Combining new technology with rapidly-developed software on newly-created hardware—where critical bugs concerning reliability, safety and performance are often overlooked—makes testing of Industrial Internet of Things (IIoT) devices difficult. However, advancements in technologies are making industries more effective and enabling machines to work smartly
The time is not far when the IoT will become essential for testers to survive in the development world. The IoT-enabled gadgets, smart device applications and communication modules play a vital role in evaluating the performance of various IoT services. As IoT technology evolves, there is a need for testing at different levels, namely, chipset, module and end host device. The IIoT monitors, collects, exchanges and analyses information to enable machines to change their own behaviour, or instruct other devices to do so, without human intervention.
IIoT testing depends on the sensitivity of devices (as specifications change from vendor to vendor), reliability of network and so on. Users’ security and performance concerns can be dealt with successfully by vendors, encouraging the use of such devices around the globe.
Need for IIoT test and measurement
IIoT testing is based on the system involved. Testers should concentrate on test-as-a-user approach rather than testing based on requirements. It is necessary to certify a complicated mesh of devices, protocols, hardware, operation systems, firmware and so on. An end-to-end quality assurance strategy covering a diverse set of embedded devices, Web applications, testing types and methodologies, test environment setups, use of tools and simulators, and more is needed.
The IIoT incorporates machine learning and Big Data to harness sensor data. It holds great potential in manufacturing for quality control, supply chain traceability and overall supply chain efficiency. Since manufacturing companies have more complex and dynamic processes, availability of real-time data on critical parameters becomes more important.
Role of sensors in the IIoT
Sensors help in predictive maintenance, asset monitoring and data analytics to achieve sophisticated machine-to-machine (M2M) communication. Automation in processes like control, safety, operations management and asset optimisation increase the demand for sensors for measurement and analytics. Smoke, proximity, IR, image, temperature, piezo and optical sensors are the different types of IoT sensors required for an industrial automation system.
Networked data loggers combined with cloud solutions enable test and measurement specialists to provide a complete digitisation solution for bringing systems, machines, vehicles and more to the IoT. Long-term measurement and monitoring are required to obtain measurement data under real operating conditions. This makes it possible for engineers to have a completely-isolated personal testbed at their desk. Different tests can be performed in parallel by multiple engineers and results saved in a common database.
Test approaches for the IIoT
The primary focus of testing during industrial automation is ease of use, ease of learning or familiarising with the system, and satisfaction of the user with the entire experience. To ensure that scalability, performance and security is up to the mark, following types of tests are recommended:
The tracking device used should be portable. Usability in terms of processing and displaying data should be tested thoroughly.
It should be smart enough to push not only notifications but also error messages, warnings and so on.
It should have the option to log all events, or store the same to a database, to provide clarity to end users.
Notifications and handling of the display should be done properly in the devices.
All connected devices operate based on available data. There is always a chance that data can be accessed or read while transferring between devices. Hence, it is important that the user interface is password protected.
Before communication starts, the hub and devices should be properly authenticated. Sent data should be in encrypted form.
Success of an IoT system depends on how well the devices and hub are connected. Loss of connection can lead to inaccurate data.
Every authenticated device in the range should be able to connect to the hub when the connection is up and running. It should be able to send any required amount of data to the hub.
In case of a down connection, if the system is unavailable on the network, timely alerts should be sent.
It is important to test the monitoring utility used to display system usage, power usage, temperature and more. IoT devices in sleep mode draw nanoamps of current but may draw tens of milliamps while processing or communicating. To deliver power-efficient IoT devices to large utility customers on time and with real-world battery life performance, following conditions must be checked:
- When a device is located at cell edge and has intermittent connectivity, carrier aggregation, higher-order MIMO and other cellular-IoT capabilities’ impact on battery lifespan must be calculated.
- Power status update should be sent to the network if the device goes low in power.
Reliability and scalability
Testing in a lab does not ensure that the system will work when exposed to real-time conditions. Hence, a test environment must involve a simulation of sensors, for reliability and scalability, by leveraging virtualisation tools and technologies.
Other than testing in real-life scenarios, there are a few common test cases that must be considered while testing IoT devices and the network.
It is a good practice to get regulatory requirements approved at the start of the development cycle, to make sure that the products are certified.
