Cognitive Radio Spectrum Sensing Based on Energy Detection and Cyclostationary Detection

This document presents a MATLAB implementation for cognitive radio spectrum sensing utilizing both energy detection and cyclostationary detection algorithms. The implementation workflow consists of the following steps: First, execute the a.m program which computes the cyclic frequency alpha value through spectral correlation function analysis. This program typically involves calculating the cyclostationary properties of the received signal to identify characteristic cyclic frequencies. Then run the zhouqi.m program to validate the computed alpha value and perform signal detection using cyclostationary detection algorithms. This module implements cycle frequency domain profile analysis to detect primary user signals based on their cyclostationary features, which offers robustness against noise uncertainty. Next, employ the energy detection algorithm by executing the nengliang.m detection program. This module calculates the energy statistic of the received signal and compares it against a predetermined threshold to determine spectrum occupancy, providing a computationally efficient alternative detection method. Finally, conduct comprehensive detection performance analysis to evaluate accuracy and reliability metrics such as probability of detection, probability of false alarm, and receiver operating characteristics. The analysis compares both detection methods under various signal-to-noise ratio conditions and signal characteristics. This combined approach effectively enables radio spectrum sensing by leveraging the complementary strengths of both detection methodologies. The energy detection offers simplicity and computational efficiency, while cyclostationary detection provides improved performance in low SNR environments and better discrimination between noise and modulated signals. Through this MATLAB implementation, researchers can gain practical understanding of spectrum sensing techniques and their implementation considerations for real-world cognitive radio applications.