USRP-based experimental platform for energy detection in cognitive radio systems

Cognitive radios grant opportunistic access to underutilized frequency bands for secondary users and assure minimum interference for primary users' transmissions through a series of measurements of the frequency environment, the first and foremost of them being spectrum sensing. Amongst the main categories of spectrum sensing algorithms, energy detection is considered an approachable solution, mainly due to its low computational simplicity. In this paper, we propose and analyze a practical implementation of an energy detector using Universal Software Radio Peripheral (USRP) software defined radio (SDR) platforms. Signal acquisition is carried out in GNU Radio Companion (GRC) and the received signal power level is calibrated using the MATLAB environment for a test signal transmitted in the 470 - 870 MHz frequency band. The Receiver Operating Characteristic (ROC) performance is estimated for the implemented energy detection scheme and it is further compared to the corresponding theoretical performance for a specified Signal-to-Noise Ratio (SNR) value.