Far Out Networking

 

Much effort has gone into wireless ad-hoc networking protocols for small, battery operated computers. Generally, these protocols do not scale well as the complexity of communication of increases rapidly, especially when mobility is required for ad hoc networks.

We approach the scaling problem of mobile ad hoc by avoiding general protocols in favor of protocols that require only limited knowledge but that cover wide ranges of problems.


Far Out Networking   Application Specific Network Patterns (0.11)

For an overview, see the presentation and the report that came out of our involvement with DARPA's Symposium on Far Out Networking.

Following are papers that go in depth into some of our results. 

 
 

 


The Census Protocol: Visiting Every Node in a Large Mobile Ad-hoc Network

Visiting every node in a network is important but in a mobile ad-hoc network (MANET) it's hard because of the high messaging overhead to maintain the structure during mobility. Census uses a gossip-style protocol with short, local gradients that guide it towards unvisited nodes. We show robustness and scaling for 150 to 4000 nodes.

This work was supported in part by Defense Advanced Research Products Agency (DARPA)’s Fixed Wireless at a Distance program under contract FA8750-12-C-0278. The views expressed are those of the authors and do not reflect the official policy or position of the Department of Defense or the U.S. Government. Approval for public release was granted by DARPA on April 25, 2014

Census: A protocol for Visiting All Nodes in MANETs Using Biased Random Walks (pdf)

 


The Repair-Time Scaling Wall for MANETs: Why it Happens and How to Extend It

Mobile ad-hoc networks (MANETs) are often viewed as not scalable beyond about 100 nodes due to insufficient network capacity for supporting related control traffic. We find that this is not the bottleneck; rather the interaction between path failure and consequent repair limits scale. We consider the factors that contribute to this scaling wall and identify techniques to extend scaling limits.

This work was supported in part by Defense Advanced Research Products Agency (DARPA)’s Fixed Wireless at a Distance program under contract FA8750-12-C-0278. The views expressed are those of the authors and do not reflect the official policy or position of the Department of Defense or the U.S. Government. Approval for public release was granted by DARPA on March 18, 2014.

On the Repair Time Scaling Wall for MANETs (pdf)