One of the presentations I attended today gave me an idea for an electronics project. In the "DTN Lake Sensor Network" talk, Stephen Farrell explained the Sensor Networking with Delay Tolerance (SeNDT) project. It's a proof-of-concept delay-tolerant sensor network that monitors water quality on a lake in Ireland. This kind of network demonstrates some of the techniques and challenges that space networks will face, thus the talk's inclusion in this conference.

Anyway, the project gave me the idea to think about how I might go about setting up a temperature monitor for a certain other lake of my acquaintance, with a remote display. This wouldn't be a sophisticated system at all—just a single sensor. SeNDT's sensors are mounted on buoys and powered by marine solar panels. I'd just hang a sensor off the dock or mooring buoy. There is mains electricity at the shore, though I imagine a wire connecting it to a water temperature sensor would tend to get in the way of shore- and dock-side activities. I'd have to look into how cheaply I could get an appropriate solar panel. 802.11b, like SeNDT uses, sounds like an easy choice for communicating to the wi-fi router at the house. The enclosure would be something to think about; I wouldn't leave it in the lake over the winter, but I would want the sensor to stay in the water over the summer. Operationally, I'd want it to provide readings at least once an hour between sunrise and sunset. So I assume I wouldn't need much in the way of a battery, or much in the way of a memory cache (it wouldn't need to buffer a bunch of readings; it'd just send 'em as it got 'em).