Estimating the Efficiency of the Algorithm for Selecting the Adaptive Signal Detection Threshold in Cognitive Radio Systems

Abstract

Due to the active development of cognitive radio systems (CRS), the task of monitoring the spectrum is again relevant for study. The basic concept of CRS is to make efficient use of the radio frequency spectrum (RF), which includes spectrum sensing in order to detect and then use the primary user's (PU) RF without compromising the quality of service for all subscribers. Previously, we proposed an algorithm for implementing CRS based on spectrum occupancy models, where spectrum sensing is used as a source of up-to-date information about the state of the RF. Thus, it should be noted that spectrum sensing is one of the important and difficult problems, the solution of which affects the implementation of CRS. There are many scientific papers in the literature describing following spectrum sensing methods: cyclostationary detection, matched filter detection, etc. In the most papers, the energy detection (ED) is the most efficient way to solve the problem of spectrum sensing in CRS. The choice of ED is valid not only by the simplicity of implementation, but also by the absence of a requirement for a priori information about PU. However, the performance of an ED is highly dependent on the signal-to-noise ratio (SNR). The performance of the ED depends on the selected detection threshold, which in turn depends on the power of the noise in the band. Thus, the efficiency of the ED decreases in a low SNR area values, which negatively affects the performance of the secondary user terminal (SU) in real-life conditions. In this article, we have proposed an algorithm for choosing an adaptive signal detection threshold to improve the performance of ED under low SNR conditions. The proposed algorithm is based on a double-threshold detection method, using a detection threshold calculated at a fixed false alarm probability, as well as information about the previous decision about the presence or absence of a PU. Compared to a fixed threshold ED, the proposed adaptive detection threshold selection algorithm for spectrum sensing minimizes the missed detection probability in a low SNR area.