Research On Polarization Spectrum Sensing Algorithms

Through the real-time spectrum sensing and using of the spectrum,cog-nitive radio(CR)is able to improve the utilization rate of spectrum in wire-less environments.In recent years,in addition to classical spectrum sensing algorithms based on time,frequency or space domain information of signal,spectrum sensing algorithms using polarization information have attracted re-searchers ' attention.The research opens up a new dimension for the spectrum sensing technology.Due to both of the arrival angle and non-ideal noise can influence the performance of polarization spectrum sensing,supported by Beijing Higher Education Young Elite Teacher Project and Fundamental Research Funds for the Central University,main contents of this dissertation are:(1)The dissertation overviews principles of spectrum sensing,and clas-sical sensing algorithms which based on time,frequency and space domain information.Polarization theories and technologies are included.Then exist-ing polarization spectrum sensing is been analyzed.Finally,the dissertation pointed that polarization sensing algorithms do not consider the presence of arrival angle and non-ideal noise.(2)We propose polarization information based spectrum sensing methods considering arrival angle,due to the presence of arrival angle,which can re-duce the received energy of primary signal and then degrade the performance of polarization detectors.Using the generalized likelihood ratio test(GLRT)paradigm,we derive a blind Polarization and Arrival angle Eigenvalue Detec-tion(PAED),which is the optimal polarization algorithm in the presence of arrival angle.Our results show that the proposed PAED exhibits 7dB better performance than other existing techniques,particularly when the number of samples is small,which is critical in vehicular applications.(3)We propose a new polarization based spectrum sensing method in CR,considering both of the arrival angle and non-ideal noise,which degrade the performance of polarization spectrum sensing algorithms.There are two as-pects of this part.The first aspect is research of polarization spectrum sens-ing in power imbalance noise.The blind GLRT based and Pre-whiten based algorithms are proposed.And the two algorithms achieves 2dB and 5dB per-formance than traditional methods perspectively.The second aspect is research of polarization spectrum sensing in Non-Gaussian noise.Based on the analy-sis of statistical information,we propose blind GLRT based sensing algorithm in presence of arrival angle in Non-Gaussian noise.Numerical simulation re-sults show this method obtains 4dB gain than those without considering Non-Gaussian noise.