Thread Links Date Links
Thread Prev Thread Next Thread Index Date Prev Date Next Date Index

Re: [802.19] Microsoft/Harvard Protocols to Fix WhiteSpace Interference



Good points Ed...and don't even talk about detecting UWB systems like that....:0).
 
Beam forming techniques seem very applicable to TVWS use such as fixed WBA, and eventually, where one end is portable/mobile (the mobile part gets really interesting :o). Even without that sophistication, fixed wireless using ordinary directional antennas can be hard to find with simple energy detection.  Looking at 802.15.3c which recently completed, which uses beam forming, there are discovery mechanisms in the protocol to deal with the problem of finding a highly directional (and hard to detect) signal - so it's been though about some.  It requires the device you are trying to find does something specifically to make it more 'findable'.
 
As to how useful a simple detection mechanism can be, it depends on what you are trying to find.  If the *only* goal is protection of incumbents, and all the incumbent services transmit omni-directional signals at high power (sufficient SNR at the receiver), then simple energy detection can work.  
 
Hasn't there been a lot of work on spectrum sensing done in 802.22, 802.20?  I'm not up on the latest drafts for either of those, but maybe someone who is can chime in.  Seems we may be reinventing the wheel...and putting the corners in different places, of course.
 
-B
----- Original Message -----
From: Reuss, Ed
Sent: Thursday, August 20, 2009 8:21 AM
Subject: Re: [802.19] Microsoft/Harvard Protocols to Fix WhiteSpace Interference

So as per Section 4.2.1, this system uses simple amplitude detection [sqrt(I^2 + Q^2)] above a predefined threshold in the time domain to detect the presence of another network. This might be fine for existing wireless technologies, such as simple OFDM (most people don?t normally characterize OFDM as ?simple? J), but beamformed networks may be nearby that simple amplitude detection won?t detect unless it implements a beamformed receiver in the opposite direction, which I suspect would be unlikely.

 

Furthermore, modulation techniques may be developed for these bands, using lots of coding, that permit data transmission close to 0 dB SNR. These transmissions will not be detected either, regardless of which direction you point the beam.

 

I don?t wish to sound like a curmudgeon, and I recognize that such fancy modulation systems may never be developed for these bands, but if we accept such a simple detection system, then it guarantees that we will never be able use them even if we wanted to. On the other hand it?s unrealistic to require every device in this band to include a full-blown SDR with wide-band I-Q spectrum analysis capability.

 

-- Ed Reuss, Principal Engineer, Advanced Technology Group, Plantronics, Inc.

 


From: Bruce Kraemer [mailto:bkraemer@xxxxxxxxxxx]
Sent: Thursday, August 20, 2009 7:25 AM
To: STDS-802-19@xxxxxxxxxxxxxxxxx
Subject: Re: [802.19] Microsoft/Harvard Protocols to Fix WhiteSpace Interference

 

Richard et al,

The technology description can be found here.

 

http://research.microsoft.com/pubs/80952/whitefi.pdf

 

 

Bruce


From: Richard Paine [mailto:richard.h.paine@xxxxxxxxx]
Sent: Thursday, August 20, 2009 9:45 AM
To: STDS-802-19@xxxxxxxxxxxxxxxxx
Subject: [802.19] Microsoft/Harvard Protocols to Fix WhiteSpace Interference

 

Everybody who has been working on TVWS in IEEE 802 should read this, the source document, and the related material.

 

From FierceBroadbandWireless:

 

-------------------------------------------

 

Microsoft claims new protocol fixes interference problem for white-space devices

 

By Lynnette Luna

http://static.fiercemarkets.com/public/newsletter/assets/comment_ico.gifComment | http://static.fiercemarkets.com/public/newsletter/assets/email_ico.jpgForward

Researchers from Microsoft and Harvard University say they have developed a protocol that could be the foundation for new white-space devices.

The researchers believe the protocol could enable devices to avoid interference in white-space spectrum--those unused slivers of spectrum in the 700 MHz band between spectrum used by broadcast TV stations. The FCC approved the use of white-space devices last November on the condition that they don't interfere with nearby services, such as broadcast TV and wireless microphones. White-space devices are supposed to undergo a rigorous certification process before it is authorized. The vision for white space is to create super WiFi technology. The broadcasting industry has vigorously opposed any use of white-space devices, citing interference concerns. It hasn't been satisfied with the FCC's conclusion that interference won't be a major problem.

Microsoft and Harvard researchers presented their ideas this week at the ACM SIGCOMM 2009, a communications conference held in Barcelona, Spain, according to an article published on MIT's Technology Review Web site. The set of protocols are called "White-Fi," which are designed to avoid interference and make efficient use of the white-space spectrum since much of the interference has to do with devices using the same frequency at the same time. The protocol enables each device to evaluate the spectrum around it, discover available frequencies and continually search for interference, the researchers said. Once the interference is found, the protocol enables the device to jump to a different piece of spectrum at any time.

Microsoft recently obtained an experimental license from the FCC to build a prototype White-Fi system on the Microsoft Research Campus in Redmond, Wash., Microsoft researcher Ranveer Chandra told Technology Review. Microsoft plans to share its findings with the FCC.

For more:
- read Cnet

Related articles:
White space: Dell to make laptops; Google downright giddy
FCC unanimously approves unlicensed white-space devices
Microsoft intensifies white-space lobbying efforts

Read more about: WiFi, White-Fi, white space, protocol

 

 

Richard Paine

Success is getting what you want, happiness is liking what you get!

Phone:  206-854-8199

Skype:  richard.h.paine

Email:  richard.h.paine@xxxxxxxxx