Statistical spectral analysis: a nonprobabilistic theory
Statistical spectral analysis: a nonprobabilistic theory
Hi-index | 0.00 |
Cognitive radios opportunistically share the spectrum with the primary users. In this paper, we address the problem of detecting the presence of a primary user while a secondary user is transmitting data, in order to avoid harmful interference to the primary user. We have proposed an inband spectrum sensing technique utilizing the OFDM guard interval for scanning the spectrum. The spectrum sensing data from the discontiguous OFDM guard intervals is combined to obtain a long coherent data segment. We have shown that cyclostationarity can be exploited for the user detection even when the sensing data is discontiguous. The 'Listen-while-Talking' technique reliably detects the ATSC DTV signal in 8.5 ms when SNR is -21 dB. The inband spectrum sensing technique provides protection to primary users and saves the system capacity compared to the conventional system which uses idle period.