Wireless Environment Resource Allocation Algorithm Based On Cross-layer Research

The evolution of wireless communication results in a vigorous growth of multimediaservice which means the wireless communication services turn to multimedia data service,high speed image transmission and so on rather than low speed voice service step by step.While the contradiction among the limitation of the radio resources, the complexity ofwireless communication environment and the severe demand for quality of service formultimedia services deepens, the disadvantage of traditional layered structure emerges.Therefore, in order to improve the efficiency of resource utilization, the wireless resourceallocation scheme that using the resource distribution of the physical layer (PHY) and datascheduling of the media access (MAC) layer has been widely studied.Firstly, we propose a queue-based cross-layer resource allocation algorithm,whichbased on the physical layer feedback channel state information and the user’s the cachequeue status information of the cross-layer module of the system, for subcarrier and bitallocation. This algorithm fully takes advantage of the greedy algorithm, selects optimaluser through channel state information and queue cache status information to avoid thewaste of resources,and after iterations，we can receive the final completion of the resourceallocation process. Simulation results show that compared with other non-cross-layeralgorithm, the proposed algorithm can effectively avoid the waste of resources, improvesystem performance and obtain lower power consumption.Then, a cross-layer resource allocation algorithm with packet dependent scheduling isproposed. According to the size of the packet, arrival time, QoS requirements and soon,We build a package independent scheduling model that the higher priority packets canbe serviced first, regardless of whether they are the same queue.After that,we decoupleresource allocation process into subcarrier allocation and power allocation,the former isbased on the user fairness and resource constraints while the latter utilizes Lagrange method. Simulation results show that compared to the scheduling algorithm with queuevalue, the proposed algorithm can effectively improve the system throughput and enhancethe efficiency of the system.