Deep Sensing: Cooperative Spectrum Sensing Based on Convolutional Neural Networks

TL;DR

This paper introduces Deep Sensing, a novel cooperative spectrum sensing framework using convolutional neural networks to improve detection accuracy in cognitive radio networks by learning from sensing data.

Contribution

It presents the first CNN-based cooperative spectrum sensing framework that adaptively learns to combine sensing results, enhancing performance with low complexity.

Findings

Improved spectrum sensing accuracy with CNN-based framework

Effective in environments with spectral and spatial correlations

Performs well with moderate training samples

Abstract

In this paper, we investigate cooperative spectrum sensing (CSS) in a cognitive radio network (CRN) where multiple secondary users (SUs) cooperate in order to detect a primary user (PU) which possibly occupies multiple bands simultaneously. Deep cooperative sensing (DCS), which constitutes the first CSS framework based on a convolutional neural network (CNN), is proposed. In DCS, instead of the explicit mathematical modeling of CSS which is hard to compute and also hard to use in practice, the strategy for combining the individual sensing results of the SUs is learned with a CNN using training sensing samples. Accordingly, an environment-specific CSS which considers both spectral and spatial correlation of individual sensing outcomes, is found in an adaptive manner regardless of whether the individual sensing results are quantized or not. Through simulation, we show that the performance of CSS can be improved by the proposed DCS with low complexity even when the number of training samples is moderate.