Cellular Automata Approach for a Low Power Fusion Center to Evaluate Spectrum Status and Coverage Area in Cognitive Radios

摘要

Efficient spectrum sensing is an important requirement for the success of the cognitive radio (CR) system. In an external sensing scenario an external agent performs the sensing and broadcasts the channel occupancy information of Primary Users (PU) to Secondary Users (SU). Cellular Automata (CA) is a discrete model used to develop wide variety of applications. CA based architectures have already proved its utility in the low power and high speed VLSI designs. In this paper a novel data fusion approach based on CA is proposed for external sensing. Wireless sensors network is deployed in the field, to sense the spectrum hole. Individual sensing result are sent to the Central Node (CN) for making the final decision. CN is considering a larger area at a time and it is important to understand the coverage area of the PU. Since we are looking for energy efficient and low cost network installation, sensing results are expected to get affected by noise, fading, and shadowing. CA based fusion rule at the CN evaluates spectrum status and the coverage area of the PU. Its performance is compared with fuzzy based methods and weighted combining methods. Performance on coverage area formation, probability of detection, false alam rate and computational cost are analysed. It is found that CA based approach is giving a better performance in all these cases. Since CA based architectures are reported to have reduced power consumption and better speed in its category of circuits, our proposed approach may be recommended for the implementation of a lowpower core at the CN for decision fusion.