Video content delivery over wireless networks is expected to grow exponentially in the coming years, driven by the desire to have ubiquitous access to TV content and interactive video services across multiple platforms, including TVs, PCs, mobile devices, and embedded devices (i.e., vehicles). This trend is supported by recent studies (Cisco Visual Networking Index) which predict that the potential traffic going to mobile devices in the future is expected to increase by a factor of 66 times by 2013. However, current wireless networks are woefully unprepared today. Since advances in compression technology only improve efficiency by 2-3x every ~10 years, and wireless spectral efficiency has only improved by ~2x over the past 5 years and is predicted to improve by 2x over the next 5 years, these trends will not be sufficient to enable new video-based usage models of the future. Opening up more spectrum, shrinking cell sizes, and leveraging multiple heterogeneous wireless networks will definitely be important, but I also believe more can be done with greater end-to-end system optimizations for video content delivery. This can include, for example, exploiting perceptual quality metrics for video transport optimizations, leveraging scalable compression techniques in intelligent ways, building video-aware heterogeneous network designs with intelligent priority based routing and network coding, enhanced video post-processing, and cross-layer optimizations like joint source-channel coding. This talk will try to motivate the need for building more ‘video-awareness’ into mobile devices as well as end-to-end systems, and I will describe some initial work we are doing evaluating the capacity of current and future WiMax and LTE systems, scalable compression, unequal error protection using Raptor codes, and joint source-channel coding for WiFi and WiMax systems. I’ll conclude by presenting some future challenges and opportunities in this area.
Dr. Jeff Foerster joined Intel in August 2000 as a Wireless Researcher in Hillsboro, Oregon, and is currently a Principal Engineer in the Wireless Communications Lab. He currently leads a team focused on Wireless Multimedia Solutions, and his past research has included Ultra-wideband (UWB) technology and related regulations, 60 GHz system design, and wireless displays. He chaired the channel modeling sub-committee for the IEEE 802.15.SG3a study group focusing on UWB channel models to be used for evaluating future UWB based proposals which has been widely used in the industry and academia, and chaired the regulatory group within WiMedia. Jeff was Vice Chair of the Technical Program Committee of the ICUWB’07 conference and has served on the Technical Program Committee for the ICC WCS, GlobeComm, WCNC, and UWBST conferences. Jeff has published 15 IEEE papers including journals, magazine, and conferences, has been an invited panelist at several conferences, was lead author of a book chapter on UWB, and has contributed to several international regulatory bodies including CEPT TG3 and ITU TG1/8. Prior to joining Intel, he worked on Broadband Wireless Access (BWA) systems and standards, and was technical editor of the IEEE 802.16 standard for a short time. He received his B.S., M.S., and Ph.D. degrees from the University of California, San Diego, where his thesis focused on adaptive interference suppression and coding techniques for CDMA systems.