5G has huge disruptive potential when it comes to serving the fundamental network needs of manufacturing operations. The industry better known as Industry 4.0 is now being digitally transformed by the ever evolving Industrial Internet of Things (IIoT).
Since 5G uses shorter (millimeter waves) frequencies, which are between 30GHz and 300GHz and of which mobile networks will use a tiny portion, it has far superior capabilities. Compared to 4G, 5G offers lightning-fast data speeds which are up to 100 times faster, extremely low latency (1 millisecond) and a much greater number of connected devices per unit area. However, to fully leverage the benefits 5G has to offer, a new set of networking infrastructure is required, including mesh networks, phased array antenna and cloud computing.
Intelligent interconnectivity of all digital devices used for factory floor automation and supply chain management, starting from measuring devices, programmable logic controllers (PLCs) and their industrial applications and all the way to smart logistics, can be called IIoT. This interconnectivity is supported by modern-day technologies including edge computing, cloud computing, machine to machine (M2M), 3D printing, advanced robotics, big data, RFID technology and cognitive computing. Currently a multitude of companies involved with technological innovation is spearheading this fourth industrial revolution.
There are several requirements for successful delivery of an IIoT solution. First and foremost, a tiny Software Defined Radio (SDR) or RFID chip is needed. which must be of low cost and should require very low power from the sensor batteries since most of their applications will range from deployment in agriculture, to factory shop floor, to hard to reach places such as oil and gas fields, to auto manufacturing and logistics operations. This enables seamless wireless communication under the unlicensed industrial band, which is typically at 900 MHz.
IIoT has four primary levels, presented in Figure 1: Devices, Edge computing, the Network (5G/Internet) and the Cloud.
Figure 1: There are four primary levels in IIoT.
Capabilities and Requirements of IIoT
As the industry rapidly moves towards IIoT solutions, the need for flexible resources with optimized costs becomes a necessity. Cloud computing offers shared computational resources along with centralized management and monitoring, it provides computing services through network and servers, central storage, advanced analytics and business intelligence. It capitalizes on these to process big data, perform machine learning and utilize data warehousing for business logic and operation management. Conversely, edge computing relies on processing data right where it is being generated through decentralized computing infrastructure placed at the network’s “edge”. Faster networks like 5G enable edge computing to process large chunks of data in real time and perform tasks such as optimization right at the source.
Such capabilities can only be harnessed through the support of a robust network, able to communicate with a massive number of devices. To help achieve this, a number of different network protocols are used to stream data among network nodes, including:
- Message Queue Telemetry Transport (MQTT)
- Advanced Message Queuing Protocol (AMQP)
- Constrained Application Protocol (CoAP)
- Modbus and OMF
This is where the key characteristics of 5G networks bring it together:
- Enhanced ultra-reliable low-latency communication (eURLLC)
- Massive machine type communications (mMTC)
- Enhanced mobile broadband (eMBB)
The table in Figure 2 summarizes some key features and use cases.
Figure 2: Different 5G network protocols are used for different purposes.
Finally, there is a need for smart devices such as sensors, actuators, trackers and PLCs, all of which may be able to connect to the network and generate data.
IIoT promises some benefits in one of the most critical areas of any industrial environment:. human and process safety. With enhanced surveillance from all the connected IIoT sensors sharing key safety parameters among themselves, probability of an accident is significantly reduced. Wearables can play an even bigger role in the event of catastrophe by pinpointing the exact location of employees in trouble and allowing for the protection of their well-being.
Security is another area where IIoT is beneficial, whether it is cybersecurity or the physical surveillance of manufacturing premises. This also includes the elimination of human error which is the most common cause of data and cyber breach.
Since IIoT provides manufacturers the ability to automate the processes extensively, improved operating efficiency is another big area which boosts productivity. This is achieved by optimizing costs related to maintenance and labor by making data driven decisions based on intelligent insights of operations provided by all connected manufacturing functions.
IIoT also helps with reliability of operations as it brings standardization in all processes by accommodating all specific industrial applications from different OEMs and vendors.
SDRs for Industrial Internet of Things
A software defined radio or SDR can be simply defined as a radio communication system where hardware components such as modulators/demodulators, filters and mixers are implemented via software in order to harness untapped potential of digital electronics. A typical high performance SDR’s front end enables tuning of wide range of frequencies through its receive (Rx) and transmit (Tx) functions. SDRs also contain multiple independent Tx and Rx channels with dedicated 16 bit DACs/ADCs for high sensitivity and Signal-to-Noise Ratio (SNR)/Spurious- Free Dynamic range (SFDR).
Under the hood, it comprises of an FPGA with digital signal processing capabilities to perform various functions such as modulation/demodulation, upconverting/downconverting and encoding/decoding, etc. These functionalities are easily configurable and upgradable to support the latest radio protocols along with different DSP algorithms.
Figure 3 shows Per Vices’ Crimson TNG, which is a high performance SDR for development and testing of new IIoT applications relying on 5G and/or other wireless networks.
Figure 3: Crimson TNG is an example of a high performance SDR.
SDRs will be playing a key role as a crucial building block for a 5G network infrastructure, which in itself is one of the prerequisites of IIoT. They will help manufacturing facilities to overcome the problem created by an extended wired network, required for a massive number of smart devices at the shop floor, and replacing it with a faster wireless network. This will help them save capital costs for installation and later operational cost for maintenance. Moreover, the flexibility of a configurable SDR adds further advantage.
Compared to the typical Wi-Fi or 4G network’s latency, which makes them unusable for mission critical time and time sensitive IIoT applications, use of low latency SDRs can help achieve massive data transfer rates for a combination of network protocols. With the further evolution of IIoT, SDRs will also help include the legacy systems in future networks. Moreover, they also offer numerous other advantages over traditional solutions, such as protocol flexibility, channelization, lower costs, lower power usage and future proofing.
