Research On Digital Preprocessing Technology Of Wideband Receiver

With the rapid development of communication technology,the replacement speed of communication equipment is also more frequent,so the situation of traditional hardware-based communication equipment has become more difficult.In this context,Software Defined Radio(SDR)technology is incubated.However,if you want to take advantage of the SDR software configuration at the receiver,you need the receiver front-end to achieve full-probability reception in the entire frequency band.However,due to the inherent nonlinear characteristics of analog devices,the resulting nonlinear distortion will seriously affect the receiver's Spurious-Free Dynamic Range and reduce the ability to detect weak signals in the frequency band.Therefore,improving the Dynamic Range of the receiver is a key point to play the advantage of SDR.This paper has done a full research on the nonlinear compensation technology of the digital receiver front end.The main contributions can be summarized in the following four aspects:1.It summarizes the linearization techniques commonly used in related fields,and analyzes the characteristics and applicable conditions of various linearization techniques in detail.Due to the lack of reference signal at the receiving end,the blind compensation technology based on nonlinear behavior modeling is emphasized,and two blind identification criteria are introduced: out-of-band distortion components and sparsity in the transform domain.2.The single-band and dual-band nonlinear distortion mechanisms of the receiver front-end are analyzed in detail.Based on this,the commonly used nonlinear behavior models and the measurement indicators of the nonlinear compensation effect are summarized.For dual-band concurrent receivers,a brand-new evaluation index,Concurrent Dual-Band Signal-to-Interference Ratio(CDB-SIR),is given to measure the nonlinear compensation effect in concurrent dual-band receivers.3.Based on the blind identification criterion of out-of-band distortion components,the nonlinear compensation of the front end of the single-band receiver is studied in detail.Due to the limitations of the adaptive algorithms LMS and NLMS,this paper introduces a more robust adaptive algorithm-Generalized Normalized Gradient Descent(GNGD)algorithm to update the parameters of the compensation model.4.Aiming at the nonlinear distortion of the front end of the concurrent dualfrequency receiver,this paper proposes a blind compensation architecture for the concurrent dual-frequency receiver based on the sparsity of the transform domain.Several measures of sparsity are given,and their scope of application is analyzed in detail.Under a specific nonlinear distortion model,the feasibility of nonlinear compensation theory based on sparsity is briefly proved.Finally,the effectiveness and superiority of the proposed algorithm are proved by software simulation.