On computing the feedback capacity of channels and the feed-forward rate-distortion function of sources

Authors:
Ramji Venkataramanan;S. Sandeep Pradhan
Affiliations:
Stanford University and EECS Dept., University of Michigan, Ann Arbor;EECS Dept., University of Michigan, Ann Arbor
Venue:
IEEE Transactions on Communications
Year:
2010

Citing 10
Cited 0

Information Theory: Coding Theorems for Discrete Memoryless Systems

Information Theory: Coding Theorems for Discrete Memoryless Systems
The capacity of channels with feedback

IEEE Transactions on Information Theory
Capacity results for the discrete memoryless network

IEEE Transactions on Information Theory
To code, or not to code: lossy source-channel communication revisited

IEEE Transactions on Information Theory
Duality between source coding and channel coding and its extension to the side information case

IEEE Transactions on Information Theory
On competitive prediction and its relation to rate-distortion theory

IEEE Transactions on Information Theory
Feedback capacity of the first-order moving average Gaussian channel

IEEE Transactions on Information Theory
On the Role of Feedforward in Gaussian Sources: Point-to-Point Source Coding and Multiple Description Source Coding

IEEE Transactions on Information Theory
Source Coding With Feed-Forward: Rate-Distortion Theorems and Error Exponents for a General Source

IEEE Transactions on Information Theory
A Coding Theorem for a Class of Stationary Channels With Feedback

IEEE Transactions on Information Theory

Quantified Score

Hi-index	0.00

Visualization

Abstract

The problem of computing the capacity-cost function of channels with feedback and the rate-distortion function of sources with feed-forward is considered. Sufficient conditions are derived on : a) the structure of the cost function for a chosen joint distribution to achieve the optimal feedback capacitycost function, b) the structure of the distortion function for a chosen joint distribution to achieve the optimal feed-forward rate-distortion function. These structural results are useful since it is infeasible in general to directly compute the optimizations. Examples are provided to show how the results can help compute the performance limits with feedback and feed-forward.