Improving Channel Estimation for Software Radios

E. Coersmeir, C. Saez, A. Hotz, M. Hoffman, M. Kosakowski, Y. Xu, A. Bauer, and F. Leder (Germany)

Keywords

Wiener filter coefficient calculation, adaptive filter length, channel estimation, OFDM, software radio, Cordic

Abstract

Channel estimator is a fundamental component in a mobile OFDM software radio receiver. From the system perspective channel estimator needs to offer high precision within various frequency bands. From the mobile software radio perspective it should require low power consumption. Best performance can be achieved by Wiener filter approach, but algorithm complexity is high and thus power consumption is not optimal for software radio platform. The channel model is assumed to be time variant. It consists out of paths, each path is described by coefficient [1]. pN )(t n hn ∑= −= PN n nn thth 1 )()(),( ττδτ (1) The time delay for the n th path is given by nτ . Channels at different time instances and time delayst τ are shown in Figure 1. Thus the paper proposes software radio processor load reduction for Wiener filter in an OFDM system. Optimal run time Wiener filter coefficient calculation as well as adaptive Wiener filter length are investigated. Additionally a simplification for trigonometric functions in software radio has been employed. τ h(τ,t) τ h(τ,t)

Important Links:



Go Back