Research On Wideband Spectrum Sensing Technology Based On Compress Sensing And Hardware Implementation

With the rapid development of wireless communication technology,how to effectively manage and apply spectrum resources becomes a particularly important problem.In response to this problem,scholars put forward cognitive radio(CR)system.It is a very promising technology,and its important part is spectrum sensing.It can avoid interference between primary users and secondary users by identifying spectrum holes,thereby efficiently utilizing spectrum resources.The CR system will operate in a wide frequency range in the future,so the research of broadband spectrum sensing technology is very important.The Nyquist rate required by traditional signal processing brings great challenges to spectrum sensing receivers.The high research and development cost of hardware applications and the high complexity of signal processing also bring great difficulties to practical applications.In order to solve these problems,some scholars get inspiration from compressed sensing.They began to study broadband spectrum sensing technology based on sub-Nyquist sampling,and developed a variety of spectrum sensing methods.The birth of these methods not only reduces the hardware requirements for practical applications,but also liberates the pressure of data compression,transmission,and processing.In this paper,a spectrum sensing method based on sub-Nyquist sampling with low computational complexity and easy hardware implementation is proposed.Then an experimental verification system based on sub-Nyquist sampling is designed and implemented according to this method.Compared with some existing research results,the method proposed in this paper has lower complexity and higher accuracy in the case of low signal-to-noise ratio.And the resource utilization of the hardware system is more efficient.The main work and innovations of this paper are as follows:(1)Research and improvement of broadband spectrum sensing algorithm based on compressed covariance sensing.The existing methods based on compressed covariance sensing cannot guarantee the accuracy of signal reconstruction when there are few signal samples.However,the difficulty of hardware system processing is raised by the increasing number of sample points.Aiming at this problem,this paper proposes an improved method with low instantaneous computational complexity.In the implementation of the hardware system,lower computing resources can be used to process more sample points.(2)A wideband spectrum sensing system based on the improved compressed covariance sensing method is designed and implemented,including front-end RF reception,sub-Nyquist sampling and FPGA algorithms.In order to receive wireless signals,the design of automatic gain control is added to the design of the front-end RF receiver;Then,the sub-Nyquist sampling samples are obtained through the multi-coset sampling architecture;Finally,the signal samples are processed by the FPGA algorithm of compressed covariance sensing to realize spectrum sensing.(3)The debugging of the hardware system is completed,and the principle verification of the broadband spectrum sensing algorithm based on compressed covariance sensing is realized.In the functional testing phase of the hardware system,the system's reconstruction accuracy and dynamic range were tested under the condition of inputting different types of signals through two methods: the signal source directly inputs the signal and the antenna receives the wireless signal.