Scalable Mismatch Compensation for Time-Interleaved A/D Converters in OFDM Reception

Realization of all-digital baseband receiver processing for multi-Gigabit communication requires analog-to-digital converters (ADCs) of sufficient rate and output resolution. A promising architecture for this purpose is the time-interleaved ADC (TI-ADC), in which several "sub-ADCs" are employed in parallel. However, the timing mismatch between the sub-ADCs, if left uncompensated, leads to error floors in receiver performance. Standard linear digital mismatch compensation (e.g., based on the zero-forcing criterion) requires a number of taps that increases with the desired resolution. In this paper, we show that oversampling provides a scalable (in the number of sub-ADCs and in the desired resolution) approach to mismatch compensation, allowing elimination of mismatch-induced error floors at reasonable complexity. While the structure of the interference due to mismatch is different from that due to a dispersive channel, there is a strong analogy between the role of oversampling for mismatch compensation and for channel equalization. We illustrate the efficacy of the proposed mismatch compensation techniques for an OFDM receiver.