Posted by Hemanth Kumar Yamjala

With 5G communications technology standard in the offing, the growth of the Internet of Things (IoT) is going to increase. According to Gartner, the number of IoT devices is going to surpass 20 billion by 2026, which by any standards is simply humongous. By definition, IoT comprises a network of physical objects such as buildings, appliances, vehicles, devices, and other stuff, embedded with software, sensors, electronics, and network connectivity designed to store and exchange data using the internet. Some of these devices can be found in smart homes, household appliances such as refrigerators and air conditioners, and self-driving cars, among others. Since the behavior and performance of these IoT devices are largely dependent on the quality of the embedded software applications and communications networks, their functioning can be validated through IoT quality assurance.

Why IoT?

Enterprises are embracing IoT devices to get a raft of benefits such as improving efficiency and productivity, optimizing operations, reducing costs, improving customer experiences, and enhancing profitability. Their ready adoption is driven by a slew of factors:

• Easily availability of low-cost sensors

• Wide usage of smartphones

• Enhancement in bandwidth and processing power

• Availability of big data analytics tools

The significance of IoT lies in the fact that connected devices can represent themselves digitally and display ambient intelligence when acting in unison. Since the IoT QA testing is critical to ensure the functioning of such devices, there are some associated challenges as well.

Challenges of testing IoT applications

Given that the IoT devices integrate and communicate with a plethora of devices, platforms, operating systems, and networks, the testing team faces a lot of challenges as mentioned below:

Security threats: Cybercriminals are using more sophisticated methods to gain access into an IoT network to steal data and wreak havoc. For example, a huge DDoS attack had impacted giants such as Twitter, PayPal, and Spotify, among others by infecting thousands of IoT devices with malware. It is assumed that more than 70% of IoT devices are vulnerable to security issues. So, any IoT quality assurance process should focus on testing the IoT network’s password policy and ensuring the enforcement of such a policy. For example, among the recommended practices for IoT devices, there should be a change in the password upon first access.

A plethora of IoT platforms: IoT devices are primarily driven by software applications embedded within the hardware. These applications issue commands to the devices and analyze data collected by the latter. Since there are a plethora of software and hardware variants within devices working alongside different versions of operating systems and firmware, it is not possible to take a singular IoT testing approach for every possible combination of software and hardware. To go about testing IoT applications effectively, it is better to know the type of devices and software versions used by the end-users. Thereafter, the information should be analyzed to choose the most popular combinations for testing. To optimize testing for IoT devices the focus should be on testing the most used combinations while running sanity tests on lesser combinations.

IoT communications protocols: IoT devices use various communications protocols to interact with each other, and with controllers. These communications protocols have their pros and cons and come with names such as Presence Protocol (XMPP), Message Queuing Telemetry Transport (MQTT), and Constrained Application Protocol (CoAP). Developers can interact with the IoT device through an API based on JSON or XML that is layered on top of the transport protocol. QA specialists can use the communications protocols and APIs for testing IoT applications automatically. However, to ensure the effectiveness of such testing the automated tools should be compatible with these APIs and protocols.

Test environment: IoT devices with embedded software applications function in actual environments where there are many subcomponents and subsystems interrelated to each other. And any issue with one of the subcomponents or subsystems can impact the whole IoT ecosystem adversely. So, with an IoT testing framework, it is very difficult to replicate the actual operating environment of IoT devices.

Conclusion

IoT happens to be a critical part of the digital ecosystem the world is going to see and use in the not-so-distant future. The devices and the embedded software applications need to be checked for glitches or vulnerabilities to ensure their superior performance across a host of operating scenarios. So, any challenges that testers might come across while conducting IoT testing should be overcome with proper planning. The focus should be on validating the connected IoT applications and their support for data velocity, veracity, volume, and variety.