Netted Sensors-based Vehicle Acoustic Classification at Tier 1 Nodes

The MITRE Corporation has embarked on a three-year internally-funded research program in netted sensors with applications to border monitoring, situational awareness in support of combat identification, and urban warfare. The first-year effort emphasized a border monitoring application for dismounted personnel and vehicle surveillance. This paper will focus primarily on the Tier 1 acoustic-based vehicle classification component. In a hierarchical network topology, distributed clusters of less capable nodes (Tier 1) at the lowest layer are in communication with more capable super nodes (Tier 2) at the next highest layer. Specifically, coarse information is aggregated at the lowest layer and communicated to the next layer in the network where more refined processing and information extraction can take place. We determined that fairly sophisticated classification processing can be performed at a Tier 1 node. This results in sizeable energy savings since data transfer rates are reduced across and between network layers. In general, acoustic-based vehicle classification is a difficult signal processing problem independent of the considerable hardware challenges. The acoustic waveforms are highly non-stationary and lack discernible harmonic structure, particularly for commercial vehicles that are designed to be intentionally quiet. In addition, the waveforms are rather complicated functions of the vehicle speed, engine RPM rate, and sensor-to-vehicle aspect angle. We discuss the development and implementation of a robust linear-weighted classifier on a Mica2 Crossbow mote using feature extraction algorithms specifically developed by MITRE for mote-based processing applications. These include a block floating point Fast Fourier Transform (FFT) algorithm and an 8-band proportional bandwidth filter bank. Results of in-field testing are compared and contrasted with theoretically-derived performance bounds.