Optimization and inference in machine learning and physics Workshop

Generalized Belief Propagation Receiver for Near-Optimal Detection of Two-Dimensional Channels with Memory

author: Noam Shental, Weizmann Institute

Description

We propose a generalized belief propagation (GBP) receiver for two-dimensional (2-D) channels with memory, which is applicative to 2-D inter-symbol interference (ISI) equalization and multi-user detection (MUD). Our experimental study demonstrates that under non-trivial interference conditions, the performance of this fully tractable GBP receiver is almost identical to the performance of the optimal maximum a-posteriori (MAP) receiver.

You might be experiencing some problems with Your Video player.

*Slides*
0:05	Generalized Belief Propagation receiver
1:20	Examples of 2-D channels
1:56	Examples of 2-D channels
3:19	Outline
4:27	System model
4:51	Outline (cont.)
5:40	System model (cont.)
6:19	Optimal detection
7:58	2-D channels as undirected graphical models
8:28	Examples of 2-D channel representations
10:00	System model
10:28	2-D channels as undirected graphical models
12:11	Exact inference – junction tree
12:25	Approximate inference: belief propagation
12:49	Generalized belief propagation (GBP)
13:17	Generalized belief propagation (GBP)
13:53	Generalized belief propagation (GBP)
21:26	Results: ISI equalization
26:27	Results: hexagonal topology cellular network
28:19	Conclusions - ITW
29:01	Current work
34:17	Approximate free energy
36:03	The connection between the free energy and the information rate
39:30	The connection between the free energy and the information rate (cont.)
40:19	The connection between free energy and symmetric information rate
41:04	The connection between free energy and the symmetric information rate (cont.)
41:30	Experimental results ISI
42:42	Experimental results: Wyner’s model
43:21	Experimental results: Wyner’s model
43:45	Why do GBP-based CVM serve so remarkably?
50:08	Local estimates and GBP

Lecture rating

People found this lecture:
Worth seeing
because it is:
Valuable and informative
Well presented
Easily understandable
Acceptably recorded
You need to login to cast your vote.

Report a problem or upload files

If you have found a problem with this lecture or would like to send us extra material, articles, exercises, etc., please use our ticket system to describe your request and upload the data.
Enter your e-mail into the 'Cc' field, and we will keep you updated with your request's status.