper the cost function is the weighted sum of the total length of the whole path and thetotal weights of all the nodes. We solve the problem by using a path fix algorithm. Toevaluate our method, we also use ant colony algorithm to generate a base TSP path anda TSP path with kinematic constraints. In our simulation, we can find that our algorithmreduce the total length of the path while has the similar value of the cost function withthe other two algorithm. In addition, our algorithm also spends much less time than theTSP path with constraints, which shows the feasibility of our algorithm.