Wideband spectrum sensing technique based on subspace compression for MIMO cognitive radio networks

Occurrence of the frequency bands seeks dynamic spectrum allocation for improving the network capacity. Further the upcoming generation in this field is expected to function in the entire unlicensed, licensed and shared spectrum for meeting the widely applicative spectral demands. Accordingly, the wireless radio has to be tuned with their significant transmission parameters like carrier frequency, detection rate, and modulation scheme based on their quality of service. When the monitoring spectrum reach gigahertz, it needs a sensing latency to prevent the receiver form spectral exploitation. To avoid this and to optimize the utilization of wide-band signals among the primary and secondary user, the study focused to prior and reliable detection of primary user through Wide Band Signal Sensing (WBSS) phenomenon. Based on the information to be extracted the proposed model employed Compressed Subspace Learning (CSL) to explore the WBSS framework. For the purpose of signal recovery, the architecture used Bayesian recovery algorithm and for spectrum sensing a hybridized CSL–Greedy approach is formulated for the efficient determination of primary user. The proposed system attains high performance in terms of detection rate, spectrum efficiency and misses detection when compared to the state of art methods. This developed framework potentially enables similar frequency bands to be smartly re-used in any spectrum.