Use Hierarchical Codebook to Improve the Primary User Detection in the Cognitive Radio’s Cooperative Spectrum Sensing

Hoai Trung Tran

doi:10.24018/ejece.2021.5.6.372

Abstract

Currently, the cognitive network is receiving much attention due to the advantages it brings to users. An important method in cognitive radio networks is spectrum sensing, as it allows secondary users (SUs) to detect the existence of a primary user (PU). Information of probability of false detection or warning about the PU is sent to a fusion center (FC) by the SUs, from which the FC will decide whether or not to allow the SUs to use the PU spectrum to obtain information. The transmission of information with a high signal to noise ratio (SNR) will increase the FC's ability to detect the existence of the PU. However, researchers are currently focusing on probabilistic formulas assuming that the channel is known ideally or there is nominal channel information at the FC; moreover, one model where the FC only knows the channel correlation matrix. Furthermore, studies are still assuming this is a simple multiple input – multiple output (MIMO) channel model but do not pay much attention to the signal processing at the transmitting and receiving antennas between the SUs and the FCs. A new method introduced in this paper when combining beamforming and hierarchical codebook makes the ability to detect the existence of the PU at the FC significantly increased compared to traditional methods.

Highlights

The cognitive radio network is proposed to solve the spectrum scarcity problem due to spectrum occupation variation [1], [2], and the bandwidths below 6 GHz are heavily used [1]
This paper uses the beamforming technique combined with the hierarchical codebook, first to find the pairs of Angle of Departure (AoD) and Angle of Arrival (AoA) corresponding to the physical paths, using the test parameters to determine it is better than using the SVD of a channel correlation matrix (TR – Robust Test) or nominal CSI information (TUA – Uncertainty Agnostic Test) [15] at the secondary users (SUs) when reporting to the Fusion center (FC)
When we use signal to noise ratio (SNR) with corresponding values of -2 dB, 2 dB, 5 dB, and 6 dB, we find that in Fig. 5 that is applying hierarchical codebook (HC) 2 beams is better than Robust Test (TR), concerning the covariance matrix, the channel estimates known at the FC and Uncertainty Agnostic Test (TUA), relating to only estimates of channel matrix available at the FC [15]

Summary

INTRODUCTION

The cognitive radio network is proposed to solve the spectrum scarcity problem due to spectrum occupation variation [1], [2], and the bandwidths below 6 GHz are heavily used [1]. The popular technique used for cognition commonly used today is spectrum sensing (SS), which finds the white spaces in the spectrum In this technique, the SUs needs to detect the signal of the PU [1], [2]. If no PU is present, the FC will make maximum use of the antennas for transmission It is not interested in the beamforming technique since this technique helps increase the signal-to-noise ratio, or in other words, it increases the decision capability at the FC. This paper uses the beamforming technique combined with the hierarchical codebook, first to find the pairs of AoD and AoA corresponding to the physical paths, using the test parameters to determine it is better than using the SVD of a channel correlation matrix (TR – Robust Test) or nominal CSI information (TUA – Uncertainty Agnostic Test) [15] at the SUs when reporting to the FC. If we combine the L transmit vectors, we have the signal at the jth antenna of the FC:

CHANNEL MODEL

SIMULATION

CONCLUSION