Efficient Sequence Detection of Multicarrier Transmissions over Doubly Dispersive Channels

Abstract

We propose a high-spectral-efficiency multicarrier system for communication over the doubly dispersive (DD) channel which yields very low frame error rate (FER), with quadratic (in the frame length) receiver complexity. To accomplish this, we combine a non-(bi)orthogonal multicarrier modulation (MCM) scheme recently proposed by the authors with novel sequence detection (SD) and channel estimation (CE) algorithms. In particular, our MCM scheme allows us to accurately represent the DD channels otherwise complicated intercarrier interference (ICI) and intersymbol interference (ISI) response with a relatively small number of coefficients. The SD and CE algorithms then leverage this sparse ICI/ISI structure for low-complexity operation. Our SD algorithm combines a novel adaptive breadth-first search procedure with a new fast MMSE-GDFE preprocessor, while our CE algorithm uses a rank-reduced pilot-aided Wiener technique to estimate only the significant ICI/ISI coefficients.

A novel receiver architecture for single-carrier transmission over time-varying channels

“…It is noteworthy that although this result coincides with the canonical band approximation approach in [9]- [12], [15], the underlying consideration [c.f. (8)] is obviously different.…”

“…2 (note that a similar structure was observed in [15]). Applying the inverse of H f,NĤ H f,N + R v,N can then be implemented by applying the inverses of G N and G H N separately using, e.g., Gaussian elimination.…”

“…This implies that it is more practical to assume a banded FD channel matrix 1 . Many equalizers exploit this banded (rather than diagonal) structure to lower the complexity, e.g., the block linear zero-forcing (ZF) equalizer in [7], the block linear minimum mean square error (MMSE) equalizer in [8], [9], the iterative serial MMSE equalizer in [10]- [12], the maximum likelihood (ML) equalizer in [13]- [15], etc. It can be imagined that to enhance the equalization performance, especially at a moderate to high signal-to-noise ratio (SNR), the band approximation error must be reduced as much as possible.…”

“…Since we need not to enforce a diagonal FD channel matrix but a banded one, a reduced-order FIR filter with just a single tap could be adequate. Such a filter is referred to as a receiver window in [9]- [12], [15].…”

“…In both data models, the full FD channel matrix will be approximated by a banded matrix for the sake of complexity. Note that unlike [9]- [12], [15] where the banded matrix is obtained by extracting the significant diagonals from the original FD channel matrix, the band approximation in this paper is achieved in a different manner: we will device a banded matrix that is close to the full FD channel matrix only in terms of the Frobenius norm. For the ODM, where the considered full FD channel matrix is the same as the original one, these two approaches are the same.…”