Since its conception nearly two decades ago, cognitive radio (CR) has been the topic of numerous research studies in different areas. CR is considered to be a viable and an important part of the wireless networks of the future because it can allow for a more efficient spectrum utilization and an increase in the overall system throughput. CR devices are envisioned to provide new services and even operate within the coverage of different technologies, to cooperate with the users of their networks, since their functional frequency and modulation are programmable
The rise of the cognitive radio systems as a concept for future networks has seen a great amount of scientific effort in the recent years. Appropriately, much attention is given to how the vital function of spectrum sensing should be executed. The cognitive radio device is required to be able to evaluate the spectral environment properly so that it may not create additional interference to the primary users. The task is further complicated by the need of optimization of the speed of the process so that the spectrum holes can be utilized. The sensing accuracy and sensing time are conflicting parameters, therefore, a suitable trade-off is necessary for an optimal efficiency. We propose a dual-approach solution. The decision about the spectrum occupancy is made using the measured signal-to-noise ratio (SNR) and the received signal levels as inputs in a fuzzy logic algorithm. The result is then compared with the one acquired using the statistical method. Finally, an optimal balance between the sensing time and accuracy is obtained for the current environmental conditions using the derived closed form expression. The algorithm has been practically implemented using a software defined radio platform comprising USRP and GNU Radio. Through simulation results, we have shown the efficiency of our proposal in relation to other existing methods. The performance of the practical implementation has also been analyzed.