Feature Extraction And Recognition Method Of Large Time-width Radar Signal

Large width of the low probability of intercept radar signals which have low power, wide bandwidth and good noise immunity and concealment are hard to be detected and identified, so they have been widely used in radar, communications and other fields. Commonly used in large pulse width are: linear frequency modulation, phase encoding and phase coherent pulse signals.Currently, the study of common radar emitter signal recognition is popular, while the research of classification and identification of LPI radar signals are not many. LPI radar has low power and wide bandwidth, so it can not easily be intercepted and identified by enemy radars, in addition, it has anti-electronic reconnaissance, anti-radiation missiles and anti-jamming techniques and has been widely used in radars, communications and other fields. It gradually becomes an important technology system and working mode in modern radar equipments. Therefore, to study the problem of anti-co-channel interference about LPI radar with large time-width and high duty cycle is of great practical significance and military reference in improving combat capability, maintaining information superiority and increasing battlefield, etc.Therefore, this article is large for some typical low probability of intercept radar signals wide, research and more of the same type of radar signal interference between the same frequency feature extraction and recognition problems, and simulation experiments, including the following:(1) Introduced the principle of pulse compression, and the advantages and disadvantages of using radar pulse compression system. From the pulse compression principle and probability of intercept factor to start, research and simulation of the large width of three typical low probability of intercept radar signals: LFM, and coherent phase encoding pulse, respectively, when given their frequency characteristics, fuzzy function and low probability of intercept characteristics, obtained a low probability of intercept radar signals with low peak power and wide band characteristics of these two, indicating a low probability of intercept radar for current and future ships and weapons and equipment in the superiority.(2) Application of WVD time-frequency distribution, respectively, fractional Fourier transform and fractal theory method to study the extraction characterization LFM, phase encoding and phase coherent pulse width of these types of large, low probability of intercept radar signal characteristics of different parameters and signal characteristics, application of these characteristics obtained detailed analysis and identification of a radar signal recognition can be effective and feasibility of the database for the study of modern radar signal sorting complex identification and the use of radar anti-jamming measures provide a reliable guarantee.(3) Under the radar system for low interception characteristics of complex signals, in order to improve the individual radar emitter signal recognition rate, a new classification method. This method can reflect the wavelet packet transform to extract the signal modulation characteristics of the signal pulse has no intention of the band energy, through the generalization and learning ability are strong mixtures of kernels support vector machines for classification, simulation results show: the method is effective and possible, performance is better than existing methods.(4) Low probability of intercept for the three common disturbance radar signal recognition, combined with fractal and fractional Fourier transform algorithm, mixed-signal disturbance extracted information dimension, box dimension and the fractional ratio of the energy domain signal components, structure feature vector, and then support vector machine, the gray relational clustering, and the accumulation of K-means clustering method for disturbance classification. Simulation results show that: This method can effectively classify the type of interference, determine the type of disturbance of the same co-channel interference, co-channel interference or unusual type of different frequency band interference, EMC design and radar system design is important .