Research On Transmission Capacity Optimization In Flying Ad-Hoc Network Via Channel Hopping Access

Flying Ad Hoc Network is a distributed communication system for multiple unmanned aerial vehicles to realize multi-hop transmission based on ad-hoc topology.The transmission requires to be real-time and highly-reliable.In the case of large geographic scale and network topology radius,the end-to-end delay of data link should be within 1?100 microseconds.Also high data payloads are generated in UAV collaborative scenario,arising congestion and decreasing network throughput.In order to maintain the successful probability of packet transmission,the TDMA-like scheduling media access control protocol allocates frequency resources and reserve time slots by increasing RTS/CTS frames.However,the timeliness cannot be guaranteed for the access of high dynamic nodes.The statistical priority media access control protocol proposed by Tactical Targeting Networks Technology(TTNT)does not rely on reply request frames and is implemented through asynchronous frequency hopping and carrier listening mechanism.In this paper,the multi-channel frequency hopping access mechanism is studied,and channel rendezvous process is introduced as a supplement.In order to improve the transmission capacity of flying ad hoc network,we analyze the optimal bandwidth partition granularity.To reduce random access delay,we propose an adaptive frequency hopping scheme based on quorum hopping set construction.Our analysis is an extension from the classical transmission capacity analysis theory,by modeling channel fading distribution for UAV datalink.Especially,the concurrent transmission interference model characterized by shadow fading channel and Rice channel is built,thus the transmission capacity is represented by outage probability on the receiver considering UAV displacement density.The contention density for UAV with regard to the number of sub-channels is derived for a given outage probability.The goal is to find a frequency reuse factor for an optimal expected contention density,which is the objective function.We find the solution for optimal channel numbers by Monte Carlo simulation,while the search range is limited by the number of nodes in the network.When calculating the expected value of the objective function,Markov Chain Monte Carlo sampling method(MCMC)is used to sample the UAV channel statistical distribution to obtain fading coefficient,and the efficiency of Metropolis Hastings sampling and Hamilton Monte Carlo sampling are compared.The delay of multi-channel frequency hopping access mainly comes from the TTR(Time-To-Rendezvous)of channel hopping algorithm.The complexity of TTR is related to the size of local hopping sets and their intersection proportion.It is assumed that the frequency hopping set will be shrink according to the perceived channel occupancy ratio.However,in order to ensure the existence of a common channel between any two nodes,a frequency-hopping set construction scheme based on Quorum set is proposed in this paper.For any number of sub-channels,the risk of non-intersection can be eliminated.As the number of concurrent transmission node pairs is guaranteed,the transmission capacity of the system is improved.