“Businesses need to protect their IoT investments by ensuring that they do not find themselves locked in by early or proprietary technology decisions”

 

In this interview, Bob Flynn describes his involvement with oneM2M’s standardization and IoT system deployment activities as well as the commercial choices that businesses face as they deploy IoT systems more widely through their organizations.

Q: Would you begin with some introductory information about yourself?

BF: My background is in embedded development and digital signal processing. I spent several years with Lockheed Martin working on space systems. After that, I joined InterDigital Communications, a member of oneM2M, where I focused on technologies for cellular chip sets. That is also when I was introduced to oneM2M. Through InterDigital’s interest in the IoT market, I got involved in ETSI’s SmartM2M group. Then, I began to do some implementation work using oneM2M standards. Only later did I start attending oneM2M’s Technical Plenary (TP) meetings, spending most of my time in the testing and protocol groups.

As a part of InterDigital’s oneM2M implementation activities, I was responsible for the development and testing of an IoT platform, or a Common Services Entity (CSE) in oneM2M terms. This experience really helped me to contribute to those oneM2M working groups.

During this time, InterDigital set up a dedicated business – Chordant - to focus on commercial implementation opportunities. I moved to Chordant as the lead architect for their oneM2M based products. I also got more involved in architecture related contributions at the oneM2M standardization meetings.

One project of interest concerns oneM2M’s support of efficient communication of cellular devices. This is an on-going work item scheduled for Release 5 of the standard. The idea behind this concept is to design IoT systems to minimize any harmful effects on mobile networks, since networks are resources that many users share. This is very much in line with what the mobile industry’s body, the GSMA, has been promoting through its guidelines.

Q: How would you describe your current business activities?

BF: In mid-2020, I left Chordant to set up my own business, Exacta Global Smart Solutions. I wanted to focus on supporting companies that wish to implement solutions using oneM2M technology. I see a trend towards more adoption of standards-based systems and a need for greater technical support. You can see this trend developing in the Indian government’s announcement that recommends use of the oneM2M standard for smart city deployments, for example.

In addition to the support that I offer to companies, I am also working on a couple market adoption and research projects. One of these is a project for ETSI to define guidelines around semantic interoperability and another project that aims to improve the usability of data generated by IoT devices for machine learning use cases. A more recent project involves working with a group of students at my alma mater, Penn State University.

Q: What is your involvement with oneM2M now?

BF: I am the Vice Chair of the Test and Development Ecosystem (TDE) working group in oneM2M. As the name of the group indicates, there are two main activities that we focus on, the development of test specifications and developer support activities.

In the TDE working group we define specifications for conformance tests and interoperability tests. The conformance tests can be used by certification bodies, such as the Global Certification Forum (GCF), to verify that a oneM2M implementation operates as defined in the specifications. The interoperability test specifications are written to make sure that the oneM2M implementations from different vendors operate together. These specifications are the ones used at oneM2M Interoperability test events around the world.

The TDE group also works on developer guidelines that provide a variety of implementation examples that demonstrate how to use oneM2M standards. Some examples involve basic systems such as a monitoring and control deployment in a home. Others demonstrate more advanced and specialized IoT capabilities. The use of oneM2M’s semantic interoperability capabilities, for example, makes it easier for different types of devices to share their data with applications in machine readable forms. Since the cellular industry is so important for the IoT market, another example involves oneM2M’s capabilities to use the 3GPP SCEF/NEF interface for IoT devices. This contains functionality that is specifically designed to enhance the operations of IoT devices by enabling lower power modes and more configuration capabilities.

In terms of other research to support new capabilities, I am working on a semantic interoperability developer guide. This aims to demonstrate an enhanced form of semantic interoperability building on the semantics features in the oneM2M standard.

Q: Would you explain a bit more about the IoT semantics project and key outcomes?

BF: Ironically, semantics seems to have several meanings. For me, when the semantics concept was introduced to oneM2M, I had a tough time seeing the benefits. Its only later, in 2020, when I worked with some talented folks that the benefits became clear to me. Earlier this year, an EU funded project, INDICO, invited Andrea Cimmino from the Polytechnic University of Madrid and me to put together a tutorial to show how to use oneM2M and semantics ontologies like SAREF to improve the capabilities of an IoT deployment. In the first part of our tutorial, we described how the SAREF ontology can be used to enhance the discoverability of devices that have the type of information we are looking for.

