Enhanced spectrum sensing in optical-NOMA for 256-QAM: a hybrid energy and matched filter detection approach

Abstract Optical non-orthogonal multiple access (Optical-NOMA) is an advanced technique of optical wireless communications that increases spectral efficiency by enabling several users to share the same frequency and power resources through power-domain multiplexing. Spectrum sensing is important in cognitive radio networks, especially in fading channels. This paper analyzes the detection capability of energy detection (ED), matched filter (MF), compressive sensing (CS), and a new hybrid scheme involving ED and MF (ED + MF). The new approach enhances detection capability with an increased probability of detection (Pd) at lower signal-to-noise ratios (SNRs). In Rayleigh fading, the hybrid scheme achieves Pd = 1 at −1 dB, better than MF (0.5 dB) and ED (2 dB), with gains of 1 dB and 3 dB, respectively. Likewise, in Rician fading, Pd = 1 is achieved at −1 dB, better than MF (0.7 dB), ED (1 dB), and CS (2 dB), with gains of 2 dB and 3 dB over CS and ED. The analysis validates enhanced detection by decreasing false alarm probability (Pfa) without compromising precision. Further, bit error rate (BER) versus SNR analysis provides a gain of 1.3 dB–4.8 dB in Rayleigh fading and 1.2 dB–5.5 dB in Rician fading. The hybrid scheme improves optical-NOMA performance with reduced BER and increased spectral efficiency in fading channels.