Creating and sharing knowledge for telecommunications

Hybrid Matched Filter Detection Spectrum Sensing

Brito, A. ; Sebastião, P. ; Velez, F. J.

IEEE Access Vol. 9, Nº 0, pp. 165504 - 165516, December, 2021.

ISSN (print): 2169-3536
ISSN (online):

Scimago Journal Ranking: 0,59 (in 2020)

Digital Object Identifier: 10.1109/ACCESS.2021.3134796

Download Full text PDF ( 1 MB)

Downloaded 1 time

The radio frequency spectrum is getting more congested day by day due to the growth of wireless devices, applications, and the arrival of fifth generation (5G) mobile communications. This happens because the radio spectrum is a natural resource that has a restricted existence. Access to all devices can be granted, but in a more efficient way. To resolve the issue, cognitive radio technology has come out as a way, because it is possible to sense the radio spectrum in the neighboring. Spectrum sensing has been recognized as an important technology, in cognitive radio networks, to allow secondary users (SUs) to detect spectrum holes and opportunistically access primary licensed spectrum band without harmful interference. This paper considers the Energy Detection (ED) and Matched Filter Detection (MFD) spectrum sensing techniques as the baseline for a study where the so-called Hybrid Matched Filter Detection (Hybrid MFD) was proposed. Apart from an analytical approach, Monte Carlo simulations have been performed in MATLAB. These simulations aimed at understanding how the variation of parameters like the probability of false alarm, the signal-to-noise ratio (SNR) and the number of samples, can affect the probability of miss-detection. Simulation results show that i) higher probability of miss-detection is achieved for the ED spectrum sensing technique when compared to the MFD and Hybrid MFD techniques; ii) More importantly, the proposed Hybrid MFD technique outperforms MFD in terms of the ability to detect the presence of a primary user in licensed spectrum, for a probability of false alarm slightly lower than 0.5, low number of samples and low signal-to-noise ratio.