Experimental Research On Blind Detection And Recognition Technologies Of Communication Signals Based On Software Defined Radio

In modern communication systems,non-cooperative communication systems are widely used in military and civil fields.Blind detection and recognition technologies of communication signals play an important role in non-cooperative communication systems.Software defined radio is widely used in communication,radar and other fields because of its good flexibility,real-time performance and portability.This paper focuses on the key technologies involved in building a blind detection and recognition system for communication signals based on software defined radio in the non-cooperative communication environment,including software defined radio technology,blind detection,blind source separation and parameter analysis of communication signals.The technical scheme is verified by building a system test platform.Firstly,this paper introduces the basic content of software defined radio technology.The signal receiving platform based on software defined radio is the foundation of the subsequent signal processing.Secondly,aiming at the problem of communication signal detection,this paper studies three blind detection methods based on power spectrum analysis,cyclic spectrum analysis and time-frequency analysis.The detection performance of these three signal detection algorithms for single signal and multi-component signal in different signal-tonoise ratio environments is compared and analyzed through simulation experiments.Then,in order to solve the problem of determined and overdetermined blind source separation under the linear instantaneous mixed model,the FastICA algorithm is studied in this paper.According to the FastICA algorithm based on negentropy maximization,this paper presents an improved FastICA algorithm based on joint diagonalization of fourth-order cumulants.In this algorithm,the first separated signal is obtained by the joint diagonalization of fourth-order cumulant matrices of the observed signal,and then the second separated signal is achieved by FastICA algorithm.The simulation results show that the improved FastICA algorithm can reduce the number of iterations and accelerate the convergence speed of the algorithm on the premise of ensuring the separation accuracy.Next,the parameter analysis technology of communication signals is studied,including parameter estimation and modulation recognition.The estimation methods of symbol rate,carrier frequency and signal-to-noise ratio of communication signals are studied,and the effectiveness of these methods is verified by simulation experiments.By analyzing and optimizing the instant character parameters of the signal,a modulation recognition algorithm based on decision tree is used to realize the modulation recognition of 12 commonly used analog and digital modulation signals,including AM,DSB,VSB,LSB,USB,FM,2ASK,4ASK,2FSK,4FSK,2PSK and 4PSK.The simulation results show that the overall correct recognition rate of the algorithm is more than 90% when the signal-to-noise ratio is greater than 15 dB.Finally,the system test platform is built and tested.The received signal is collected by the software defined radio equipment.The blind detection,blind source separation and parameter analysis of the collected data are completed by the computer software.The test results are in accordance with the expected target,which verifies the feasibility of the system test platform.