Research On Robust Channel Estimation For SC-FDE UWB Systems

UWB affected by interference and fading in the signal transmission will produce distortion, it is usually possible to compensate for the distortion at the receiver, prerequisite is to know the channel information system, so for UWB system channel estimation is one of the key technical factors to ensure the performance of communication. In this paper, research is focused on robust channel estimation algorithm under the SC-FDE UWB. SC-FDE UWB system share bandwidth with traditional narrowband and often subject to strong narrowband interference system, so the channel estimation of SC-FDE UWB system must consider the influence of narrowband interference on the estimation result.The robust channel estimation is not sensitive to narrowband interference, whether or not the system existing narrowband interference it can get a better channel estimation result. When the system is subjected to discrete narrowband interference, its interference can be regarded as a singular point, in order to remove the discrete narrowband interference, a new method which based on wavelet modulus maxima dynamic threshold is proposed in this paper. According to the relationship of wavelet transform and singular point, we first get the singular point location by wavelet modulus maxima algorithm, and then, restore the original data through dynamic threshold and the corresponding interpolation algorithm. When the system is subjected to continuous narrow-band interference, according to the principle of transform domain narrowband interference suppression, in this paper presents fast dynamic double threshold robust channel estimation algorithm, first the frequency domain signal is segmented, after using piecewise spectrum to dynamically estimate the high and low threshold and determine the location of the interference, then restore the original data based on the least squares fitting and compression ratio algorithm.The simulation and experiment results show that the dynamic threshold wavelet modulus maxima robust channel estimation algorithm on the detection probability of discrete narrowband interference in the system are much better than the traditional algorithm and lower probability of misjudgment. In estimation performance, the mean square error of traditional channel estimation than the LS channel estimation is lower, but not obvious, the main reason is the threshold selection is not accurate. The wavelet modulus maxima dynamic threshold channel estimation adjusts the threshold according to the specific data for each of the receiver in real time, so it has a good adaptability to quickly change channels, and improve the mean square error and the accuracy dramatically. From the simulation of fast dynamic double threshold robust channel estimation algorithm can be seen that: Segmentation algorithm reduce the number of iterations dramatically, when data is bigger, the operation time and the complexity are greatly reduce. Traditional algorithm is not flexible to determine the high threshold and low threshold, the dynamic algorithm can flexibly get the corresponding threshold according to the different receiving data, it’s more close to reality and the accuracy is higher. Although the algorithm of setting zero on data recovery is simpler, but its accuracy is not high, in this paper, using the algorithm of fitting compression ratio achieved a good result.