Overview

Acoustic ENSBox is an ad-hoc deployable wireless system designed to support distributed acoustic sensing applications. The Acoustic ENSBox is a multi-node system in which each node hosts an array of 4 microphones, enabling it to perform local analysis to compute bearing estimates to acoustic sources, as well as signal-enhancing techniques such as beamforming. The system also hosts special system software that enable automatic self-localization. The current version of the Acoustic ENSBox is not intended for long-term unattended deployments. The system will run for about 24 hours continuously on a single 12V 7.2AH gel cell. Longer term deployments could be achieved by duty-cycling the system.

Approach

The Acoustic ENSBox platform is a self-contained unit containing an ARM-based CPU module, a wireless network interface, a 4-channel acoustic sampling interface, and a battery. The system connects to a “head unit” that hosts an array of 4 microphones and 4 piezo tweeters in a Lucite and aluminum chassis. The microphones are condenser microphones with a custom pre-amplifier board. The photo at right shows an Acoustic ENSBox node deployed in the James Reserve.

In addition to the hardware, the Acoustic ENSBox includes a complete stack of system software designed to support distributed acoustic sensing. The system autonomously forms an ad-hoc wireless network that supports inter-node coordination, hosts routing services and reports diagnostics to a user with a laptop. It supports accurate time synchronized sampling, enabling application programmers to trivially compare time series data taken at the same time at two or more nodes. An acoustic localization system (described below) autonomously and accurately estimates relative position and orientation for all nodes in the system.

With this stack of system software, this platform is ideal for many types of collaborative sensing, especially target localization algorithms based on “beam-crossing”, where multiple states estimate bearing to a target and combine their estimates to compute a location.

Systems/Experiments

We performed several test deployments of the system to assess the system's performance. Controlled testing of the ranging and bearing estimation algorithms yielded average range errors of 3.5 cm, for the most part independent of distance. The bearing estimates were found to be accurate to within one degree. System tests were performed in the UCLA Court of Sciences and in the James Reserve. In these tests we placed 10 acoustic nodes in an area approximately 50m by 80m, with an average inter-node spacing of 15-20 meters. The ground truth locations of the nodes were measured using professional survey equipment. We found the self-localization to be accurate with an average 2-D position error of 4 cm in the Court of Sciences and 10 cm in the more complex and obstructed forest environment of the James Reserve.

Accomplishments

During the reporting period, the system was completely designed and assembled. In addition, the platform software development has been completed to the point of being a viable prototype. Several experiments have been undertaken to assess the capabilities of the system. This process is documented in detail in Lewis Girod's PhD thesis, completed December 2005.

Future Directions

We hope to see the Acoustic ENSBox platform taken up by several groups at UCLA who are involved in acoustic localization projects. We are currently working with Prof. Kung Yao's group to compare their bearing estimate algorithms to those developed for the position estimation application. We are also working with Vlad Trifa, a student from Prof. Charles Taylor's group who is developing software on the Acoustic ENSBox platform to detect acorn woodpecker calls. In addition, we are loaning several nodes to a Gaurav Sukhatme at USC, and some students who are using them in an embedded sensing class. There has been interest from other groups in replicating our system, although the current prototype system could use design improvements to make it easier to replicate.

People

• Lewis Girod (UCLA/CENS)
• Martin Lukac (UCLA/CENS)
• Jefferey Tseng (Aevena Corporation)
• Hanbiao Wang (UCLA/CENS)