University of Southern California WiDeS - Wireless Devices and Systems Group The USC Andrew and Erna Viterbi School of Engineering USC



Video transmission is currently the main driver for the increase of both wired and wireless data traffic.  In wired networks, movie streaming currently accounts for 50% of all internet traffic during evening hours and further growth is expected. A similar trend is observed for wireless networks. Wireless data traffic is expected to increase by a factor of 40 over the next five years, from currently 93 Petabytes to 3600 Petabytes per month. This explosive demand is fueled mainly by video traffic, which is expected to increase by a factor of 65 times, and become the by far dominant source of data traffic. 
This increasing demand offers new business opportunities but also poses an enormous challenge from a technical and economical perspective. Traditional methods for increasing data throughput in wireless networks have relied on the following three approaches: (i) increase of spectrum usage, (ii) increase of the per--link spectral efficiency, and (iii) increase of spatial reuse. Increasing the amount of spectrum is limited by the fact that spectrum is a finite resource. Increasing the spectral efficiency per link is also approaching its limits: fourth--generation cellular systems such as LTE have a near--optimal physical layer, using OFDM together with capacity--approaching  codes and multiple antenna elements.This leaves as the main measure the decrease of distance between transmitter and receiver, and thus the increase of the area spectral efficiency. This approach --which can be combined with increased spectrum and more efficient links --is highly scalable, and can thus provide the order--of-magnitude increase in capacity that is required for future video networks. The Achilles heel lies in the fact that every Base Station (BS) --including femto-stations --needs a high-speed backhaul


In our work, we have developed a radically new approach that is based on the following two key observations: (i) a large amount of video traffic is caused by a few, popular, files and (ii) disk storage is a quantity that increases faster than any other component in communications/processing systems. The essential idea of our approach thus is to trade off backhaul capacity with caching of video files at local base stations or helper nodes from where they can be transmitted very efficiently. In other words, we envision a proliferation of low-complexity base stations with weak backhaul links, which we call "helper nodes". Those nodes obtain the most popular video files by downloading through their weak backhaul links (or even completely without backhaul, through methods discussed later on). Whenever a user close to them needs a video file that they have stored, they transmit to it over a wireless link that is short-distance and thus has extremely high area spectral efficiency. Our concept is pushed even further by introducing the notion of mobile devices themselves as helper nodes. Recent years have seen an enormous proliferation of smartphones and tablets that have anywhere between 10 and 64 GByte of storage (not to mention the 500 GByte on typical laptop hard disks). By enabling device-to-device communications, the ensemble of mobile devices can become a distributed cache that allows a much more efficient downloading.    


Under Constructions




Journal Papers

N. Golrezaei, A.G. Dimakis, and A.F. Molisch, "Wireless Device-to-Device Communications with Distributed Caching", Submitted for Publication.

N. Golrezaei, A. F. Molisch, A. Dimakis, and G. Caire, “Femtocaching and Device-to-Device Collaboration: A new architecture for wireless video distribution,” IEEE Comm. Mag., in press. 

Conference Papers

N. Golrezai, A. F. Dimakis, and A. F. Molisch, “Device-to-Device Collaboration through Distributed Storage”, IEEE Globecom 2012

N. Golrezai, A. G. Dimakis, and A. F. Molisch, “Device-to-Device Communications for Wireless Video Delivery”, Asilomar 2012 (invited).

N. Golrezai, A. G. Dimakis, and A. F. Molisch, “Wireless Device-to-Device Communications with Distributed Caching”, ISIT 2012.

N. Golrezai, A. F. Molisch, and A. Dimakis, “Base-Station Assisted Device-to-Device Communications for High-Throughput Wireless Video Networks”, IEEE ICC Workshop on Video-Aware Wireless Networks 2012.

N. Golrezai, K. Shanmugam, A. Dimakis, A. F. Molisch, G. Caire, “Wireless Video Content Delivery through Coded Distributed Caching”, IEEE ICC 2012, in press.

N. Golrezai, K. Shanmugam, A. Dimakis, A. F. Molisch, G. Caire, “FemtoCaching: Wireless Video Content Delivery through Distributed Caching Helpers”, IEEE Infocom 2012.

N. Golrezaei, A.G. Dimakis and A.F. Molisch, "Wireless Video Content Delivery through Distributed Caching and Peer-to-Peer Gossiping," Asilomar Conference on Signals, Systems, and Computers, 2011 (Invited).