Research On Nonstationary Interference Suppression In Dsss Communications

Direct sequence spread spectrum(DSSS) communication has been widely used in both civil and military areas because it can provide desired features such as energy density reduction, confidentiality, interference suppression, anti-multipath fading, and code-division multiple access, etc. In military communications, DSSS systems face serious threat of enemy jamming because of the rapid development of electronic warfare as well as its importance in information warfare. In civilian communications, DSSS systems and narrow-band communication systems coexist in the same band in order to make full use of spectrum resources. Although DSSS system has immunity to narrow-band interference, it still faces threat of interference because it is enslaved to processing gain, distance, transmission power and other factors. It is necessary to suppress interference in DSSS communications to ensure validity and reliability of communication. On the summarization of existing researches, this paper discusses on the suppression of linear frequency modulation(LFM) interference, frequency-hopping interference, broadband comb-spectrum interference in DSSS communications, and classification of interferences.The existing LFM interference suppression methods have several drawbacks: excessive computational cost, less research on time-varying LFM and multi-component LFM interference rejection. This paper presents single-component LFM interference rejection method based on two close fractional Fourier power spectra, multi-component LFM interference rejection method combining High-order Ambiguity Function with Fractional Fourier Transform. Algorithm validity is verified by means of simulation. Compared with current algorithms, the proposed algorithm avoids two-dimensional search process, so it has lower computational cost under the same performance condition. Furthermore, it can suppress time-varying amplitude LFM interference.In this paper, frequency hopping interference suppression is implemented. Because of inherent property of wavelet transform on processing frequency break signals, interference parameters are estimated through wavelet ridge extraction and interference rejection is implemented by adaptive notch algorithm in frequency domain according to interference characteristics. In order to estimate parameters accurately, wavelet denoising and Least-Square algorithm are employed. Simulation results show that the algorithm is effective. In the proposed algorithm, fast wavelet ridge extraction iterative algorithm is used to reduce computational cost.Broadband comb-spectrum interference suppression based on adaptive wavelet packet decomposition is also presented in this paper. Through analyzing broadband comb-spectrum interference features in time-frequency domain, adaptive wavelet packet decomposition is employed to target interference first, then interference suppression complete through weighted wavelet packet coefficient and reconstructing the signal without interference. Simulation results show that the algorithm can suppress broadband comb-spectrum interference effectively, and the performance is better than the DFT-based methods.Recognition of nonstationary interference researched in this paper is implemented. After extracting interference features, the classifier is structured using decision tree support vector machine. The classifier training algorithms and recognition algorithms are given, and optimal parameters are selected. Recognition rate statistics show that, the classifier can identify interference accurately when the identification system SNR is greater than 0dB.