Top 3 Internet of Things Trends Behind the Scenes at CES
Shawn Wasserman posted on January 06, 2016 | | 11260 views
With the Consumer Electronics Show (CES) in full swing, various outlets have been reporting on IoT trends to try and predict the types of gadgets we will soon see on the store shelves. However, engineers designing these gadgets will need to know what Industrial IoT (IIOT) trends will power these CES gadgets for years to come.

One of the interesting sources of these IIoT trends comes from National Instrument’s (NI) Trend Watch 2016. NI provides much of the hardware and software that keeps the IoT running.

According to NI, engineers looking to design and build the gadgets of tomorrow should pay attention to these three industry trends:

  • IoT standards will grow out of industrial system integrations (machines, power grids and transportation)
  • Growth of low-cost big data will force intelligent sensors to perform data management and analytics
  • Vast growth of IoT devices will force IoT device testing to look at the architecture, not the device

“As the world becomes more connected, the incredible amount of real-world data available today promises engineers and scientists great insight, but getting that insight can be a challenge,” said Eric Starkloff, NI executive vice president of global sales and marketing. “We strive to help our customers stay ahead of the deluge of analog data by investigating the trends that will help them handle the increasingly complex amount of information they’re acquiring to make the right decisions at the right time.”

Evolution of IoT Standards Fueled by Industry, Not Consumers

Although there will without a doubt be many options for IoT standards presented at CES, Ni predicts that standardization may well come not from the consumer electronic world but rather from the IIoT domain. Major corporations, industries and utilities look to save a considerable amount of money using IoT, big data and intercommunication of devices. The utilities, manufacturing and transportation industries will be pushing the IoT industry to consolidate much faster than the end consumer with a Fitbit and a smart fridge.

Stephen Lawson of PCWorld agrees with NI that the answers to IoT standard consolidations will not be seen at this year’s CES. “When Internet of Things devices debut at this year's CES, one of the biggest questions will be how they'll connect to all the other smart-home gear on display,” he reported. “The fact is, it's too early to say what standard or protocol will become the glue that can turn a pile of cool gadgets into a system that runs your whole house for you. New systems are just starting to emerge and though they may eventually work with each other and with older platforms, buying one of each and expecting harmony is still wishful thinking.”

NI predicts IoT standardization will come from the industry, not consumer electronics. (Image courtesy of National Instruments.)
NI predicts IoT standardization will come from the industry, not consumer electronics. (Image courtesy of National Instruments.)
NI’s report outlined the many advantages IoT can bring to industry. It reported that “once networked together, [IoT can] create a smart system of systems that shares data between devices, across the enterprise and in the cloud. These systems will generate incredible amounts of data […]. This Big Analog Data will be analyzed and processed to drive informed business decisions that will ultimately improve safety, uptime and operational efficiency.”

Unfortunately, these advantages all break down once IoT-enabled devices from different vendors refuse to cooperate with each other or when the network that connects these devices is unable to handle the increased data flow.

NI explained that traditional networked infrastructure works on the control hierarchy of rigid bus layers optimized for certain tasks, each layer with a differing latency, bandwidth and quality of service. Unfortunately, this isn’t optimal for the timely data collection needed for real IoT interoperability between systems.

One potential solution to this standards issue is Time-Sensitive Networking (TSN). If standardized across the IoT industry, TSN may be one standard that will ensure that devices will be able to communicate and be able to address the data transfer and security needs of the IIoT.

The creation of the TSN standard and ecosystem is headed by the AVnu Alliance. This alliance consists of Broadcom, Cisco, Intel and NI. The aim will be for AVnu to certify devices similar to the certification process of IEEE 802.11 standards by the Wi-Fi Alliance.

IoT Device Certification Moving to Architecture-Centric Testing

IoT industry could be worth $11.1 trillion a year by 2025 (Image courtesy of McKinsey & Company.)

IoT industry could be worth $11.1 trillion a year by 2025 (Image courtesy of McKinsey & Company.)

The IoT industry is looking to become a $11-trillion-dollar industry by 2025, according to the consulting firm McKinsey. No standardized test procedure will be able to cope with the resulting proliferation of disparate devices.

In response, NI suggests that the standardization of device testing take on an architecture-centric perspective.

At the end of the day, all IoT devices will have three main components: a power source, wireless communication and a sensor. As a result, the testing procedures should be designed to adapt to these three components despite their different formats and how they might change in the future.

For instance, look at this year’s CES. Timothy Torres from PC Mag reported, “Bluetooth LE, Lutron ClearConnect, Wi-Fi, Z-Wave and ZigBee are all home automation and connectivity protocols and they'll all rear their heads at CES 2016. But the one to look out for may be something called Bluetooth mesh networking, a new standard that will extend the distance for Bluetooth technology.”

Using a device-centric testing point of view, each of these protocols will need a different standardization procedure—which makes for a very messy set of instructions. And this doesn’t even include the hundreds of sensors and power sources an IoT device might have.

To combat the disparate nature of IoT devices, NI suggests that we learn a lesson from the smartphone industry. Why not design multiuse testing hardware with varying software and low-budget hardware upgrades?

For example, you could choose to use a smartphone’s built-in speaker or buy another one for more power. What if your IoT testing hardware could contain a standard set of testing devices and have the ability to connect to upgradable testing platforms?

If your device uses a thermocouple, Bluetooth and a AAA battery, you buy the additional hardware you might need and connect it to the main device. At that point the testing procedure becomes only a matter of numerous software apps that change depending on what testing devices have been connected. Press start on the testing device and let the software do all the work for you.

Cheaper Big Data Will Grow the Need of Analytics and Management

As the costs and size of sensors and processors drop, NI reports that more designs will implement data management and analytics at the sensor level. NI reports that current examples of such designs can be seen from ARM, Intel and Xilinx. This trend can allow engineers to reduce the bandwidth and data storage needs of the IoT and big data.

Intelligent sensors and data management will significantly reduce the storage and processing needs for big data. (Image courtesy of National Instruments.)
Intelligent sensors and data management will significantly reduce the storage and processing needs for big data. (Image courtesy of National Instruments.)
Traditionally, log systems would keep track of every data point, even when the status of the sensor hasn’t changed. This can create a significantly larger data set. When data analytics is pushed to the sensor, however, many more devices can be managed as not all the data collected by the device will need to be sent to the central hub.

However, this method of data collection does offer one challenge: the consolidation of metadata. Without proper documentation and metadata collection units, testing parameters and testing outcomes can be lost. Once this metadata is lost, managing and processing the big data can become impossible.

NI reports by automating its metadata collection, Jaguar Land Rover was able to go from analyzing 10 percent of its data to 95 percent. Imagine the operational, development and maintenance cost savings potential of so much data analysis.

The IoT can offer a lot for both the consumers at CES and the engineer in a factory. However, it hasn’t yet made it to silver bullet status for solving our industry and household problems. But one thing is for sure: necessity is the mother of invention and the greatest IoT necessity lies in industry.