We design and analyze noncoherent detection schemes for multi-user/multi-node communication systems where neither the transmitters nor the receiver knows the channel. First, we propose differential decoding schemes for two-user MIMO systems based on orthogonal space-time block codes (OSTBCs). We derive low complexity differential decoders for users with two transmit antennas. We also present differential decoding schemes that achieve full diversity, perform significantly better than the existing schemes, and work for any square OSTBC, but need higher decoding complexity compared to our low complexity decoders. Moreover, we analyze the diversity of the proposed schemes. To the best of our knowledge, our low complexity schemes are the first low complexity differential schemes for multi-user systems.
We then propose differential decoding schemes for asynchronous multi-user MIMO systems based on OSTBCs. We derive novel low complexity differential decoders by performing interference cancelation in time. The decoding complexity of these schemes grows linearly with the number of users. We also present differential decoding schemes that perform significantly better than our low complexity decoders and outperform the existing synchronous differential schemes but require higher decoding complexity compared to our low complexity decoders. The proposed schemes work for any square OSTBC, any number of users, and any number of receive antennas. Furthermore, we analyze the diversity of the proposed schemes and derive conditions under which our schemes provide full diversity. To the best of our knowledge, the proposed differential detection schemes are the first differential schemes for asynchronous multi-user systems.
Finally, we present novel distributed beamforming (DBF) algorithms using feedback control based on Tree-Structured Vector Quantization (TSVQ). We develop TSVQ-based DBF algorithms for static channels. To the best of our knowledge, the proposed algorithms are the first deterministic DBF methods that can feed back more than 1 bit per time slot for faster phase synchronization. We analytically prove that our TSVQ-based DBF algorithms attain phase synchronization in probabilistic senses. Moreover, we modify our TSVQ-based DBF algorithms to enable them to track time-varying channels without the knowledge of the channel. Simulation results demonstrate that our algorithms significantly outperform the existing adaptive DBF algorithms for static and time-varying channels.