Abstract—In this paper we revisit the minimum meansquare error (MMSE) pilot-aided channel estimation forbroadband orthogonal frequency division multiplexing(OFDM) systems. The careful design of OFDM symbol leadsto the proposal of a simplified time-domain (TD) MMSEestimator. By exploring the Fourier properties of thesymbol, the investigated method eliminates the need to usedirect or inverse discrete Fourier transforms (DFT/IDFT)before the channel estimation, by using the TD receivedsymbol samples as the input to the MMSE filter. Moreover,performing the estimation in TD (where the channel impulseresponse (CIR) energy is mainly limit to a small set ofsamples), makes way to a simple, nonetheless efficientestimation of the filter parameters. The performance of thechannel estimation scheme, using the estimated parameters,presents a tolerable degradation when compared with theideal situation (a-priori knowledge of the channel correlationand noise variance), while exhibiting a reducedcomputational load. The feasibility of the investigatedmethod is substantiated by system simulation using indoorand outdoor wireless channel models.
 
Index Terms—OFDM, pilot-aided channel estimation, timedomainprocessing.

Cite: Carlos Ribeiro and Atílio Gameiro, "An OFDM Symbol Design for Reduced Complexity MMSE Channel Estimation," Journal of Communications, vol. 3, no.4, September, pp. 26-33, 2008.