Information Transmission over Two-Way Networks

Abstract

Shannon's two-way channels (TWCs) allow two terminals to exchange data streams in a full-duplex manner and enable interactive adaptive coding to improve transmission reliability. However, TWCs are often used in conjunction with orthogonal multiplexing to mitigate the interference incurred from simultaneous transmissions over a shared channel. To date, TWCs with non-orthogonal multiplexing are still not fully explored. This thesis examines channel capacity problems for TWCs and identifies coding methods to facilitate and enhance two-way simultaneous transmission.

We first make use of channel symmetry properties to determine the capacity region of three types of two-way networks: (a) two-terminal discrete-memoryless TWCs (DM-TWCs), (b) two-terminal TWCs with memory, and (c) three-terminal multiaccess/degraded broadcast DM-TWCs. For each network, symmetry conditions under which a Shannon-type random coding inner bound (under independent non-adaptive inputs) is tight are given. The results broaden the class of TWCs whose capacity region can be exactly determined and imply that interactive adaptive coding does not enlarge the capacity region and is hence unnecessary for such channels. Moreover, we generalize Shannon's push-to-talk TWC and analytically derive this generalized channel's capacity region, which is a convex hull of at most four rate pairs. For general two-terminal DM-TWCs that lack channel symmetry properties, a simple outer bound is further derived to obtain approximation capacity results.

In addition to examining capacity problems, we also study joint source-channel coding (JSCC) for TWCs. Specifically, we propose an adaptive lossy JSCC scheme for sending correlated sources over two-terminal DM-TWCs. Our idea is to couple the independent operations of the terminals via an adaptive coding mechanism which can mitigate cross-interference resulting from simultaneous channel transmissions and concurrently exploit the sources' correlation to reduce the end-to-end reconstruction distortions. Our adaptive JSCC scheme not only subsumes existing lossy coding methods for the same setup, but it also improves on their performance. Several examples are given for illustration. Moreover, we derive outer bounds for our two-way lossy transmission problem and establish complete JSCC theorems in some special settings. In these special cases, a non-adaptive separate source-channel coding scheme achieves the optimal performance, thus significantly simplifying the design of the source-channel communication system.