Abstract
Cooperative spectrum sensing (CSS) is an important topic due to its capacity to solve the issue of the hidden terminal. However, the sensing performance of CSS is still poor, especially in low signal-to-noise ratio (SNR) situations. In this paper, convolutional neural networks (CNN) are considered to extract the features of the observed signal and, as a consequence, improve the sensing performance. More specifically, a novel two-dimensional dataset of the received signal is established and three classical CNN (LeNet, AlexNet and VGG-16)-based CSS schemes are trained and analyzed on the proposed dataset. In addition, sensing performance comparisons are made between the proposed CNN-based CSS schemes and the AND, OR, majority voting-based CSS schemes. The simulation results state that the sensing accuracy of the proposed schemes is greatly improved and the network depth helps with this.
Highlights
The rapid development of wireless communication technology has led to more and more wireless network services
This paper considers the scenario where the received energy vector of each node is sent to the fusion center (FC) and a two-dimensional matrix is formed at the FC
Deep reinforcement learning (DRL)-based Cooperative spectrum sensing (CSS) algorithm is proposed, which is employed to decrease the signaling in the network of secondary user (SU)
Summary
The rapid development of wireless communication technology has led to more and more wireless network services. Centralized CSS denotes that each node aggregates the local sensing information to the fusion center, and the fusion center makes a decision according to the fusion rules. There is no fusion center for distributed CSS [17,18], and each local node exchanges the detection results with the others, and combines the local fusion decision. The sensing accuracy of the CR system is effectively improved for the distributed CSS [19,20,21] This is at the expense of network burden and system overhead [22,23,24]. This paper is devoted to the CNN-based CSS for the possible performance improvement.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have