“It’s Not Dead Yet!” - MPEG-2 Video Coding Efficiency Improvements

Learn about significant new improvements to MPEG-2 Video coding, how this enables more services to legacy set-top boxes and receivers, and about the revolutionary new TANDBERG EN8100 Advanced MPEG-2 Broadcast Encoder. ABSTRACT Broadcasters are required by federal regulation, and locked into in practice due to the millions of consumer digital televisions (DTVs) purchased, to use the MPEG-2 Video format for free over-the-air DTV service to the home. The 19.39Mbps bandwidth is already a constraint for those broadcasters that offer multiple DTV services in a single 6MHz channel. With the advent of new Mobile/Handheld services being introduced in 2009, the need for more efficient coding is paramount. Any available bandwidth in a 6MHz channel is at a premium; thus, improvements in MPEG-2 Video coding efficiency are extremely valuable. Contrary to widespread belief, MPEG-2 Video has not yet reached its maximum performance. Improvements in integrated circuit speed and intelligence mean that the amount of information that can be processed in real-time is enabling new approaches to coding MPEG-2 Video that before were not possible. Examples include fully exhaustive motion searches and predictive processing, rate-distortion optimization (RDO) algorithms, and lessons learned from implementing MPEG-4 AVC coding retrospectively applied to MPEG-2 Video to improve the use of the compression algorithms. This paper explores new approaches to coding MPEG-2 Video efficiently. Wherever possible, simulation results have be used to demonstrate the improvements. The MPEG-2 Video standard was ratified in 1994 and the selected set of algorithmic tools and the decisions made on the theoretical processing models were based on the state of integrated circuit technology of that time. Over the past decade, there has been an estimated 40 fold increase in integrated circuit processing power, yet the past several years have not produced improvements in MPEG-2 Video coding efficiency of more than approximately five percent. As such, the general industry belief is that further improvements to MPEG-2 Video coding efficiency is limited by the standard itself. However, taking a fresh look at MPEG-2 Video, by combining the new integrated circuit processing power with knowledge obtained through the design of MPEG-4 AVC encoders, has shown that it is possible to increase the coding efficiency of MPEG-2 Video by more than 15%. Such a Revolutionary Optimized MPEG-2 Video Encoder is practical to implement today. Three areas are investigated: • Look-ahead encoding • Two-stage motion estimation • Pre-processing Look-Ahead Encoding Look-ahead encoding is the process of using a …