A QoS controlled spectrum switching resource allocation technique for cognitive Wi-Fi networks

Abstract

Cognitive networks (CN) introduce an adaptable architecture that enables efficient sharing of transmission resources among heterogeneous networks. While quality-of-service (QoS) provisioning is always an important issue, the channel switching frequency at the secondary networks has a clear impact on the QoS performance of a CN. A secondary network returns a borrowed channel when demanded by the primary network. In this paper, we propose i) a novel secondary channel allocation algorithm and ii) a flexible CN architecture for IEEE802.11 based WLANs. The proposed CN architecture utilizes a Geo-location Database (GDB) as the repository of shared channel information that can be used by secondary networks. The proposed channel allocation algorithm allocates shared spectrum to secondary networks by matching the available channel characteristics with the secondary network's requests to maximize the QoS of traffic in secondary networks. An OMNET++ based simulation model is used to analyse the performance of the proposed network. Simulation results show that the proposed channel allocation algorithm offers significant QoS advantages and stability of secondary networks compared to conventional channel sensing and GDB based secondary channel allocation approaches.