Channel Allocation In Wireless Mesh Networks Based On Partially Overlapped Channels

Wireless Mesh Network has become an important networking form for wireless access in emergency communication scenarios because of its features of multi-hop,self-organization,non-visual transmission,as well as its excellent performance in coverage,cost control and networking speed.Channel interference is the main factor affecting network performance.Although the introduction of Multi-Radio Multi-Channel technology can significantly reduce co-channel interference,it also brings problems such as spectrum utilization and link conflict.In the wireless network environment where available spectrum resource is scarce,the rational use of some overlapping channels can improve the spectrum utilization rate and network capacity to a certain extent.In this paper,based on the different traffic distribution characteristics and network performance requirements of different wireless access scenarios,the partially overlapping channel allocation methods in Wireless Mesh Networks are studied.Firstly,a minimization interference channel allocation method based on the improved bat algorithm is proposed to solve the problem that expansion in nodes leads to an increase in network interference links and a low efficiency of channel allocation in traffic decentralized mode.An optimization function aimed at minimizing link-weighted interference is designed,taking into account the number of nearby links and channel interference.The static channel allocation process is transformed into a linear programming model with constraints,and solved by the improved discrete bat algorithm.The chain behavior of salp group is introduced to improve the local search ability of the algorithm,and the dynamic inertia weight is designed to effectively balance the global search and local development ability of the algorithm.Simulation results show that the proposed algorithm can reduce global link interference and improve channel allocation efficiency with low time complexity.Secondly,a potential game channel allocation method based on an incentive mechanism is proposed to solve the problem of unfair bandwidth reservation and uneven load distribution in traffic aggregation mode.The minimum spanning tree topology control algorithm is used to delete redundant links and balance load.The channel allocation problem is transformed into a game model,the link priority factor is designed while taking into account the service load and the network factors that affect the fairness in the topology structure,the benefit function of the game model is established by combining the interference measurement,and it is proved that the proposed game is an exact potential game with Nash equilibrium point.On this basis,a better response Nash equilibrium method based on incentive mechanism is introduced to improve the speed of game convergence.Simulation results show that the scheme can effectively reduce network interference and achieve fair channel allocation under different network sizes.Finally,an adaptive channel allocation algorithm based on multi-agent reinforcement learning is proposed to solve the problem that static channel allocation in traffic awareness mode cannot adapt to traffic changes in the network environment,resulting in the degradation of link communication quality.Under the circumstance of unknown service demand modification,the channel assignment problem is mapped to the multi-agent reinforcement learning process.The state space based on interference,load and other factors is established,and the relationship between exploration and utilization in the algorithm is balanced by the improved greedy action selection mechanism.At the same time,the return function based on link satisfaction is designed to guide the channel selection to improve the direction of link communication quality.Simulation results show that the proposed algorithm can effectively adapt to the traffic and source node changes in the network and guarantee the quality of user service.