A simple two-level channel estimation scheme for OFDM system

Abstract

Due to its robustness to multipath delay spread and impulse noise, and also elevated spectral efficiency like characteristics, OFDM technology has been commonly adopted in contemporary wireless communication systems like LTE. The heart of an OFDM system is Channel estimator and the efficiency of channel estimation (CE) has a direct impact on the bit error rate performance of the system also the CE performance itself depends on effective estimation of fading channel hence the channel estimation problem has to be addressed. The popular LS based CE method is accepted for its simplicity but is found to be vulnerable to noise whereas the simple MMSE based CE that provides fine-tuned estimation still suffers from complexity as it depends on the knowledge of channel statistics (KCS). Hence a two level channel estimation technique that is a combination of kalman filter and thresholding scheme which denoises the LS initial estimates is proposed. The kalman filter tracks the estimates with initial data as LS estimates and threshold helps in selecting the significant data thus denoising the estimates and also filters the redundancy in estimated data. On comparing the existed CE techniques with that of the proposed scheme, the MSE performance of proposed scheme is found to be optimal and simple to that of the LMMSE technique.