In the second part of the tutorial, we demonstrated these discovery concepts using Chordant’s CSE implementation as well as the interoperability that can be achieved by using the oneM2M ontology. The demonstrator scenario involved a variety of different smart streetlights and pole-mounted environmental sensors. The streetlights use different on/off commands, which simulates the effect of different commands for devices which is likely when they are supplied by different vendors. The benefit of semantic interoperability in such a multi-device and multi-vendor scenario is that a member of a city’s operations team can issue a query to locate the type of device they wanted, such as a streetlight, as well as the commands to turn the light on and off. The command can apply only to streetlights from a given vendor. It could also apply to streetlights, from potentially several vendors, along a given street. Using the SAREF and oneM2M ontologies applications can send the proper command to each streetlight, without the operator having to configure each command for each type of streetlight. This feature is particularly attractive in a smart city deployment where there could be multiple vendor products. It is also useful for device developers because they will be able to provide this type of control without making changes to the logic on the devices. That means application developers can support many more vendor devices with their applications with very few changes.

If you want to learn more, you can find these tutorials on the oneM2M YouTube channel.

Q: What are your aims with Exacta GSS in the commercial market?

BF: I think that the potential that can be realized by IoT technologies and systems is at the initial stages of development. So many great ideas for innovative IoT solutions are hindered by the need to develop the entire system from scratch or using proprietary systems that can limit flexibility. Using an IoT platform can save a lot of time; using an open standards-based platform can bring additional benefits such as protection from vendor lock-in. The oneM2M standard has been supported for several years in Europe and S. Korea, with plenty of pilot projects and use cases to validate the platform. Since adding support for interworking with the 3GPP SCEF/NEF there has been further interest by mobile network operators and smart cities, with India being a notable case that I spoke about earlier.

My company offers services to companies and cities that are interested in adopting the oneM2M standard. Right now, I find myself spending a fair amount of time introducing interested parties to the benefits of oneM2M and how to get started. This is one of the things that led me to participate in a project at Penn State University.

Q: Tell us a bit more about the project with Penn State University.

BF: Some time back, I read an article where the City of Philadelphia did a smart city-based CAPSTONE project with Penn State. The Penn State University CAPSTONE project is a graduation requirement for final year students to apply what they learned in classes to solve a business sponsored project. I felt that supporting a CAPSTONE project using oneM2M would benefit both the students and oneM2M.

I worked with Professor Kyusun Choi at Penn State’s Department of Computer Science and Engineering and Katie Oberti of Nordic Semiconductor, Inc. to define a project around the use of the Nordic Thingy:91. The Nordic Thingy:91 is a battery-operated prototyping platform for cellular IoT using LTE-M, NB-IoT and GPS. It is easy to use and great for proof-of-concept demonstrations. It has several sensors for motion, impact, and air quality measurements. Our aim was to build oneM2M compliant smart devices with the intention of providing software that demonstrates working sensor and actuator devices to the open-source community through the Zephyr RTOS project.

The project scope intentionally left the specific application use cases up to the student team to identify. The team of 5 students chose to demonstrate how they can improve the operations of a local irrigation company using oneM2M to integrate and automate irrigation sensors. The team is also using this as their entry into a oneM2M hackathon sponsored by the Korea Electronics Technology Institute and ETSI. The results of the hackathon will be announced sometime in Q4 of 2021.

I must say that I am excited about the outcome of this project. This team of students will exercise what they learned at Penn State to develop and contribute to an open-source project. They will learn about deploying IoT devices over a cellular network using standards-based components, like the oneM2M and the certified Nordic radio modules. They will also be interacting with students from more than 10 universities in the USA, Korea, India, France, Spain, and Italy.

By early December, I expect that there will be hackster.io articles in the oneM2M channel from the participating teams and GitLab repositories containing project code from each of the teams.

Q: Finally, what advice would you offer to organizations interested in IoT?

BF: Right now, IoT is complex, and it can be costly to build systems. My advice is to protect the investment that you are making by ensuring that you do not find yourself locked into your first decisions. That is because the technology is growing and will continue to change. You want to be able to adapt to changes in your business needs and technology enhancements. I think the best way to protect your investment is to make sure that you can grow the deployment with the flexibility to add and replace parts of the system as needed. Use of standards-based technology is my recommended approach.

Do not expect to be able to have a homogenous system; the selection of the best products or applications for your evolving deployment is likely to have a variety of underlying components. Use of a standard like oneM2M offers greater opportunity for interoperability of different components and applications.