Blurry Box Encryption Scheme and Why It Matters to Industrial IoT

Manufacturing plants, devices, and end products require new protections. While the Industrial Internet of Things (IIoT) offers new avenues for monetizing software, unprotected software puts the underlying system at risk, whether it is an industrial computer, an embedded system, a mobile device, a Programmable Logic Controller (PLC), or a microcontroller.

Methodology of Modern Cryptography

Cryptography is based on a systematic approach of defining and proving security properties of systems. Two steps are needed to define a system's security:

1. The security property has to be described precisely, and;

2. There must be as few restrictions placed on the attacker as possible.

Today’s cryptography is not only about keeping messages private. The field has broadened its purview. A major step in this direction was the introduction of public key cryptography. This is a different kind of message encryption, where different keys are used for encryption and decryption. Several other primitive approaches such as digital signatures, key exchanges, and commitment schemes have been designed and are used in more complex systems, such as voting schemes, online banking, crypto currencies or general secure multi-party computation.

The Blurry Box Scheme

At its core, Blurry Box is based on the assumption that a hacker lacks the domain knowledge necessary to create a software product. This lack of domain knowledge can be exploited to achieve secure protection. The main idea is to split the program code into small pieces to make it practically infeasible to retrieve all pieces by running the code. The hacker’s lack of domain knowledge prevents him from creating additional pieces on his own.

Only programs that are sufficiently complex can be protected effectively. Note that not the whole program has to fulfill this requirement; it suffices if only a part of it is sufficiently complex. Typically, this complex part is exactly the critical part that needs to be protected. Programs that meet this requirement include video games, raster graphics editors, and feedback control systems.Results, Insights and Best Practices from IIC Testbeds: Time-Sensitive Networking Testbed

Results, Insights and Best Practices from IIC Testbeds: Time-Sensitive Networking Testbed

Time Sensitive Networking is a new technology that enhances Ethernet, a foundational piece of the Internet of Things (IoT). The IIC’s TSN Testbed endeavors to apply new TSN technology in a manufacturing system with a wide range of automation and control vendors, to display the new capabilities and value of TSN. Manufacturers are dealing with automation and control systems that make up a large part of the 50 billion things we are trying to connect, and they usually require the pieces and parts and the overall system to be very deterministic in nature.

This deterministic characteristic is important to many industrial automation and control applications for manufacturing in oil and gas, utilities and transportation. TSN is an enhancement of Ethernet (standards IEEE 802.1 and 802.3 comprise what is generally considered Ethernet) to bring more deterministic capabilities to the network. It is enabling more and more devices, applications and systems to use a standard, open, inter-connected network that is the basic concept and driver of the IoT. If manufacturers do not connect things because they are worried about how the network is going to perform, they cannot implement IoT applications. This technology is viewed as a very important enhancement and upgrade to the standard networks to enable the overall Industrial IoT.

The goals of the TSN Testbed are to:

1. Show TSN’s readiness to accelerate the marketplace and

2. Show the business value of accelerating the adoption of this technology.

Time Sensitive Networking - A Brief Overview of TSN

TSN brings TSN brings a number of enhancements to Ethernet. TSN comprises a number of enhancements, but three key capabilities form the core of TSN. They include:

1. Time Synchronisation - A distributed, precise sense of time

2. Sending scheduled Traffic flows - Based on the precise sense of time, the network infrastructure must be capable of forwarding critical automation and control traffic on a timely basis.

3. Central, automated system configuration - Consolidate application requirements, develop paths and schedules for the traffic flows and distributes that to the relevant network authorities.

Testbed Objectives

This testbed is designed to grow and achieve objectives in a progression. The three key capabilities listed above represent the initial technical milestones of the testbed. In initial plugfests, the focus was first on establishing time synchronization. Once this was accomplished, devices began to send scheduled traffic over the network. In our upcoming plugfests, we will focus on the central, automated configuration aspects of TSN.

In addition to the technical milestones, the TSN Testbed is looking at interfacing with a range of tools and applications within these spaces, from the programming and configuration tools, down to the actual operational input and output devices typically found in these environments. The testbed is trying to handle the whole process from designing a plant to operations of a plant using these new capabilities.

Influencing Standards

Specifically, the testbed has been influencing two Ethernet standards: IEEE 802.1 and IEEE 802.3. All TSN Testbed work is filtered to Avnu, a standards organization developing the interoperability and certification based on IEEE’s TSN. Through Avnu, it is channeled back into IEEE [Institute of Electrical and Electronics Engineers].

When TSN gets into the manufacturing area, the TSN Testbed team anticipates the TSN aspects will influence OPC Unified Architecture (OPC-UA), the Object Management Group Data Distribution Services (OMG DDS) standard and the ODVA, among others such as SERCOS International or Profinet International are looking to adopt TSN. Common Industrial Protocol (CIP), SERCOS and Profinet are industrial automation and control protocols often used between Programmable Logic Controllers (PLCs), input/output devices, motors and drives, and robots. They support tight synchronization and tight deterministic systems and can benefit from the capabilities TSN brings.

Making Factories Smarter Through Machine Learning

This article highlights a specific company that is not only aware of the value of the IIoT but is truly recognizing value from unlocking the meaning of the data. The predictive maintenance presented in this article is representative of actions being taken as a result of IIoT and I4.0. To become more competitive by leveraging the convergence of the IT and the OT in a secure, safe and connected manner, smart factories see this as one of the key ways to increase productivity and operate more efficiently through higher availability and increased machine and asset utilization.