SDRs’ use of smart technology along with ruggedized design with custom form factor makes them ideal for their use in industrial setup, while their frequency hopping capability eliminates the problem of interference in IIoT applications. Another challenge which SDRs help overcome is the connectivity of heterogeneous devices being used in the IIoT along with their use of different frequency bands and protocols.
Using SDRs for testing new IIoT applications
SDRs with their programmability make it easier for identification and development of new use cases, support development of testbeds, and help develop best practices in order to provide guidance for the application of over the air and related communication. They also enable digital transformation technologies in support of the IIoT use cases.
Using SDR is particularly useful in tackling the rampant challenge of interference due to low powered devices in a dense network where these devices can always be transmitting data independently. Moreover, they are used in development of data security and encryption methods in communication such as device-to-device (D2D). This flexibility of SDRs make them an ideal product for design and development of a proof of concept for different IIoT applications and network topologies.
The customization and configurability of the on-board FPGA makes it an excellent device for testing new frameworks and protocols. It also supports the wide range of evolving IIoT gateway requirements through flexible protocol switching/bridging and also allows for remote in-field upgrades. ‘Edge computing’ is also possible through FPGAs, with their higher processing power, to perform analytics while also allowing for time sensitive networking (TSN).
Figure 4 shows typical architecture of a 5G based smart manufacturing facility.
Figure 4: A diagram of the architecture of 5G-based IIoT for Industry 4.0 is displayed.
IIoT is increasingly creating ripples in almost every industry, while many industries have already found some great uses for this technology. The aviation and aerospace industry in particular is taking great advantage, with the use of artificial intelligence and machine learning models, to understand the multiple thousand steps which operators follow during the assembly of an aircraft. These models then help automate and simplify operator tasks while ensuring their correctness, resulting in greater production efficiency.
Similarly, the auto industry, including big names like BMW, is tapping into this potential by transforming their production sites, working in key areas such as:
- Smart data analytics
- Smart logistics
- Innovative Automation
- Additive Manufacturing
Other use cases for factories, like creation of digital twins for planning and space optimization, and AR/VR solutions for remote maintenance and commissioning, are being used by manufacturers to improve their processes and adapt new ways of working. Efficient energy management, especially for solar panels, is another area where costs are being saved by real time energy monitoring and advanced analytics.
IIoT for the Logistics industry
IIoT is doing wonders for the logistics industry, from automated deliveries via autonomous vehicles and UAVs to optimization of fleet management with real time tracking and monitoring. Warehousing has been made ‘smart’ with the implementation of IIoT based integrated management system where smart sensors have taken control of inventory management. AGVs and automated forklifts have taken over the production facilities and loading decks all the while providing accurate inventory levels and generating alerts, for things like low inventory levels, that will allow timely intervention from management.
There is also huge potential for IIoT in the packaging industry which can optimize the throughput based on advanced analytics and machine learning algorithm to automate packaging of inventory for shipping based on the product it is dependent upon. Similarly, the distribution side of the supply chain can also benefit via automated labeling and sorting of products. Goods can be collected in customer order fulfillment warehouses and delivery routes can be optimized based on real time data along with other last mile delivery innovations.
Most of the e-commerce and logistics giants, including Amazon, UPS, FedEx and Walmart, have already adopted and are actively pursuing future IIoT solutions, while others too are aggressively embarking on the journey for such digital transformation for their operations.
Further benefits of IIoT include the following:
Value addition — Data gathered from the IIoT assets can help find out the actual process efficiency of processing lines and thus can be improved significantly. Data can also be fed to 3D printing technologies which produce parts more flexibly and efficiently while using far less raw materials.
Quality Control — Proactive quality management via IIoT enables the early detection of defects and quality issues than conventional quality control methods.
Predictive Maintenance — Tiny sensors mounted on critical process equipment can measure and transmit abnormal temperature, speed and vibrations to accurately predict or alert of failure; thus saving equipment and reducing unplanned downtime.
Employee health — Repetitive tasks can be eliminated with help of machine learning and process automation which helps employees spend time on solving more complex problems and relieving people of ailments such as repetitive strain injury.
Flexibility — Products are made according to real time demand; thus more SKUs of a product can be made on one assembly line.
Logistics — An efficient ‘just in time’ part movement system from warehouse to manufacturing facility can be implemented using robots.
Sustainability — Effective utilization of materials, manpower, energy, etc. leads to sustainable operations.
Big data analytics and machine learning
A recent study from IBM suggests that up to 2200 terabytes of data is generated in single month for an average manufacturing facility, which usually goes unanalyzed, thus rendering the data of no use. This happens because most of the manufacturers are not equipped with the right infrastructure or organizational structure to gain valuable insights to drive actions through such vast amount of data.
Multiple industrial data analytics tools now exist which can harness potential via analyzing these large chunks of data. One such solution is Mindsphere from Siemens, which lets users generate, process and analyze the data and lets them build industrial applications for better insights and informed decision making. Amazon also offers its cloud computing platform, AWS, for IIoT applications.
In summary, 5G networks will form the backbone of IIoT as they offer far greater data transfer rates, incredibly low latency, and superior reliability when compared to previous generation networks. With the rise of Industry 4.0, there has been an exponential rise in the smart devices at the shop floor and an ever-increasing amount of new data being generated which, if automated and analyzed right for intelligence & insights, has significant business impact. This calls for finding and investing in new solutions to new challenges and opportunities including robust networking solutions. Commercially available, high bandwidth SDRs have been making it easier for developing and testing such solutions, thus opening a whole new era for digital transformation at manufacturing facilities.