Jovie Sylvia

IoT Automation — How Feasible It Is?

Blog Post created by Jovie Sylvia on Aug 28, 2017

Abstract

IoT (Internet of Things) is a rapidly developing market, where the opportunities are huge on an end-to-end basis. The number of devices in the IoT spectrum is rapidly increasing with every passing day. The opportunities are also huge as IoT, as a technology in itself and its application across various sectors, is gaining prominence quickly. Testing IoT brings in a lot of opportunities as well as challenges.

This blog focuses on and provides a brief idea about IoT, its trends, opportunities, and challenges for automated testing in IoT Domain.

Background — What is IOT?

The concept of IoT was introduced and documented in early 1999 by Kevin Ashton. IoT is a network of physical devices or things that are embedded with sensors, actuators, or embedded with a programmed software, building connectivity for exchanging data with the manufacturer, operator and/or other connected devices. These devices communicate to the internet in the dual-enabled mode to transmit data to backend systems as and when required.

Gartner says that by 2020, there will be over 26 billion IoT connected devices; another research (IDC, ABI, and Cisco) predicts that the estimate may go over a 100 billion. IoT is a huge network of connected ‘things’ which includes devices and well as people. The relationship will be between human to human, human to objects, and objects to objects.

Key Components of IoT

Things

Embedded objects residing within products, which collect data from sensors and actuators and send them to the backend systems.

Network Communications

Communication components form the bridge between things and back-end analytics systems. Internet-enabled things communicate using different types of network protocol (4G/3G, LTE, Wi-Fi, NFC, Bluetooth) through wired or wireless medium.

Computing Infrastructure

This component receives data from different things and applications and provides different business views of KPIs and performance parameters with the help of analytical tools, thus helping the consumer/enterprise take business-critical decisions.

Solution: Application of IoT Automation

Combining Hardware and Software

Devices, sensors, and applications from the core of IoT. However, there is a phase switch from the absolute testing of applications in a specific environment, to testing the experience in a dynamic environment. Simple functionality validation is not enough in an environment for the complicated IoT.

Many working systems

To deploy an IoT system or service, it is not enough to only ensure a mere working set of device software, multiple back-end servers. Sensors and algorithms must also be installed for the same. For instance, a marine tracking system with sensors and devices that communicate with multiple software in back-end will require intelligent algorithms to track the accurate location of the ship. This entire system requires a robust QA validation process. The user cases are extremely complex in real time and thus become a dispute for QA.

Signaling Instruments

The creation of a working environment to test a real-time IoT implementation is complex. It means not only configuring an environment but also includes scalability and reliability. The system is built on numerous analytics engines and requires more hands-on experts in simulation to build out a test environment. While the hardware is well tested by device makers, understanding application intelligence and the device’s working in real-time is essential for the creator of every module that is involved in the IoT process.

Device Middle layer interaction

This layer proves to be the connectivity between the real –time software and the hardware components of IoT. Example: A Bluetooth device transmitting real-time data to a mobile device app.

However, other types of testing could also be necessary.

User Interfaces and Firmware interfaces

Here is the user interface with which a computer or a software interacts. It determines how inputs are given to the computer or the program and how information is displayed on the screen. Three major types of user interfaces are Machine language. The user must know

  1. All the necessary and valid machine instructions or codes.
  2. Menus: User chooses the commands from lists displayed on the screen.
  3. GUI: user gives commands by selecting and clicking on icons.
Testing Architecture: End to End IoT Testing

Functional Testing

To check the customer requirement, based on specific inputs given to the IoT app.

Compatibility Testing

Varieties of verifying and validating the possible combination of device versions, communication protocol versions, mobile devices, resolutions, and mobile OS version.

Usability Testing

Verify the ease of using experience of IoT app, with respect to its usage, visibility of text, and usefulness of the content to the end user

Network Testing

Validate the IoT app with all available network connections and ensure app to sync with all different backend combination protocols

Security Testing

Verifying the privacy of data, the reliability of IoT app, Also verifies if the IoT app uses any weak password or missing data encryption. Checking the application followed network security standards and authentication.

Performance Testing

To ensure the overall scalability and performance of the IoT app, test the response time based on data loads, optimize the code to improve the performance, and also like low battery, less memory, switch between different networks.

Implementation — Simple Framework outline

Protocol simulators

These are communications medium that transfer messages. Protocol simulators are majority used in changing the system settings frequently. Protocol simulators can be used to test a huge variety of device endpoints and interfaces to validate.

Data recorders

Data recorders from a variety of devices can be helpful in smart validation across device sets. The recorded data can be used across different device endpoints and ensure the compatibility testing of apps across different device sets and communication layers.

Virtualization:

This is very complicated and an important aspect of IoT validation. Due to highly complex IoT interfaces and the necessity of validating everything in real time, virtualization implementation will be helpful into the services on which IoT applications are built. Virtualization of an IoT reduces the costs due to minimal dependency on the real time environment for testing.

Summary
I here conclude that IoT automation is a possible but complex and game changing concept. It will play an essential role in our smart cities. IoT testing scope always comes to a tester with a lot of challenges that they have to deal with while testing. The testing team has to focus on core components of the IoT system to treat functional and connectivity testing as significant elements of overall IoT testing. As newer devices will start to connect the need, will enforce the technical folks to come up with most advanced concepts and keeps experimenting the new stuff.

Outcomes