all-

here are some notes from a friend who attended the IEEE meeting. i
haven't had a chance to pick mark's brain on what the tone of the
presentation was but there's some interesting stuff in there.

> Attended IEEE Communication Society meeting in Denver. Presentation
> was "Next Generation Wireless: Who Needs Base Stations?" Presented
> by  Dr. Tim Brown, University of Colorado at Boulder.
> 
> The basic premise of his presentation was that cellular networks are
> not well suited for high bandwidth services, and that ad-hoc 802.11
> WiFi networks were delaying the acceptance of 3G wireless.
> 
> Some of the highlights:
> 
> Economic challenge:
> +What price for high quality, on demand, anywhere content...assume 1
>  cent/kbps per minute 
> +Services cost: MPEG video at 1.2mbps=$12/min, or MP3 at
>  128kpbs=$1.28/min 
> +This doesn't seem very affordable
> 
> What can be done?
> +Price=BW x bit-rate-price
> +Reduce BW demand, Increase supply, or use an Alternate model
> 
> The 3G challenge:
> +Spectrum is expensive
> +Base stations expensive
> +Performance maybe only 2x of 2G cellular
> +Handsets expensive
> +Killer app is voice, 2G already does that well.
> +So business proposition is uncertain
> 
> The WiFi comparison
> +Spectrum
> +Base stations $200
> +Performance, proven multi megabit rates
> +Handsets, many already own laptops w/wireless LAN cards
> +2G is complimentary service
> +Technology is already deployed
> 
> Ad Hoc networks or community networks
> +Cooperative wireless that emerges when wireless nodes are brought together
> +Communication dynamic, has to adjust to people moving away
> +Peer to peer, no one is in charge
> +Capacity & coverage increase w/more users [i.e. more access points]
> +Note, won't get coverage everywhere
> +Security also an issue. 
> Where are these networks?
> +Free nets in 50+ cities
>    example, Seattle; www.seattlewireless.net [showed map of several
>    hundred networks in downtown Seattle] 
> +Available in many airports, for like $8 day
> +Neighborhood, Joltage franchise, joltage does billing; 100k nodes by 2005?
> +Capacity is proportional to the number of base stations
>    FCC likes model, because they don't have to manage each little
>    piece of spectrum, no concern about companies not paying for
>    spectrum
> 
> Technical issues [resource management, QOS, battery]
> +Power issue, a 1 watt wireless lan card is a big percentage of
>  portable device
> +Power & distance, power goes up with distance
> +Routing in Ad Hoc networks
>    topo discovery
>    route choice
>    route maintenance
> 
> Discussed Dynamic Source Routing, current status:
> +power aware DSR design
> +implemented in ns-simulator [open source simulator]
> +implementing in laptop test bed using "click"
> +Also showed simulation and experimental routing results.
> 
> Conclusion:
> +Wide area cellular networks are not well suited for high bandwidth services
> +Wireless LAN networks complement wide area networks
> +Can in-vision tri-mode handsets that can hand-off
> +Ad hoc networks can expand coverage and capabilities of WLAN networks
> +Routing layer has a role in conserving battery energy


-- 
steve ulrich                       sulrich at botwerks.org
PGP: 8D0B 0EE9 E700 A6CF ABA7  AE5F 4FD4 07C9 133B FAFC