Research And Implementation Of Wideband Signal Real-Time Analysis Technology Based On Channelization

With the progress of social life and science technology,electronic communication technologies in human society is occupying an increasingly important position.Facing this increasingly complex electromagnetic environment,whether it is spectrum sensing and information reconnaissance in military applications,or spectrum monitoring in civil systems,etc,a powerful signal receiving,analysis and processing system is required.At present,the commonly used technical methods mainly adopt narrow-band frequency sweeping methods,but for TDMA signals or sudden short-term signals,they may miss detection.In this regard,in response to the above situation,this article focuses on the research and implementation of a channelization-based real-time reception and processing system for wideband signals,which can analyze and process non-cooperative signals in real time while satisfying large bandwidth and high dynamics.First of all,this article studies the channelization technology.According to the processing requirements of the whole system for broadband signals,this article analyzes the dynamic range and sensitivity of the receiver processing,and proposes a two-stage channelization scheme.Through the combination of front-end RF channelization and the back-end digital channelization,this method meets the needs of the system in the dynamic range of broadband signal reception and processing.At the same time,through the improvement of the digital channelization technology based on the polyphase filter bank in the subsequent stage,a full coverage channelization solution is proposed to solve the problem of channel blind zone and signal cross-channel problem.The dynamic range can reach 75 d B at 300 MHz sub-channel bandwidth.Then this paper studies the blind detection technology of carrier basic parameters.First,the Welch periodogram method of parallel fast Fourier transform based on serial number mapping is used to estimate the power spectrum of the received signal collected at high speed,and then the carrier detection is performed on the power spectrum to obtain basic information such as signal bandwidth,carrier frequency and signal-to-noise ratio to guide subsequent processing.Aiming at the problems of variable received signal patterns and uneven noise floor,a LAD detection algorithm with flattened noise floor is proposed to reduce the probability of signal misdetection and missed detection.The realization result shows that the dynamic range of the power spectrum estimation is about 75 d B,and the error of the actual signal carrier parameter detection result does not exceed 5%.Finally,this paper studies the real-time recognition technology of signal characteristic parameters oriented to FPGA,which mainly includes four types of characteristic parameters: modulation mode recognition,amplitude domain characteristics,power spectrum characteristics,high-order moment characteristics and fractal dimension.A hardware coprocessor architecture of the signal characteristic parameter real-time recognition unit is proposed.The implementation results show that the recognition accuracy of BPSK,8QAM,and OQPSK modulation methods can reach95% under 10 d B signal-to-noise ratio;under 10 d B signal-to-noise ratio,The signal classification accuracy rate of characteristic parameters can reach 95%.