Processing, Protocols, and Protection: A Closer Look at IoT and IIoTJanuary 20, 2020 by Lisa Boneta
While IoT and IIoT are based on similar hardware, their specific applications warrant different data processing, protocols, and security considerations.
The Internet of Things has taken the connectivity of devices to a whole new level. According to a Statista article by Shanhong Liu, the IoT market reached 100 billion dollars in revenue in 2017 and is expected to grow to 1.6 trillion dollars by 2025.
A study from Juniper Research revealed that there will be over 38 billion IoT devices in 2020—a 285% increase from 2015.
IoT is used in various applications and markets, from autonomous vehicles to smart home devices.
Another important role of IoT is on the manufacturing floor, where it's better known as Industrial IoT (IIoT). IIoT completely transforms traditional supply chains into more efficient digital supply networks (DSN) by using sensors that enable features such as predictive maintenance and location tracking.
While IoT and IIoT are similar and related, there are key differences in deployment that separate the two.
Processing Data: In the Cloud or at the Edge?
IoT data tends to gets processed either “in the cloud” or “at the edge.” The cloud generally refers to a server whereas the edge refers to a logical layer rather than a physical location. That is, the edge can be user-defined, but generally refers to the physical device collecting data itself.
It is important to note that the edge and cloud work together in IoT systems, but serve different purposes and have different advantages.
Smart home devices such as Amazon’s Alexa offload a majority of tasks to the cloud because of the system's complexity. Companies are releasing products that can integrate Amazon Alexa with commonplace hardware, such as Toshiba’s latest Smart Microwave debuted at CES 2020.
Toshiba’s new smart microwave is the latest IoT smart home device from CES 2020 Image used courtesy of Midea Microwave & Cleaning Appliance Division
Such a task becomes even more complex when using a voice-activated system to control the device itself, such as starting, pausing, or specifying cooking times.
Immediate response time in a device like a smart microwave isn’t as consequential, however, as response time in an IIoT device. IIoT devices deal with time-sensitive and critical data that often require immediate processing and computing with low-latency.
For instance, ZeroKey recently showcased its latest real-time 3D location IIoT sensor at CES 2020. This sensor provides real-time location of assets, personnel, and equipment across large environments in manufacturing settings.
ZeroKey's IIot sensor technology. Image used courtesy of ZeroKey
Any points of failure or maintenance should be detected in real-time to prevent any backlog or fatal errors within the processes themselves, which is one of the main selling points of ZeroKey’s new sensor.
While ZeroKey has not provided specific details of how data is processed, it seems that the device does promise real-time results with edge computing delivering the power and speed necessary.
Scalability and Protocol of Devices
At home, we can have many IoT devices connected using Wi-Fi or Bluetooth. As we move toward more digital and smart home devices, scaling can be an issue to ensure sufficient bandwidth, connectivity range, and compatibility to support all devices.
In a previous article, we discussed how Google, Amazon, Apple, and other tech giants are teaming up to bring consumers more user-friendly and device-compatible connectivity solutions under project Connected Home over IP.
The working group believes that “that smart home devices should be secure, reliable, and seamless to use” and will accomplish this by taking an open-source approach for the development of a new, unified connectivity protocol that is built upon Internet Protocol (IP).
To encourage scalability in IIoT, the presence of a strong data network is vital. While smart home devices tend to be wireless, there is an unavoidable latency. Thus, engineers and developers are looking toward Ethernet to help curb this undesired effect.
Industrial Ethernet switches are another important component to address connectivity as they must withstand extreme vibrations, temperatures, and shocks.
Different real-time requirements of Ethernet may lead to different implementations. Image used courtesy of Mouser
Time-sensitive Networking (TSN) is used to develop a “time-oriented” style of traffic scheduling, which makes Ethernet a true-industrial grade, real-time communications protocol.
It is still up to the developers to make sure the system supports scalability in large, heterogeneous networks and the multiple protocols often used in legacy equipment.
Hardware must be rugged to withstand any extreme conditions as to not lead to failures in connectivity.
According to a (PDF) study of the smart home environment and associated security threats, there are several different security threats that can occur: eavesdropping, masquerading, replay attacks, message modification, denial of service (DOS), and malicious codes.
The most common of the threats is eavesdropping, which is when an attacker monitors data traffic without authorized users knowing.
Examples of security attacks that can occur in a smart home environment. Image used courtesy of (PDF) Shafiq Ul Rehman and Selvakumar Manickam
Sensitive and private data could potentially be transmitted, such as passwords and private information. To help keep a smart home secure, authentication and authorization are needed, so intruders are denied access into the network.
Security challenges for IIoT can be similar but lead to different consequences. Lack of authentication and security in process sensors, protocols, gateways, and data is a major concern because any compromised data can lead to equipment damage, regulatory issues, and personal safety hazards.
IIoT security must be multi-layered and protect the data plane, management (network and element), and control (protocol) planes.
Have you worked with the hardware driving IoT or IIoT? What design distinctions have you noticed between the two? Share your experiences in the comments below.