Research On Non-intrusive Identification And Analysis Technology Of Communication Emitter

At present,with the rapid development and wide application of wireless communication technology,the communication security of wireless communication system is also facing unprecedented challenges,especially the monitoring and management technology for physical layer signal security is currently relatively weak.The identification and analysis of physical layer signals under non-intrusive conditions for various communication emitters are becoming more and more important.Based on this problem,this paper has done the following work in the single-carrier modulation systems and the LTE-A(Long Term EvolutionAdvanced)system which is one of the multi-carrier multi-antenna modulation systems respectively:(1)For single-carrier modulation systems,in order to realize highprecision identification of multiple communication emitters,a method of communication emitter identification based on steady-state cyclic spectrum characteristics is proposed.By using the strong robustness of cyclic spectrum’s cross-sectional spectrum in frequency domain to Gaussian noise,the intrinsic differences between shaping filters of different emitters are extracted for identification.Specifically,the cyclic spectrum’s cross-sectional spectra in frequency domain are extracted from the received steady-state signals,and the dimensions are reduced by principal component analysis.Then the emitters’ categories are determined by Pearson correlation coefficient method,probabilistic neural network and Frechet distance method,etc.The simulation results show that the proposed feature is superior to the traditional slice feature in cyclic frequency domain by using probabilistic neural network and Pearson correlation coefficient,which proves that it has certain application value.(2)Aiming at the problem of identification and analysis of communication emitters of multi-carrier modulation,the LTE-A system is mainly studied,a complete and feasible scheme of blind decoding of physical layer signals is proposed,and an actual measurement platform is built to verify the feasibility of the scheme.Then,based on the above scheme,combined with VGG16,ResNet18,ResNet50,MobileNetV3large and MobileNetV3-small five types of convolutional neural networks,the LTE-A downlink base station signal identification and uplink VoLTE(Voice over Long-Term Evolution)users’ quantity identification and analysis research are carried out.For downlink base station signals,a multi-antenna channel estimation plane is proposed as a base station feature map to individually identify multiple base stations in the airspace,so as to monitor the operation status of base stations and check the interference of pseudo base stations.Both the base station individual identification experiment and the pseudo base station interference investigation experiment prove that this feature is effective.For the uplink VoLTE user data,a complete set of data acquisition methods in FDD(Frequency-Division Duplex)or TDD(Time-Division Duplex)mode is proposed.In the method,the number of uplink VoLTE users and whether it is a data traffic service are classified and judged by cyclically collecting time slices,time slot slicing,removing cyclic prefixes,time-frequency transformation,removing noise,removing interference data blocks and buffering identification.The power ladder decision method is used for the initial identification,and then the time-frequency feature maps are brought into the five types of networks for identification and decision.Considering the large amount of data and the limited performance of the hardware platform,the feature data is subjected to multi-granularity processing experiments according to the wireless frame structure,and the experimental results show that the scheme is practical and effective.Finally,in order to improve the recognition accuracy of low-precision data judgment,a multi-granularity data ladder fusion training is proposed,which has been preliminarily verified through experiments.Finally,the main work and shortcomings of this paper are summarized.In the future,the related research methods based on the LTE-A system will be verified and developed in 5GNR(5th Generation New Radio)system.