Skip to Main Content
A novel on-line channel estimation and tracking method for wireless OFDM systems based on comb-type pilots is developed. Long training sequences are used to estimate the channel response in the frequency domain, and a scalar Kalman filtering scheme is designed to refine the channel estimation. To track the channel variations during a long burst packet, pilots estimated via vector Kalman filtering are used to interpolate linearly and modify the estimated channel response. Both channel AWGN and state variable noise variance are adaptively estimated. Furthermore, to overcome phase changes across symbols owing to the frequency offset estimation error, a pilot based phase compensation scheme is incorporated. The complete algorithm can be processed in real time with much better performance than training sequence based or pilot based channel estimation methods. Simulation results based on the IEEE 802.11a wireless LAN standard are presented.