Performance Evaluation of Spectrum Sensing Implementation using Artificial Neural Networks and Energy Detection Method

摘要

A Cognitive Radio (CR) is a radio that is able to sense the spectral environment over a wide frequency band, then use it temporarily without causing any interference to the primary user (PU). The spectrum sensing operation (SS) is one of the most challenging issues in cognitive radio systems. In this paper, we are interested in the implementation of the spectrum sensing operation, using a real signal generated by Raspberry Pi 3 card and a 433 MHz Wireless transmitter (ASK (Amplitude-Shift Keying) and FSK (Frequency-Shift Keying) modulation type), and captured under MATLAB-Simulink software by an RTL-SDR hardware using two detection method: the energy detection technique and the Artificial neural network (ANN). In ANN, we have tested different training algorithms that can be applied on the set of input data patterns to find out the best ANN architecture for signal detection. The performance evaluation of the used approaches is evaluated in terms of its ability to detect the transmitted signal by two parameters: probability of detection and false alarm probability.