When an upgrade is performed, be it for the system or for any involved item, thorough regression testing should be carried out to overcome upgrade-related issues.
Signal and power integrity. The IoT is successful if the integration test plan is accurate and robust enough to catch flaws in the system. Devices need carefully-controlled signal and power integrity.
There is a large number of devices available for particular applications of sensors, protocols and platforms. As a result, a test matrix is formed to achieve the right compatibility testing.
Communication protocol and device interoperability. This assesses the ability to interoperate protocols and devices across different standards and specifications.
Prerequisites for testing
The prerequisites that need to be taken care of are:
Setting up the IoT hub
An IoT hub is a server that can talk to IoT devices and collect data from them. It may be a Web server on cloud, or an app on a mobile device.
Setting up the IoT device
It is essential to make sure that the IoT device is turned on, accessible and creates a genuine use case for testing.
Setting up the network
The device(s) and hub can be connected via Wi-Fi, Bluetooth, satellite signals, NFC, cellular, Zigbee or LoRa. Each technology has its pros and cons. It can be a challenge to decide which connectivity protocol to use and then to figure out how to test to make sure the rollout goes smoothly.
Design and simulation tools
There are various tools used during testing IoT systems. Greater integration delivers lower cost, lower energy consumption and better performance. It is important to know how to choose a solution that can solve the challenges, and to know the best measurement tips and practices that need to be followed.
An open source application called Wireshark can be used for monitoring traffic in the interface, source/destination host addresses and more. It helps users in displaying TCP/IP and other packets that are transmitted or received over a network, as in Tcpdump.
This helps in debugging the target platform code and shows variable step-by-step procedure, for example, JTAG dongle. It can be used to check various events with time stamps, glitches in power supply, signal integrity using a digital storage oscilloscope and so on.
A software-defined radio is used to emulate the receiver and transmitter for a large range of wireless gateways.
Architecture for the IoT comprises multiple layers built on top of each other. To create industry-specific solutions, components in each layer need to work in sync, to effectively convert data into information.
Latest IIoT testing and monitoring devices
Some testing and monitoring devices for the IIoT are:
- imc’s CRONOScompact provides adaptable measurement and control system for mixed-signal testing.
- Analog Devices’ AD8233 electrocardiogram integrated AFE has microamp-range power requirements that enable extended battery life and continuous monitoring.
- Texas Instruments’s industrial monitor with Sitara processor addresses the increasing demand for industrial communication connectivity and processor performance with superior video processing and graphic performance.
Challenges for IoT testers
Common challenges that IoT testers face are calibration of sensors that can vary from device to device, setting up the environment, connectivity, power, security and privacy, hardware quality and accuracy, among others.
The IoT is an architecture comprising hardware, software, sensors and communication gateways. And testing in terms of environment, data transfer and the like is tedious.
Device interaction module. When hardware and software integrate with each other, things such as security, backward compatibility and upgrade issues become a challenge for the testing team, as related sub-systems and components can be owned by third-party units.
Testing devices for every platform—iOS, Android, Windows and Linux, among others—is a challenge.
An IoT architecture must be tested for all kinds of network connectivity speeds. For this, virtual network simulators are used to vary network load, connectivity, stability and so on. Real-time networks can be challenging in the long run.
The IoT has a dynamic environment with millions of sensors and devices in conjunction with intelligent software. A complex set of use cases must be created to test cases and data.
IoT applications can have multiple real-time scenarios, and use cases are extremely complex. Getting regulatory check points or the system deployed is tough.
Anisha Nikash Dumbre, research analyst, Frost & Sullivan, says, “By 2020, a number of IoT-enabled test systems are expected to be launched, catering to diverse services and fields within the industrial ecosystem. Democratisation of the IIoT will lead to cheaper software-as-a-service (SaaS) models.”
Combining new technology with rapidly-developed software on newly-created hardware—where critical bugs concerning reliability, safety and performance are often overlooked—makes testing of IIoT devices difficult. Emergence of smart testing for the IIoT will have a significant impact on the test and measurement space. Advancements in technologies are making industries more effective and enabling machines to work smartly. Predictive analysis will make test systems intelligent, and their use will maximise digital connectivity.