Visual Sensor Networks

VSN (Visual Sensor Networks) intends to address the issues in a visual sensor networks related to network protocols, computer vision algorithms and their interaction. Current work focus on: setting up a Visual Sensor Network testbed energy consumption issues

Energy Models for Sensor Networks

The main goal of this work is to understand the energy consumption trade-offs between computation and communication in sensor networks in general, and visual sensor networks, in particular. In order to do so, we need to evaluate and model: energy spent to process data by sensors (e.g., how much energy vision algorithms require, etc.) as well as energy required for communications.

The first step to evaluate energy consumption due to communications is to obtain an accurate energy model. Such energy model must be as close to reality as possible, taking into account all radio states, i.e., energy spent not only while transmitting and receiving a packet, but also while in idle, overhearing, or sleep modes. To this end, we developed such a model and used it to instrument the QualNet and GloMoSim network simulators.

The next step is to understand energy consumption due to sensing and processing. Our goal is to develop a model that captures the behavior of sensing and processing activities. More specifically, video processing should consider not only capturing, but also compression and computer vision algorithms that provide several different levels of data representation.

We intend to evaluate energy consumption in three different platforms: laptops (ongoing work), Crossbow Stargate single board computer, and Crossbow Mica2 Motes.

Next a methodology for nodes to decide what level of data representation should be used will be developed. The decision-making process must consider factors such as: energy spent on communications and processing, current state of the network (available bandwidth, latency, errors), and current state of neighbors (i.e., energy).

People

Cintia Borges Margi
Katia Obraczka
Roberto Manduchi

Publications

Publications are available at the i-NRG publications page.