Research On Communication Signal Feature Extraction And Recognition Technology Based On ZYNQ Platform

The feature extraction and recognition technology of communication signals has important research significance,which can provide parameter basis for receiving and processing signals.Identifying the modulation type of the signal is a prerequisite for realizing the correct reception and demodulation of the communication signal.In addition,signal modulation identification technology can also be used in the field of radio signal monitoring and electronic warfare.In radio signal monitoring,equipped with a software radio platform to identify signal modulation types,it can monitor and evaluate the performance and operating status of the radio communication system,which is of great significance for maintaining the security and stability of the communication system.In the field of electronic warfare,signal modulation identification technology can be used to monitor and interfere with the enemy’s radio communication system,which is of great significance to the grasp of battlefield information and the destruction of the enemy’s communication system.Therefore,the research and application of signal modulation recognition technology has important theoretical significance and practical application value.This thesis studies the communication signal feature extraction and recognition technology based on the ZYNQ platform.The focus of the work is as follows:(1)First,analyze the theoretical knowledge.The basic principles of communication signal feature extraction and recognition technology are analyzed,as well as various characteristic parameters commonly used in signal modulation pattern recognition,and the characteristic differences between different modulation signals are analyzed through formula derivation,which lays a theoretical foundation for subsequent algorithm design;at the same time,it introduces the software The basic frame structure of the radio platform introduces the main components of the software radio in detail from the perspective of the software and hardware system framework,and provides design ideas for the construction of the subsequent experimental test platform.(2)Next,the construction of the experimental platform is introduced in detail.The software radio system model based on the ZYNQ platform is studied,including the hardware design scheme of the platform,the data processing module design and the upper computer software design.FPGA in the ZYNQ platform implements logic resources to develop data processing high-speed algorithms,and ARM uses processor resources to develop logic control programs and configure the radio frequency transceiver unit.Through the start-up test,network data transmission test and joint debugging of software and hardware data communication on the built experimental platform,the real-time performance and portability of signal processing of the software radio system based on ZYNQ are verified under the premise of low power consumption and small size.(3)Optimize the design of the algorithm again and conduct a simulation test.According to the key research points of the project and the task requirements of project realization,15 types of modulated signals were selected as the research objectives,and the inter-class characteristics of different modulated signals and the differences in the intra-class characteristics of the same modulated signal were compared respectively,and simulation tests were carried out.Classification indicators set threshold thresholds.In view of the fact that some signals have small feature differences and are difficult to classify in the process of intra-class recognition,an FSK signal optimization algorithm based on spectrum peak statistics and a QAM signal mixed feature recognition algorithm based on constellation diagram and time-frequency characteristics are proposed.The simulation results show that the above algorithms the recognition performance has been greatly improved.Based on this,a joint recognition algorithm based on signal feature values is proposed to achieve reliable signal modulation identification and classification.(4)Finally,the performance of the algorithm is verified by the experimental platform.Test the effectiveness and functional integrity of the communication signal feature extraction and recognition system based on the ZYNQ platform.In order to prevent the influence of other factors in the actual measurement process,the fifteen types of modulation signals selected in this thesis are all generated by the signal generator,and the signals are received and processed by the radio frequency receiving end.When the signal-to-noise ratio of the collected signal is 12 d B,the recognition accuracy is above 89%.When testing the platform,repeated experiments were carried out on the platform from the perspectives of signal transmission power and signal-to-noise ratio.The test results verified the effectiveness of the signal joint recognition and classification algorithm in this thesis,and also provided a realistic basis for further application in engineering practice.