The Research On Transform Domain Narrowband Interference Suppression Techniques In Spread Spectrum Communication System

Spread spectrum technique, which is widely used in the third generation mobile communication systems, broadband wireless access systems and satellite communication systems, especially in military satellite communication systems, inherently offers some degree of protection against interference, and thus can be used to improve reliability. Although any level of protection can be achieved by using a spreading waveform with sufficient processing gain, it is often necessary to supplement the innate processing gain associate with the spreading code by using additional signal processing techniques.Firstly, this thesis briefly introduces the spread spectrum communication systems and the developing narrowband interference suppression techniques at present.Secondly, this thesis deeply researches narrowband interference suppression technique based on lapped transform. Lapped transform principle, structure and basic type are analyzed. On this basis, a receiver is developed, and lapped transform incorporating modified median filter algorithm is proposed, furthermore, the bit error rate (BER) formula of the receiver is obtained. Simulation results show that the system BER performance is improved greatly and the narrowband interference is suppressed greatly.Finally, aiming at the disadvantage of the lapped transform being insensitive to time-varying interference, this thesis proposes a new transform-domain narrowband interference suppression technique based on undecimated wavelet packet transform (UWPT). Unlike the tree structuring algorithm (TSA), UWPT does not perform decimation operation after the analysis filter. It combines with frequency shift of the signal to place the interference inside one of the passband filters, and adopts a scheme from bottom to top to confine the interference in one of the first level filters and remove it. Simulation results show that the algorithm can track the variations of the input signal and provides a very robust performance with independence of the interference frequency location. So the algorithm is particularly suitable to suppress the time-varying narrowband interference.