Digital integrated circuit implementation of an identification stage for the detection of illegal hunting and logging

Abstract

Results of the VLSI implementation of an acoustic classification system's identification stage, intended for the detection of gunshots and chainsaws in a protected tropical area, are shown, with the idea of later building a surveillance wireless sensor network with similar nodes. The system performs from signal preprocessing to feature extraction, with results of the HDL description of the system tested on a FPGA against a golden reference, using real data taken from a protected rain forest area. Final classification of signals, using HMM, is in the final stages of testing. Some post-place-and-route results of the code ported to a commercial 130nm CMOS technology are also given.