Dissertation Defense
Design of Detectors and Decoders for MIMO Wireless Systems
Add to Google Calendar
Abstract:
Multiple-Input-Multiple-Output (MIMO) technology makes use of multiple transmit and receive antennas to improve the spectral efficiency and reliability by spatial diversity and multiplexing. However, MIMO systems require a complicated detector to cancel the Inter-Antenna Interference (IAI). This talk presents three high-performance and energy-efficient MIMO detector designs.
First, an Iterative Detector and Decoder (IDD) design is presented. It improves the error rate performance by iterating the information between the detector and the decoder, thereby relaxing computation requirements. The idea is demonstrated in a 4×4 MIMO IDD ASIC incorporating Minimum Mean Squared Error (MMSE) detector and Non-Binary Low-Density Parity Check (NBLDPC) decoder. The use of NBLDPC decoder provides a superb error-correcting capability and simplifies the detector-decoder interface. Second, a Message-Passing Detector (MPD) for a massive MIMO system is presented. The MPD exploits the favorable massive MIMO channels to reduce the complexity while still maintains close-to-optimal BER-SNR performance. With the architectural optimization and the adaptive precision control, the MPD chip achieves a high detection throughput and an excellent energy efficiency. Finally, an iterative expectation propagation detector (EPD) is designed to address challenges of fast varying and highly correlated channel in a practical massive MIMO system for mobile applications. The EPD ASIC is implemented with condensed systolic array architecture and an approximate moment-matching circuitry to support link-adaptive processing with scalable energy consumption.