Volume-based method for spectrum sensing

Performance Analysis of Volume-Based Spectrum Sensing for Cognitive Radio

“…Note that the volume-based test statistic in (12) has been derived in [29] for accurate and robust spectrum sensing. Nevertheless, its theoretical performance analysis has not yet been investigated.…”

“…Indeed, besides the non-IID noise, the radio frequency (RF) IQ imbalance [21], [22] can also lead to performance degradation for the spectrum sensing methods, which, however, is beyond the scope of this paper. Some approaches have been suggested for robust spectrum sensing in the literature, such as the GLRT test [23], independence test [24], Hadamard ratio test [25], [26], Gerschgorin disk test [27], locally most powerful invariant test (LMPIT) [28] and volume-based test approaches [29]. Being robust against the non-IID noise and derived in the GLRT paradigm, the Hadamard ratio approach [30]- [32] has received much attention in the community of spectrum sensing, such as [25], [26].…”

“…On the other hand, the performance of the Hadamard ratio algorithm for spectrum sensing has been analyzed in [23], [24]. However, as pointed out in [29], although the Hadamard ratio test is robust against the non-IID noise, its detection performance needs to be further enhanced. Unlike the Hadamard ratio approach, the volume-based detector [29] is able to employ the correlation among the primary signals to further increase the denominator of the test statistic, leading to significant reduction in the test variable under the signal-presence hypothesis.…”

“…However, as pointed out in [29], although the Hadamard ratio test is robust against the non-IID noise, its detection performance needs to be further enhanced. Unlike the Hadamard ratio approach, the volume-based detector [29] is able to employ the correlation among the primary signals to further increase the denominator of the test statistic, leading to significant reduction in the test variable under the signal-presence hypothesis. This eventually leads to considerable improvement in detection performance for the volume-based approach.…”

“…It is worth pointing out that the volume-based detector is originally developed in [29] for real-valued observation and its detection performance has not yet been analyzed, which is the major contribution of this work. In particular, by computing the first and second moments for the signal-absence and signalpresence hypotheses together with using the Gamma distribution approximation, we derive accurate analytic formulae for the false-alarm and detection probabilities for the scenario of IID noise.…”