Machine Monitoring Without Predictive Maintenance

It is important to state that a machine learning-based monitoring system could detect the first failure peak, giving enough time to stop the line in a controlled manner. A production and workforce could then be reassigned to reduce the failure impact over production line productivity. In this case, a relatively simple machine learning algorithm is able to detect these types of anomalies within a variable, effectively warning about the problem when a first peak is detected or before.

Machine Learning For Predictive Maintenance

Fast forward a few years to a new system where machine learning is being used to analyze the data to predict potential system failure14. First an understanding of the machine learning approach is needed to gain full appreciation of the scope of the predictive analytics performed to predict when the component/system may fail and even more importantly why the failure may occur. This, in turn, allows system optimizations that can extend the lifetime of the asset and the overall system.

To build the complete system, the workflow with data reduction would be used. The data is taken from the manufacturing system and sent to a machine learning algorithm that uses the new data and other information to produce the predictive system. While data is travelling within this process, a summarization is performed, helping to only move data that is needed to solve the asked question. This helps to reduce the bandwidth utilization and the response speed.

Summary

This system represents the convergence of the OT and the IT. Through this convergence and advances in sensor fusion, edge and cloud computing, machine learning is seeing adoption in many ways and is now playing a central role in smart Factory predictive analytics. 

Maximizing system operation and lifetime provides valuable advantages to further advance the manufacturing facilities.

Results, Insights and Best-Practices from IIC Testbeds: INFINITE Testbed

INFINITE was designed to develop software-defined infrastructures to drive the growth of Industrial Internet products and services. INFINITE is a state of the art innovation platform and ecosystem of technologies and business expertise covering sensors, gateways, connectivity, cloud, analytics and security. By recognizing that no one organization can do it alone in Industrial IoT (IIoT), INFINITE established strong collaborations between key stakeholders from different domains and sectors working on the common goal of seamless and interoperable infrastructure.

INFINITE was purpose-built as a ‘horizontal’ platform meaning that the combination of the many technologies made available by the INFINITE ecosystem will satisfy almost all of the anticipated technology requirements of future IIoT application domains across all sectors and verticals.

From the many use cases, in their varying levels of maturity, the INFINITE Testbed team has learned specific lessons, discovered and documented best practices and experienced outcomes that will help improve the lives of those living within the boundaries of the deployments and be the foundation for expanding the innovations to reach larger populations.

The INFINITE team developed a use case engagement process that takes into account the multi- vendor and multi-partner composition of the INFINITE ecosystem. This unified process is an end- to-end integration of different partner and technological processes that combine to support all activities covering initial use case engagement with INFINITE customers, requirements specification, design, deployment, results and use case sign-off.

Smart Factories and the Challenges of the Proximity Network

In the Industrial Internet Consortium’s (IIC) Industrial Internet Reference Architecture (IIRA), examples of architectural patterns for Industrial Internet of Things (IIoT) are described; two of which we select for this Smart Factory discussion: the 3-Tier Architecture and the Gateway- Mediated Edge Connectivity and Management Architecture (Figure 1). In both architectures IIoT gateways and edge devices form the boundaries of the proximity network. The challenges and corresponding solutions in the proximity network will be viewed in terms of this general architecture.

As we studied the proximity network in the factory, the Synapse team noticed a prominent "progression of challenges" (Figure 2) that begins with specific use cases. Each use case creates a unique set of integration challenges that drives the selection of connection technology such as wired Ethernet, Wi-Fi or 802.15.4. The connection technology creates a secondary set of challenges that are often overlooked in the initial system design phases. We have grouped this secondary set into three general categories: Distributed Intelligence, Deployment and Long-term Management. We will focus on a subset of use cases found in the Smart Factory and follow the corresponding progression of challenges and proposed solutions.

Use Cases Drive Integration Challenges

Most factories have well established and highly efficient processes already in place. However, plant managers still focus on ways to improve their process efficiencies. In some cases, they operate in a commodity market where the business operates on a few cents of margin per unit. In other cases, the plant manager has tight deadlines or safety concerns driving the need for greater visibility into problems earlier in the process. Most of the possibilities to gain additional efficiency occur at the integration points between different automated processes. Quality and production flow need to be first understood and then ultimately controlled at these integration points.

While the majority of factories explore how Machine-to-Machine (M2M) and IIoT technologies improve process, few factories have been able to successfully implement predictive maintenance. Many factories have mechanisms for scheduled preventative maintenance and redundancy on critical systems, but few predict failure in time to perform maintenance. Predictive maintenance generally involves collecting sensor data such as temperature and vibration from critical components (such as motors) and analyzing it over time. The algorithms for detecting maintenance conditions can be as simple as threshold crossing or as complex as a trained neural network.

Integration Challenges Drive Connectivity Decisions

The IIoT System Architecture provides three discrete connectivity decision points (Figure 6): the sensor/actuator interface, the proximity network and the backhaul network. The various integration challenges directly impact the choice of connectivity at each of these interfaces.

Sensor/Actuator Interface: Powering edge devices and integrating sensor/actuators are challenges with a direct bearing on how the sensor/actuator interface is chosen.

Backhaul Network: Dictates a large portion of the gateway functions, in turn affecting the proximity network. Typically runs IP over wired Ethernet, Wi-Fi or cellular.