IDNC vs. RLNC: Mission Impossible?
Mingchao Yu (ANU)
APPLIED SIGNAL PROCESSING SERIESDATE: 2012-11-15
TIME: 10:00:00 - 11:00:00
LOCATION: RSISE Seminar Room, ground floor, building 115, cnr. North and Daley Roads, ANU
CONTACT: JavaScript must be enabled to display this email address.
ABSTRACT:
Network coding is an evolution of traditional data communications. By allowing simple encoding at the sender- s/intermediate nodes and simple decoding at the receivers, network coding has shown great capability to improve the throughput and delay performance in a wide range of scenarios, including single/multi-hop wired/wireless transmissions. In this talk, two popular network coding techniques for single-hop wireless broadcast are presented and compared. One is random linear network coding (RLNC), which is throughput optimal but is inefficient in decoding delay. The other one is instantly decodable network coding (IDNC), which allows a subset of the receivers to decode in every transmission. While RLNC is highly structured and can be well described mathematically, IDNC encoding is recognized to be highly opportunistic and thus is hard to be tracked and evaluated. Hence, the performance comparison between IDNC and RLNC is yet a amission impossiblea, which is the obstacle to adaptively chose between them in order to maximizing the benefit of network coding. To overcome this obstacle, in this talk we will establish a framework which offers a fair transmission baseline for both RLNC and IDNC. Under this framework, IDNC exhibits interesting properties which reveal that it is actually trackable and can be mathematically described. We then derive throughput bounds and design optimal encoding algorithm for IDNC. We then compare the throughput and decoding delay performances of IDNC and RLNC. The results show that there is no clear winner between them. They outperform each other under different parameter settings and we can easily refer to these parameters to implement adaptive transmission schemes.
BIO:
Mr. Mingchao Yu has received his Bachelor degree in 2008 in Northwestern Polytechnical University, China. He then pursued MPhil in ANU from 2010-2011, aiming at the evaluation, implementation and improvement of digital video broadcasting standards. Currently he is a first-year Ph.D. student in ANU under the project aOptimizing Throughput and Delay in Network Coded Systemsa. His research interests are OFDM techniques, channel estimation techniques, physical layer transceiver design, FEC code design, and network coding.
