| 
  • If you are citizen of an European Union member nation, you may not use this service unless you are at least 16 years old.

  • Finally, you can manage your Google Docs, uploads, and email attachments (plus Dropbox and Slack files) in one convenient place. Claim a free account, and in less than 2 minutes, Dokkio (from the makers of PBworks) can automatically organize your content for you.

View
 

Ad-hoc Networking

Page history last edited by Ralph 11 years, 4 months ago

 

Ralph Morelli, Trishan de Lanerolle of Trinity College,  and their

students (Antonio Alcorn, UConn,  Christopher Fei and Prasanna

Gautam, Trinity,  and Qianqian Lin, Wesleyan)

presented POSIT (http://posit.hfoss.org/),  an Android-based search and rescue

tool that implements  a form of ad-hoc networking called Random Walk Gossip (RWG). 

This is work in conjunction with the Prof. Simin Nadjm-Tehrani, Mikael Asplund, and Gustav

Nykvist of the Real Time Systems Laboratory at Linkoping University in Sweden, where

RWG was developed (http://www.ida.liu.se/~rtslab/HFN/rwg/index.shtml). 

 

  • The POSIT project is looking for users to test the system.
  • RWG is based on rap sockets, not UDP.  It randomly propogate messages through a partitioned or intermittently connected network.
  • Field tests demonstrate that phone-to-phone broadcast messages are transmitted  at up to  400 meters, assuming  line-of-sight.
  • RWG is designed to minimize power usage and bandwidth requirements.
  • Other possible application were discussed:
    • pH sensor data aggregation?
  • Possible optimizations were discussed:
    • Simulators: optnet, ns2
    • Possibility of p2p using cell radio
    • Adjust power level to conserve battery
    • DD-WRT for example.
    • Optimize algorithms: instead of complete randomization, look at contact history. A person may migrate between 2 groups for a reason, or regularly, e.g. for work. ("stability metric")
    • Oscar's AODV algorithm: modified to transmit video, photos, etc. high-bandwidt and uses a stability metric based on how long 2 nodes are in touch, on average. Simulated in NS2.
  • SMS
    • Robert Kirkpatrick reported that SMS: didn't go down in Aceh; latencies were high.  In practice, SMS is usually last to go down, first to come back.
    • Data encoding/structured SMS?

    • Air traffic control over SMS?
    • MMS: less reliable, but messages queue up, then are sent once network comes back up
  • User experiences are important.
    • Must be independent of protocol/technology.
    • User Interface: Firefighters can't read POSIT screen on the job.

    • Pony express message delivery
    • See Project Guardian (http://www.netsquared.org/projects/guardian-secure-private-anonymous-telephone-built-google-android) for ideas on privacy and security for on Android.
  • Design Issues/Suggestions
    • Use preferred system, then failover to SMS

    • Get text out if you can't send the pictures because of bandwidth

    •  In disaster area, get coverage by driving around

    • Get into an exercise.

      •  Don't try to test prototypes in a disaster zone; becomes a nuisance to relief efforts.

    • Try to attend Camp Roberts exercise (in CA in November) and the Google Hacathon.

    • Generic synchronization of data structures; server drives around and updates extended RSS feeds

  • Observations from disaster/crisis scenarios.

    • In Afghanistan: cell towers are autonomous, fenced in, with generator, guarded, satellite up link.

    • In US, most critical infrastructure is in the hands of the private sector

    • In Mexico Zapatistas want internet;

    • Mexico has a  cell # database -- security issues.

    • Military capability to create WiFi corridor: drop long-range heavy duty military routers every mile.

    • Mexico 1988 computer voting fraud: system "crashed" at a crucial time during the election

 

Comments (0)

You don't have permission to comment on this page.