Research On Channel Allocation Strategy For Wireless Ad Hoc Networks

Channel allocation is a momentous research direction in wireless ad hoc networks.Its process includes three steps: access control,resource allocation and channel adjustment.Due to the limitation of channel resources,it is a challenge to allocate channel resources fairly and reasonably for network nodes.Moreover,the development of multi-radio and multi-channel technology makes the problem of channel allocation more complex.Through the research on the development status of channel allocation strategy in wireless ad hoc networks,this paper summarizes three problems of channel allocation,which are inter-channel interference,low channel utilization and difficult to guarantee the quality of service(QoS).Accordingly,specific solutions are given.Aiming at the problem that the node request for idle channels is rejected in the process of access control,this paper proposes a dynamic channel allocation strategy for QoS enhancement(DCASQE)to reduce the rejection rate of idle channel requests.Firstly,the channel availability model is modeled as a Continuous Time Markov Chain in a bid to calculate the channel availability which is used to compute the channel availability factor(CAF)and channel availability threshold(CAT).Then,the dynamic channel allocation process is achieved based on the comparison result of CAF and CAT.Ultimately,the earliest deadline first algorithm is exploited to ensure the QoS and the packet segmentation algorithm is used to enhance the QoS.NS2 simulation results prove that the DCASQE increases the packet delivery ratio and throughput,decreases the end-to-end delay and total overhead and shows a QoS enhancement than existing algorithms.Aiming at the problems of channel utilization reduction due to node selfishness and the inter-channel interference in the process of resource allocation,this paper advances a dynamic channel allocation based on reinforcement learning(DCARL)algorithm.Firstly,the selfishness of nodes is described by the proportion of channel resource allocation.Secondly,the interchannel interference factor is used to quantify the degree of inter-node interference.Finally,an optimization model is established to improve the channel utilization and reduce the interference.The reinforcement learning method is applied to select the optimal channel resource allocation ratio.NS2 simulation results show that different combinations of channel resource allocation ratios and inter-channel interference have different network performances.Selecting the optimal channel resource allocation ratio can reduce the node selfishness,degrade the interchannel interference and upgrade the channel utilization to a certain degree.To sum up,the algorithms proposed in this paper can ensure and enhance the QoS,reduce inter-channel interference and improve the channel utilization.