Amazon entered an increasingly crowded space with the launch of its AWS IoT platform, but through acquisitions, industry partnerships, and its inherent size and scale, the company is well positioned to become the cloud backend of choice for Internet of Things developers and enterprises. AWS IoT is a managed cloud platform that lets connected devices easily and securely interact with cloud applications and other devices. In this article, we have an actual case about a smart air conditioner system to see an excellent solution.

Challenges and Approach

In Smart Home domain, the question “How to build smart domestic electric appliances which allow users to control real-time via Internet and allow the recommendation of best utilization scenarios?” always poses many challenges about technologies, reliability and cost.

Proposed Solution

In order to illustrate the proposed solution, in this section, we present a case study about a smart air conditioner system. The architecture of the system is described in Figure 1.

Figure 1: Architecture of the system

The air conditioner system includes heaters, coolers, fans and air sensors (humidity and temperature) which are controlled by Wall Controller via Modbus protocol (turn on/off, increase/decrease temperature/humidity, increase/decrease fan speed, select program modes, etc.).

The IoT device communicates with Wall Controller via Modbus protocol using RS-485 transceiver and communicates with AWS via HTTPS protocol using WiFi connection (see Figure 2). It will receive information about the air conditioner system and send the information to AWS (via AWS IoT). After that, the information will be displayed on users’ mobile application (iOS, Android, Windows Phone). When users want to control the air conditioner system, they use their mobile application to send commands to AWS (turn on/off, increase/decrease temperature/humidity, increase/decrease fan speed, select program modes, etc.). These commands will be sent to the IoT device via AWS IoT, and the device will transfer them to Wall Controller via Modbus protocol to control the air conditioner system.

Information related to the system management is stored in Amazon RDS. The system will also store logs in Amazon S3 for Tracking, and store information about user behavior in Amazon DynamoDB for Analytics.

2-15-2017 10-45-07 AM
Figure 2. Components of the IoT Device

The technologies used in the system include:

  • Low-power ARM® Cortex®-M4 Core for MCU (microcontroller unit) of the IoT device.
  • Xamarin to build the mobile application for iOS, Android and Windows Phone.
  • AWS IoT, Amazon EC2, Amazon DynamoDB, Amazon RDS, Amazon S3.
  • AWS for system monitoring: Amazon CloudWatch, Amazon CloudTrail, Amazon SNS.
  • In addition to the foregoing AWS, we can also use Amazon Elasticsearch Service, Amazon Mobile Analytics, Amazon EMR, Amazon Redshift, Amazon Machine Learning, Amazon QuickSight, Amazon Kinesis for analysis of user behavior to recommend best utilization scenarios.

For example, the following data will be stored and analyzed:

  • Temperature, humidity, fan speed, program mode selected
  • Time to turn on/off, increase/decrease temperature/humidity/fan speed
  • Power consumption.
  • Thanks to this analytics, the system can recommend optimal settings for users as well as provide producers with useful information to improve their products.

Benefit and Conclusion

The approach helps us build appropriate solutions in Smart Home domain, which allow users to control real-time domestic appliances via Internet and allow the recommendation of best scenarios about economy, health, environment, comfortableness.

The utilization of AWS also assures the reliability, cost effectiveness and calculation power of the system. Thus the combination between AWS and IoT really brings a perfect solution for Smart Home.

Dang Kim Dung – FPT Software (Source: Potato – FPT Software)
(Published on the technology magazine Cuder World #3, released by FPT Software)

Related posts: