An adaptive threshold energy detection mechanism using GNU Radio on USRP

Cognitive Radio (CR) effectively addresses the under-utilization of spectrum in the wireless environment. Energy Detection (ED), a spectrum sensing mechanism, plays a vital role in identifying the spectrum holes. However, the ED performance and the static threshold are negatively affected by uncertainty in the noise power. In this paper, we propose an adaptive ED mechanism using dynamic threshold detection. The mechanism calculating the estimated noise power from the received random samples. This calculation is iterated for minimizing the false alarm rate and maximizing the detection rate of the Primary User (PU) signal. The proposed model is implemented using ISM 2.4GHz frequency bands in USRP NI-2932 and GNU radio processing blocks. In our results, the sensing errors are minimized, and the sensitivity of ED is also improved by mitigating the adverse effects of the noise.