Cross-Layer Architecture in Cognitive Ad Hoc Networks

In this paper, we proposes a cognitive cross-layer architecture for the cognitive ad hoc networks, where each node in the networks can sense and learn from wireless environment, network environment and user´s requirements, and then it will respond the changes of environment by adapting system parameters and network services based on the learned results. In this way, it can provide the end-to-end goals of a data flow while fully utilizing network resources. The core of the proposed architecture is a cognitive module. The proposed architecture can support the design requirements of cognitive ad hoc networks, such as autonomy, self-configuration, self-organization, scalability, adaptability, and simplicity, to provide multimedia services such as video, voice, and emails. The implementation of the proposed cognitive crosslayer architecture is briefly illustrated in this paper with an example of using case-based reasoning. We will research the related algorithms which the cognitive module needs, and the end-to-end performance optimization of cognitive ad hoc networks.