Channel Estimation for IFDMA - Comparison of Semiblind Channel Estimation Approaches and Estimation with Interpolation Filtering

In this paper, Discrete Fourier Transform (DFT) precoded Orthogonal Frequency Division Multiple Access (OFDMA) with interleaved subcarrier allocation per user is considered which is denoted as Interleaved Frequency Division Multiple Access (IFDMA). In order to estimate the channel for IFDMA, in general, each subcarrier of an IFDMA symbol is used for pilot transmission to fulfill the sampling theorem in frequency domain. If the data of different users is additionally separated by a time division multiple access component, a limited number K of successive IFDMA symbols is transmitted and at least two of these K symbols are used for pilot transmission if the channel is estimated with the help of interpolation filters. In this paper, we propose to estimate the channel with an iterative decision directed channel estimation with Wiener filtering in order to reduce the pilot symbol overhead in time domain. We additionally introduce the combination of this approach with semiblind subspace based channel estimation in order to reduce the pilot symbol overhead in frequency domain. By this means the channel can be estimated even if the sampling theorem is not fulfilled. The new approach outperforms the iterative decision directed channel estimation with Least-Squares estimation on each subcarrier and the Wiener interpolation filter for Signal-to-Noise Ratios larger than 20 dB and for velocities up to 15 km/h.