New robust ICI estimation using distributive PM-sequences in OFDM systems

The intercarrier interference (ICI) matrix for the orthogonal frequency division multiplexing (OFDM) systems usually has a fairly large dimension and changes over time for mobile communications. Obviously, the traditional least-square solution for the ICI matrix estimation based on the pseudo-inverse operation has its limitation. In addition, the provision of a sufficiently long training sequence to estimate the complete ICI matrix is not feasible, since it will result in severe throughput reduction. In this paper, we design padded maximum-length shift-register sequences (pm-sequences) to serve as the training sequences for the ICI estimation. Such new training sequences will induce the diagonal correlation matrices which can lead to the optimal estimation. We propose a new distributive allocation scheme of short pm-sequences for the ICI estimation with a linear ICI estimate interpolation, which can lead to the multi-ICI-vector estimation to combat the realistic non-circulant ICI matrix problem, where almost all other existing schemes such as the ICI matrix estimation using the Hadamard sequences, the convolutional coder/decoder and the ICI self-cancellation coder/decoder could not perform effectively