Pattern-Search-Based Nonconvex Cooperative Sensing in Multiband Cognitive Radio Systems

Abstract

In this paper, a pattern search (PS)-based solution is proposed for nonconvex multiband cooperative sensing (NCMCS) problem in cognitive radio systems. This problem consists of maximizing cumulative throughput subject to constraints on cumulative interference, probability of detection, and probability of false alarm. Initially in existing literature, this problem was solved under constraints that make it convex. However, removing the conditions for convexity and solving the NCMCS problem have been shown to improve performance. A two-step PS-based solution is presented: The first step uses uniformly distributed random sets of input points to find a solution. The set of points that gives the maximum throughput is chosen as input to the PS algorithm. Numerical examples show the improvement of the proposed method over existing genetic-algorithm-based solution, as well as PS-algorithm-based solution that uses a single set of random points as inputs. The proposed two-step solution gives higher cumulative throughput and is not sensitive to the choice of input, unlike the PS-based solution using a single set of random points as input.

Spectrum sensing in cognitive femtocell network based on near‐field source localisation using genetic algorithm

“…PS algorithm comprises of two main steps that is exploratory and updating. In exploratory step , a mesh, which is a group of points, is selected around the existing points [22]. These existing points can be either initial points provided at the start of the algorithm or points obtained from a previous step.…”

SMCSS: A Quick and Reliable Cooperative Spectrum Sensing Scheme for Cognitive Industrial Wireless Networks

“…By controlling the reporting order of local statistics, the global statistic at the fusion center accumulates faster than the case without order control. Apart from the fundamental aspects of CSS, there are also many other branches of the CSS research, such as multi-band sensing [25]- [27], energy-aware sensing [28]- [32], malicious user detection [33]- [35] and so on. More related works can be found in the recent CSS surveys [36], [37].…”