I-LENSE Directed Diffusion

Directed diffusion is a novel data-centric, data disemmination paradigm for sensor networks. Directed diffusion has some novel features: data-centric dissemination, reinforcement-based adaptation to the empirically best path, and in-network data aggregation and caching. These features can enable highly energy-efficient and robust dissemination in dynamic sensor networks, while at the same time minimizing the per-node configuration that is characteristic of today's networks.

Directed diffusion consists of several elements. Data is named using attribute-value pairs. A sensing task (or a subtask thereof) is disseminated throughout the sensor network as an interest for named data. This dissemination sets up gradients within the network designed to "draw" events (i.e., data matching the interest). Events start flowing towards the originators of interests along multiple paths. The sensor network reinforces one, or a small number of these paths.

More details can be found in the publications on directed diffusion.

Software

There are two implementations of diffusion, some running in different environments:

• full directed diffusion: runs on 32-bit sensor nodes over native hardware and radios, and on PCs over Ethernet
• full directed diffusion in ns: runs diffusion embedded in the ns-2 simulator
• tiny diffusion: runs on UCB Motes

Full diffusion in ns: Directed diffusion is fully supported into ns-2.1b9. You can download the latest version of ns-2 with diffusion in the ns-2 webpage.

Full diffusion: Diffusion 3.3.0-beta is the latest version of directed diffusion and includes support for two-phase pull, one-phase push, and GEAR algorithms. In addition, it includes both RMST and RBP filters (developed by Fred Stann). It can be used to wired Ethernet and with the Sensoria radios (when running on the Sensoria nodes). In addition to running on Red Hat Linux versions 7.2, 7.3, 8.0, 9.0, and the more recent Fedora Core Releases, diffusion also compiles and runs on other versions of Unix (such as Mac OS X).

For more information, see Diffusion 3.3.0-beta (released 9-Nov-06).

Current Tarball: The current repository of diffusion (available read-only) is at CENS.
The most recent checkin to the repository (as of 20-Sept-05) is also available here

The mailing list for directed diffusion is [email protected]. You can subscribe to this mailing list at the diffusion users info page.

Download older releases: 3.0.1, 3.0.2, 3.0.3, 3.0.4, 3.0.5, 3.0.6, 3.0.7, 3.1.0, 3.1.1, 3.1.2, 3.1.3, 3.2.0

Tiny diffusion: Tiny difufsion is currently maintained at CENS. More information can be obtained here.

RMST: A new version of RMST (a transport layer for diffusion) is included in the current diffusion tarball. The latest version of RMST has been debugged on the ISI testbed. The original version used for the paper is still available in ns-2 (2.28). More information can be obtained here

TopoGen: TopoGen is a tool for generating random topologies for simulation experiments. It can be configured to generate EmSim run files or ns-2 tcl files. For source code & information, topo_gen (released 10-May-05).