To code, or not to code: lossy source-channel communication revisited

By Michael C. Gastpar, B. Rimoldi, Martin Vetterli. IEEE Transactions on Information Theory, 2003

545 citations recorded by the source corpus.

Subfields: Computer Networks and Communications, Electrical and Electronic Engineering

View Paper: https://www.semanticscholar.org/paper/406166145d0a322c16aa91c2dba0c11a772f066d

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Networks and Communications

In Electrical and Electronic Engineering

Citations in the corpus

Beyond Transmitting Bits: Context, Semantics, and Task-Oriented Communications (2022, unknown): “Necessary and sufficient matching conditions are given in [195] for general source and channel distributions.”
On the capacity of large Gaussian relay networks (2005, background): “Moreover, it also shows that in a scaling sense, as becomes large, an optimal relay “recoding” strategy is simply uncoded forwarding.”
Transporting information and energy simultaneously (2008, background): “[22] for undetermined b(·), but this path is not pursued here.”
Information Theory of Decisions and Actions (2011, background): “This view is of particular interest due to the discovery of the general existence of optimally matched channels (Gastpar et al. 2003).”
Source-Channel Communication in Sensor Networks (2003, methodology): “For the point-to-point transmission of an iid Gaussian source across an additive white Gaussian channel, it is well-known that uncoded transmission is optimal [10], see also [5,6,7].”
Multiuser MIMO Achievable Rates With Downlink Training and Channel State Feedback (2007, background): “It is well-known (see for example [56] and references therein) that “analog transmission” (the source signal is input directly to the channel after suitable power scaling) is an optimal strategy to send an i.”
Fast transfer of channel state information in wireless systems (2006, methodology): “In particular, if an uncorrelated discrete-time Gaussian random process has to be transmitted over a discrete-time additive Gaussian white noise channel having the same sampling-rate, and if the criterion for reproducing the message at the receiver is minimum mean-square error, then direct uncoded a”
Estimation Diversity and Energy Efficiency in Distributed Sensing (2007, methodology): “It is well known ([19], [20]) that for a single Gaussian source with an Additi ve White Gaussian Noise (AWGN) channel, both the digital and analog approaches can retain the optimal power-distortion tradeoff.”
Uncoded transmission is exactly optimal for a simple Gaussian "sensor" network (2007, methodology): “Straightforward evaluation of the measure-matching conditions as given in [14], [15] leads, not surprisingly, to the mean-squared error distortion measure.”
DeepJSCC-f: Deep Joint Source-Channel Coding of Images With Feedback (2019, background): “Although there have been many research efforts on joint source-channel coding techniques, they have mostly focused either on theoretical analysis under some idealistic source and channel distributions [6]–[10], or the joint optimization of the parameters of the components (vector quantizer, index as”
Information Bottleneck for Gaussian Variables (2003, background): “In that information theoretic sense, IB isself dual, where the resulting source and channel are perfectly matched (as in Gastpar and Vetterli, 2003).”