Overhead Analysis for Radio Environment Mapenabled Cognitive Radio Networks

Abstract

This paper presents a novel and general approach to cognitive radio (CR) networking based on the Radio Environment Map (REM). REM is envisioned to be an integrated database that consists of comprehensive multi-domain information for CR, such as geographical features, available services, spectral regulations, locations and activities of radio devices, policies, and past experiences. Disseminating and sharing REM information offers a proper vehicle of CR network support, which can be exploited by cognitive engine (CE) for most cognitive functionalities such as situation awareness, reasoning, learning, planning and decision support. Tradeoffs have to be made between the performance gain and the cost of overhead. This paper focuses on analyzing the overhead associated with REM dissemination under various scenarios. With analytical models and network simulations, it is shown that the overhead of REM dissemination can be significantly reduced by extending the optimized link state routing protocol (OLSR). Application-specific ad hoc methods have also been proposed and can be employed to further reduce the overhead. Simulations are presented, comparing the overhead of REM dissemination for different network size, topology, node density and mobility. Preliminary results show that the speed of wireless nodes has little impact to the load of overhead if the REM dissemination rate is fixed. The size of REM information element is estimated for the emerging cognitive wireless regional area networks (IEEE 802.22 WRAN).