A Highly Efficient Spectrum Sensing Approach Based on Antenna Arrays Beamforming

Abstract

In cognitive radio (CR), the sensitivity of the spectrum sensing (SS) algorithms depends mainly on several factors such as the receiver sensitivity, the antenna gain, the antenna efficiency and the SS algorithm itself. Researchers depend mainly on a single antenna element and sometimes on a uniform linear antenna (ULA) arrays to increase the system sensitivity. But actually the ULA suffers from the large side lobe level which in turn reduces the antenna gain. Furthermore, the separation between elements in the ULA is almost chosen to be half wavelength at the operating frequency which results in a dense array of close elements. If this separation increases slightly in order to reduce the mutual coupling between elements, the side lobes will increase and thus the realized gain is reduced more and more and as a consequence, the SNR will be reduced also which has a direct effect on the sensitivity of the system. In this paper, a low complexity and highly efficient SS technique based on antenna arrays beamforming and the generalized likelihood ratio test (GLRT) denoted as BF/GD is introduced. The proposed beamforming approach allows the utilization of the limited number of ULA elements and their corresponding RF chains to synthesize the desired radiation patterns of larger size antenna arrays for maximum gain realization. The desired pattern is realized by optimizing the excitation coefficients and inter-element spacing of the array. Consequently, the enhancement in the realized array gain will lead to a direct increase in the received signal-to-noise ratio (SNR) which significantly improves the sensitivity and the SS capability of the CR receiver without the need for extra hardware.

Keywords

• Beamforming (BF)
• cognitive radio (CR)
• spectrum sensing (SS)
• multiple antenna elements (MAE)
• uniform linear array (ULA)
• generalized likelihood ratio test (GLRT)