From Constrained Signaling to Network Interference Alignment via an
Information-Estimation Perspective
Shlomo Shamai. EE Dept. Technion-Israel Institute of Technology.
Abstract
A subjective overview of selected topics in Information Theory,
Communications, and Signal Processing will be presented taking an Information-Estimation
perspective.
We first review basic information-estimation (I-MMSE) relations and properties
in Gaussian additive channels, and then address via this framework specific
problems and results in:
- Nonlinear optimal filtering;
- Constrained signaling;
- Multi-terminal models, touching upon broadcast, interference, wiretap,
and cognitive settings as well cellular networks of the Wyner type,
and novel aspects of interference alignment;
- Efficient codes for point-to-point, interference and relay channels;
A statistical physics view of I-MMSE related relations,
is shortly mentioned. So are, time permitting, different
other applications motivated by the incremental channel decomposition,
namely deterministic and random vector Gaussian channels,
and the broadcasting strategy (variable-to-fixed channel coding).
Based on the overviewed framework, we emphasize specific
research challenges in these areas motivated by the
information-estimation perspective and close the lecture
with an outlook examining general channels and network
information theoretic concepts.