Intelligent cross-layer adaptation scheme for wireless links with QoS-guaranteed traffic

Quality-of-service (QoS)-guaranteed traffic is of critical importance in real-time multimedia communications. In this paper, a novel approach to systematic cross-layer monitoring, learning, and optimization using artificial neural networks is developed for wireless links with QoS-guaranteed traffic. The proposed method is used to jointly optimize the design parameters of the adaptive modulation and coding (AMC) scheme at the physical layer with the truncated automatic repeat request (ARQ) protocol at the data link layer. The structure, performance, and cooperation of the artificial neural networks used in the proposed adaptive system are disclosed in detail and the capability of the proposed method in maximizing the overall system throughput and maintaining the QoS guarantees in non-stationary wireless links is discussed and verified by a set of illustrative simulation results.