Recipe for a home-made IoT surveillance system
Icon made by Freepik from www.flaticon.com

Recipe for a home-made IoT surveillance system

At the beginning of this year we started an IoT specialisation course on Coursera with Mika Majakorpi. From our uber-busy lives it stole a few evening hours and some weekends but it all ended with a huge sense of fulfillment for us as IoT enthusiasts. As an Electronics Engineer, for me this was a welcome opportunity to work once again with microcontrollers and sensors and be hands on with key IoT concepts.

In my job and sector, being on top of latest technology and emerging trends is a must for all leaders and in my opinion there is no better way than getting one's hands dirty on such concepts - that way one knows sufficient and concrete details, not just high level overview. Being hands on also helps to develop an agile, innovative and design mindset; one learns to quickly design and develop prototypes, figure what works, what could be improved and how to create big value out of ideas - this is hardly possible when working merely with theories.

As part of the Coursera specialisation, we worked on a Capstone project applying our learning of sensors and actuators to a real-life IoT application. We made a prototype for a home surveillance system using the Dragonboard 410C from Qualcomm. The prototype home surveillance system alerts the user of the humidity and temperature levels in the house and also functions as an intrusion detection system. We named the capstone prototype application "house bug"

Components

The key components of the home surveillance system are listed below.

These components were chosen based on their function, availability, inter-compatibility and price. The home surveillance system is powered by regular home electricity and as such there is no need for a battery. The power supply is rated at 2250mA, 12V and 27W. The load from the components is a maximum of 586mA, 2.9W on top of the DragonBoard's typical 4W when under load, so the power supply is sufficient. The components are standard 5V components. This prototype is not assembled in any housing.

The concept

Specifications

The prototype surveillance system has 2 modes: static monitoring mode and intrusion detection mode.

Static Monitoring

Static monitoring is always enabled and alerts the user when low or high thresholds are passed for either temperature or humidity. The data for these alerts is read from the DHT22/AM2302 sensor.

Intrusion Detection

Intrusion detection uses the HC-SR501 passive infrared sensor to detect motion. When motion is detected, a photo is taken with the Logitech Quickcam Orbit camera and an alert is sent with the photo attached.

Controlling the surveillance system

The user can use a browser over the internet (e.g. on their mobile phone or desktop) to control the threshold settings and configure other parameters like the email address for receiving alerts. The intrusion detection can be switched on or off as well.

Data display

The system has a local display implemented using two MAX7219 LED dot matrix displays. The display shows the humidity and temperature, switching between the two values every 5 seconds. This is so that the sensor information can be simply seen inside the home even when thresholds are not passed.

Operating System, Software and User interface

The Dragonboard 410C runs the Debian operating system. The software to interface with the sensors and the Internet facing HTTP server is implemented using Python.

The user interface for controlling the application is accessible with a browser. This way one interface can be used on different mobile device platforms without the need for native applications.The alerts are sent as emails which is also platform independent.

Functionality Overview video

 

Key references for building the circuit 

HC-SR501 Specification

DHT22/AM2302 Specification

MAX 7219 Specification

Link to the code on github

https://github.com/mmajis/iot_capstone

要查看或添加评论,请登录

社区洞察

其他会员也浏览了