Font Size: a A A

The Research Of Spectrum Sensing Algorithms Based On Eigenvalue Structure

Posted on:2015-12-31Degree:MasterType:Thesis
Country:ChinaCandidate:Y MiFull Text:PDF
GTID:2298330431994663Subject:Communication and Information System
Abstract/Summary:PDF Full Text Request
Cognitive radio(CR) technology can intelligently and dynamically manage spectrum resources, changed the original mechanism of statically allocated spectrum, therefore CR can effectively relieve the tension situation of spectrum resources usage so as to significantly improve spectrum utilization, in recent years, CR is generally concerned by people. One of the most critical components of CR technology is Spectrum sensing(SS), that require the wireless communication devices detect and found the "spectrum holes", its purpose is the reasonable use of the free spectrum bands via analysis and judgment of the spectrum. By means of SS technology for real-time monitoring of the spectrum condition, once the primary user(PU) needs access to an interesting radio spectrum, the cognitive user(CU) should withdraw from the spectrum instantly to minimize the interference. Thus, SS is the key technology to the development of CR applications.Firstly, in this paper, the research backgrounds of CR and SS technology are briefly presented, the present research situations are outlined after that, then the system model of spectrum detection is introduced, the current spectrum detection algorithms are briefly presented in classification, mainly investigated the classical energy detection(ED) algorithm and its improved algorithm. By MATLAB simulation for verification, although ED algorithm is simple and does not need any prior knowledge of PU, but it is susceptible to noise uncertainty and the fading environment. An improved ED algorithm mainly considers the minimum error probability criterion for analysis, introduced the multi-antenna and multi-node, the simulation results show that an improved algorithm can achieve arbitrary small probability of error, that is, minimize interference to PU, then meet the given wireless communication standard.Secondly, in response to the existence of the "hidden terminal" problem in actual wireless communication environment, the multi-node cooperative SS algorithms are researched. Random matrix theory(RMT) and array signal theory are applied to the cooperative spectrum detection, by multi-node, to obtain the sample covariance matrix of the received signal, then through constructing the different test statistics, in the case of the given probability of false alarm deducing the real-time decision threshold, while an improved threshold decision algorithms are given. By MATLAB simulation, the performance of the algorithms are compared in the case of different parameters, an improved algorithms are also verified. The simulation results show that the algorithms based on eigenvalue structure are blind SS algorithms, and do not need any prior knowledge of PU in advance, but also good resistance to the effects of noise uncertainty, thus this kind of SS algorithms have a good detection performance.Finally, the work of this paper are summarized, and the future research contents are discussed. In short, CR technology can fundamentally improve spectrum utilization of the limited spectrum resources for secondary utilization, which provide a reliable resource protection and far-reaching actual significance for the development of the next generation of wireless communications technology. SS as a key and basic technology for CR, facing great challenge and many problems, has a more broad prospect and urgent research in the future.
Keywords/Search Tags:Cognitive radio, Spectrum sensing, Eigenvalue, Random matrix theory, Sample covariance matrix
PDF Full Text Request
Related items