Iterative Receiver Architecture for MIMO HSDPA Evolution

This paper investigates the combination of spatial multiplexing and overloading as an evolution of the UMTS-HSDPA (high speed downlink packet access) toward very high spectral efficiencies. The base station employs single-carrier space-time bit-interleaved coded modulations with linear preceding for each user. From an information-theoretic point of view, this scheme is potentially capacity achieving but puts high stakes on the receiver complexity. Channel estimation is made particularly difficult by the superimposition of pilot symbols and data. Indeed, the classical channel estimation method based on matched filtering and accumulation clearly fails to give satisfactory performance in a spatial multiplexing and overloading context. Thus, we propose a sophisticated iterative receiver which includes channel estimation, MMSE-based chip-equalization and MMSE-based multiuser detection together with decoding of the turbo-code. Simultaneous downlink transmission to several users is investigated in this paper which brings out the overloading flexibility in terms of space-time modulation and coding scheme (ST-MCS) design and radio resource allocation