Resource Allocation in UAV-assisted Mobile Edge Computing Systems

In a crowdsensing system, effective transmission of sensing data is an important step in improving task distribution efficiency. In real-life scenarios, when resources are limited or geographic conditions are harsh, Unmanned Aerial Vehicles (UAV) can often be used for data transmission. This paper proposes a group intelligence sensing task distribution method under dynamic topology conditions to solve the problems of unreachable links or high network overhead caused by the dynamic topology of UAVs. This method uses the Markov model to model the data transmission process, and on this basis proposes an improved task distribution method based on topology awareness to achieve efficient task forwarding. The final experimental results show that compared with the traditional self-organizing network data transmission scheme for opportunistic networks, the scheme proposed in this paper can better solve the problems of routing voids and blind paths, and effectively improve the network lifetime while realizing a lower end-to-end delay, with better transmission performance.