Narrowband Interference Compressed Detection And Mitigation Based On Priori Information

Impulse radio ultra-wideband (IR-UWB) is a new short-distance high-speed wireless communication technology possessing advantages such as low power consumption, high-bandwidth and multipath immunity. IR-UWB systems share spectrum resources with existing narrow-band wireless systems in a manner of overlapping. The radiation power of an IR-UWB system is much lower than that of its narrow-band rival systems. Thus, UWB systems have to face tremendous difficulties imposed by narrowband interference. How to mitigate the narrowband interference is an important issue need to be addressed in UWB system design.Compressed Sensing (CS) theory is a new signal sampling and reconstruction methodology proposed by Donoho, Candes et al recently. The CS theory states that:If a signal exhibits sparsity over a certain domain or basis, then it can be reconstructed with high probability from a few random observations. The CS theory makes it possible to sample and reconstruct IR-UWB signals with lowered rate. In this thesis, CS is applied to IR-UWB systems. The narrowband interference from wireless local area network (WLAN) signals is mainly considered. Based on prior information of WLAN signals, we investigate narrowband interference detection and mitigation schemes. The main contributions of work include the following two aspects:1. An extraction method of narrowband interference subspace is proposed based on the prior information of narrowband interference, and narrowband interference nullspace and observation matrix are constructed. In the coexistence of UWB signal and WLAN signal, the existing nullspace technology can not accurately extract narrowband interference subspace, and thus results in false interference information and/or missing useful information. Consequently, the performance of narrowband interference mitigation is poor. To solve this problem, we propose an improved method to extract narrowband interference subspace. In the proposed method, the spectrum structure of WLAN signals and the effect of spectrum leakage are taken as prior knowledge and used to reject false interference information and amend useful information. Based on proposed method, the narrowband interference nullspace and the observation matrix are constructed. Simulation results show that, the proposed extraction method of narrowband interference subspace can accurately locate interference and the constructed observation matrix can be used to mitigate narrowband interference effectively.2. A CS-UWB receiver model with the capability of interference mitigation is designed by using the proposed narrowband interference detection and mitigation method. In this model, pilots are used to assist the implementation of interference subspace extraction periodically. After that, the interference nullspace is constructed. The measurement matrix is updated dynamically according to the interference nullspace and the UWB signal subspace based on eigen-based dictionary. In this way, the narrowband interference can be mitigated during the measurement of incoming signals. Simulations have been done to evaluate the BER performance of the receiver in typical propagation environments. Simulation results show that, the designed CS-UWB receiver can mitigate narrowband interference effectively and gain satisfactory bit error rate (BER) performance.