by the webcam is viewed and the time corresponding to each
intrusion is first noted down. These intrusions are verified
against the log maintained by the server. Any intrusion that has
been missed by the PIR system will not have a corresponding
entry in the log. Verifying missed detection in this manner
would not be possible if the webcam and server are not
time-synchronized. The time synchronization is achieved by
using a simple UDP client-server utility.
4) A PIR-sensor ANimation-based simulAtion (PIRANA):
Supervised machine learning classifiers require training
data. Collecting data corresponding to animal intrusions is
time-consuming. PIRANA1 tool is developed that simulates
signals corresponding to human and animal intrusion and some
limited models of vegetative clutter [1]. Animated 3D models
are developed using a software called Blender. The simulation
tool was useful both for evaluating various competing designs
as well as for developing an intuition for the kind of signals
the STP would generate without the need for time-consuming
animal motion data collection. Initial testing of the algorithm,
which was later implemented on the mote, was performed on
the simulated signals