Spectrum Sensing And Accurate Reconstruction Algorithms Based On Digital Channelization Technology

In the military field,in order to improve the utilization of spectrum,related technologies of spectrum sensing are increasingly being applied in military communications and military reconnaissance.On one hand,it is necessary to use the perceptual technology to judge whether there is a self-contained device in the monitoring frequency band,so that other selfcommunication signals avoid using the frequency band.On the other hand,it is necessary to monitor whether there is a communication state in the frequency band through the sensing technology.The enemy's equipment determines the communication frequency range occupied by the enemy signal through the sensing technology,which provides a key reference for the enemy information interception and the selection of the interference strategy.This research is based on how to carry out efficient and wide-band communication signal reception and spectrum sensing and obtain accurate reconstruction characteristics.The main research contents are as follows:First of all,this paper studies the structure of broadband spectrum sensing,including the processing of multi-rate signals.Because digital channelization technology has good spectrum segmentation characteristics,low implementation complexity and high processing efficiency,a multi-phase filter bank is constructed.The digital channelization receiving system perceptual structure,secondly studied the basic theory of spectrum sensing,and proposed a spectrum sensing algorithm based on digital channelization structure and cyclic characteristics.By setting the dual power detection threshold,the subband signal detection and screening are calculated.The unoccupied and occupied subchannels are combined with the cyclic spectrum feature detection method to further detect the uncertain suspicious channel.The simulation verifies the correctness of the channelized output and the feasibility of improving the spectrum perception by the structure.Secondly,the stochastic resonance technique is introduced into the spectrum sensing theory of communication signals.The output response of the bistable stochastic resonance system is analyzed.The energy of the disordered applied noise energy is transformed into the weak signal in the stochastic resonance system.Mechanism principle,in order to improve the detection probability of energy detection algorithm under low SNR condition,an energy detection algorithm under stochastic resonance system is proposed,and the noise intensity is optimized by analyzing the offset coefficient under the fading channel.The amplitude causes the added stochastic resonance noise to randomly resonate with the direct current in the signal.The algorithm starts from the detection probability and the number of detected sample points.The theoretical derivation and simulation analysis show that the algorithm is lower than the traditional energy detection algorithm under low SNR conditions.Effectively improve the detection probability.At last,in order to complete the perception of all output subchannel signals and obtain the time-frequency waterfall graph of the signal,this paper studies the full-frequency domain situation sensing technology based on accurate reconstruction,which uses the channelized output as the reconstruction system.The input signal and the reconstruction of the signal need to design a prototype filter that satisfies the two characteristics of power complementation and high stopband attenuation.The optimal prototype filter and lattice structure filter bank are designed based on the Parks-McClellan algorithm.And the improved analysis of the reconstructed characteristics of the pseudo-orthogonal image filter bank,the prototype filter bank with good attenuation characteristics and the narrowest transition bandwidth is analyzed and verified.Based on the reconstructed output signal,the whole is drawn.The time-frequency waterfall map in the frequency domain acquires key time-frequency information and some characteristic parameters.