Clustered OFDM with channel estimation for high rate wireless data

Clustered OFDM can provide in-band diversity gain for wideband wireless channels. It is a promising technique for high-rate wireless packet data access, such as mobile multimedia communications. Due to the smaller size of each cluster for clustered OFDM than for classical (non-clustered) OFDM, edge effects can be very large. In this paper, we present new transforms for channel estimators in clustered OFDM systems. The new transforms are independent of the channel delay profiles and can effectively mitigate the edge effects. Computer simulation shows that the performance of clustered OFDM with the estimator using the new transforms is very close to the performance with the optimum estimator that depends on the channel delay profile. For the typical-urban or hilly-terrain delay profiles, clustered OFDM using the new transform based estimator is almost as good as classical OFDM with transmitter diversity.