Cooperative Source and Channel Coding
In wireless communications, the channel conditions vary over time due to signal fading and multipath propagation. If a wireless device can transmit simultaneously over different channels, there is a chance that some of these channels may have better conditions than others; therefore the overall performance will improve. The widely known Multiple Input-Multiple Output (MIMO) technology works on this principle. While MIMO provides spatial diversity through multiple antennas on the same wireless device, user cooperation does so through a partner's antenna. This means that spatial diversity can be obtained in small size wireless devices which cannot have more than one antennas. The idea behind cooperation is, when a wireless device transmits signal, the signal is "overheard" by another wireless device (the partner). If the partner also sends a copy of that signal to the destination, spatial diversity is achieved.
In order for cooperative diversity to benefit a particular application, such as wireless video, there are many questions to be answered. For example:
- How do we allocate bandwidth between partners to optimize end to end source quality?
- Since resources are limited, a partner may not be able to retransmit the entire signal that it "overhears". How can partners selectively retransmit a portion of each other's signal to optimize overall performance?
- Since only a portion of the signal is transmitted cooperatively (given the above bullet), how does a higher layer application divide its transmission session into cooperative and non-cooperative? And how would this affect the end to end quality of service for the underlying application?
In this project we are interested in answering the above questions both from an information theoretic perspective and also for practical source and channel coders. In particular, we investigate real-time multimedia communications over wireless. We will include video simulations and channel simulations while jointly optimizing parameters at both layers. We will gather performance measures from simulation to determine how cooperation can benefit wireless multimedia communications.
Philips Research USA, Braircliff Manor, NY
- D. Gunduz and E. Erkip. Source and channel coding for cooperative relaying. In Proceedings of 2005 International Workshop on Signal Processing Advances for Wireless Communications (SPAWC), New York City, June 2005(invited paper).
- X. Xu , Y. Wang and E. Erkip . Layered cooperation for wireless multimedia communications . In Proceedings of 2004 Picture Coding Symposium, San Francisco , December 2004 (invited paper).
- X. Xu, D. Gunduz, E. Erkip and Y. Wang. Layered
cooperative source and channel coding. 2005 ICC Multimedia Communication
and Home Networking Symposium,
- D. Gunduz and E.Erkip. Joint source-channel cooperation: Diversity versus spectral efficiency. IEEE International Symposium on Information Theory, Chicago, June 2004.
Faculties:
· Elza Erkip (homepage)
· Yao Wang (homepage)
Students:
· Deniz Gunduz(homepage)
· Hoi Yin Mok (homepage)
· Ozgu Alay (homepage)