PhD Prospectus Defense: Wenbo He

Starts:
9:30 am on Friday, December 19, 2014
Ends:
11:30 am on Friday, December 19, 2014
Location:
15 Saint Mary's Street, Room 301C
Consider a multiple-input multiple-output (MIMO) Gaussian multiple-access channel (MAC) with channel matrix H and a MIMO Gaussian broadcast channel (BC) with channel matrix transpose of H. It is well known that the achievable rate tuples for these two channels are the same under the same total power constraint, which is called the uplink-downlink duality relationship. In this work, we consider the integer-forcing strategy where the channel is steered towards an integer-valued effective channel matrix by beamforming at the transmitter(s) and projection at the receiver(s) such that the receiver(s) can decode integer-linear combinations of the transmitted codewords. Recent efforts have demonstrated the benefits of this strategy for uplink, downlink, and interference alignment scenarios. In this work, we establish an uplink-downlink duality relationship for integer-forcing. Specifically, consider an uplink channel that consists of L multi-antenna transmitters, a single N-antenna receiver and a channel matrix H with target integer matrix A. The corresponding downlink dual channel consists of a single N-antenna transmitter (base station), L multi-antenna receivers and a channel matrix transpose of H with target integer matrix transpose of A. We show that any computation rate tuple that is achievable in the uplink is achievable for the same total power in the downlink and vice versa. Furthermore, we introduce dirty-paper integer-forcing for the downlink channel and show its duality relationship with successive integer-forcing for the uplink channel.