Subhadeep Bhattacharyya

SOFTWARE ENGINEER

INTRUSION DETECTION SYSTEM


Internet Of Things has led to the realization of ubiquitous computing and has enabled the practical existence of concepts such as Smart Home, Smart City, Smart Garage etc. Similar concepts of interweaving different objects to act as a single logical unit being connected to the internet can also be used to security systems which can detect intrusions and alert stakeholders as well as surroundings at real-time. This coursework group-project aims to implement an end-to-end IoT solution for an intrusion detection system.

As a part of this group project we used PIR (Passive Infra-Red Sensors) for detecting movements. PIR sensors have sensitivity range upto 20 feet(6 metres) and are used in automated doors,etc. Also since it is cheap and can easily be integrated with a Photon particle, it was our primary choice for a sensor.

We used a Particle Photon as the IoT hardware which gave us access to IoT device cloud and acted as the primary IoT platform. We further connected the motion sensor with the particle Photon and then used the particle programming to publish messages (0-2047 if no motion is detected and 2048-4095 if motion is detected) to a local Mosquitto MQTT topic. We also used a LED with the particle photon which would light up if motion was detected or otherwise remain off. The entire setup can be seen below.

Finally we designed a Java web application that would subscribe data from the Mosquitto topic and then based on the type of data generate alerts in a web page. To complete the end-to-end system design we also went ahead and implemented email features using SMTP to send emails to the user based on intrusion detection. The source code for the application can be found here.

Technologies Used:

  • Role: Developer
  • Event: Coursework
  • Location: Carnegie Mellon University, Pittsburgh
  • Year: Spring 2018

Project 01a
A deeper look into a PIR Sensor
Project 01a
Final Setup - PIR Sensor, Particle Photon, LED light, Resistors