Low-complexity multipath diversity through fractional sampling in OFDM

Orthogonal frequency division multiplexing (OFDM) enables low-complexity equalization and has been adopted in several wireless standards. However, OFDM cannot exploit multipath diversity without computationally complex coding and decoding. We show here that by sampling at a rate higher than the symbol rate, which is also known as fractional sampling (FS), one can improve the diversity that the wireless channel can provide in an OFDM system. We propose maximal ratio combining at each subcarrier for the FS-OFDM system, argue that the diversity gains acquired through this approach are related to the spectral shape of the pulse and its excess bandwidth, and derive analytical bit error and symbol error rate expressions for our scheme. We also explore extensions to differentially encoded systems that do not require channel status information at the receiver, multiple-input multiple-output (MIMO) systems that exploit space diversity, and low peak-to-average (PAR) options such as zero-padded (ZP) and cyclic-prefix only (CP-only) transmissions. We corroborate our approach with simulations.