Matched Filter Based Spectrum Sensing And Power Level Recognition Algorithm

The need for higher data rate is increasing as a result of the transition fromvoice-only communications to multimedia type applications. Given the limitations of thenatural frequency spectrum, it becomes obvious that the traditional static frequencyallocation schemes can not accommodate the requirement of an increasing number ofhigher data rate devices. It results in situations in which some radio bands areovercrowded, while other bands remain moderate or rarely occupied. Cognitive radio (CR)has been recognized as a potential technology to improve spectrum utilization and solvespectrum scarcity. In a CR network, a secondary user (SU) is allowed to access thespectrum licensed to a primary user (PU) if the spectrum is not in use or the interferenceto the PU from SU is under a certain level by spectrum sensing. Hence, spectrum sensingserves as a key technology in CR network.This thesis mainly studies matched filter based spectrum sensing and powerrecognition when PU has multiple transmit power levels. Compared to the traditionalbinary sensing where only the presence of PU is checked, the targets of this thesis notonly detect the on-off state of PU, but also recognize the power level of PU. First, thebackground, key technologies, development status and the challenges of CR aresummarized in this thesis, and the existing spectrum detection key technologies isanalyzed. Secondly, spectrum sensing algorithms under a more reasonable scenario wherePU transmits different power levels are proposed that are grouping spectrum sensing anddirect spectrum sensing, and both the channel phase known and unknown is analyzed ineach strategy. The states of PU are divided into on and off two groups in groupingspectrum sensing algorithm, the first step of the algorithm is deciding the on-off state ofPU, the second step is recognizing the power level; in strategy direct spectrum sensingalgorithm, the power level of PU and the state of PU are recognized at the same time,rather than two steps.Finally, a cooperative spectrum sensing and power level recognition algorithm inmultiple primary power transmit power (MPTP) scenario is studing in this thesis, and themajority decision fusion and the optimal decision fusion are discussed separately.According to the simulation, the performance of the two cooperative spectrum sensingways and the performance of a single SU spectrum sensing and cooperative spectrum sensing are compared.