Experts At The Table: What’s Missing In The IoT

Semiconductor Engineering sat down to discuss the future of the IoT with Oleg Logvinov, director of market development for STMicroelectronics’ Industrial and Power Conversion Division; Martin Lund, senior vice president of the IP Group at Cadence; Naveed Sherwani, president and CEO of Open-Silicon; and Damon Hernandez, a member of the Web3D Consortium. What follows are excerpts of that conversation.

SE: When it comes to wearable electronics, having these devices next to your body means they need to extremely temperature-sensitive, particularly for things next to your skin. That requires a much more complex design, doesn’t it?
Lund: That’s a given and it’s going to happen. But it’s just a different design constraint. It takes a couple generations and we’ll get it right.
Logvinov: Over time, a smartphone might be right next to your temple. How real is this in terms of energy harvesting? Very real. If you look at devices that harvest from solar light or ambient light, it’s enough to power the sensors. There are sensors to provide temperature and vibration data.
Sherwani: Energy harvesting is a very old idea. There have been watches that harvest the kinetic energy in our hands for at least 70 or 80 years. It’s not a very difficult idea. It will be done.
Logvinov: It’s there today. There are devices to measure the temperature in your home based upon ambient light.
Hernandez: I just hope we find more clever ways to get that done. There’s all this energy next to you and you’re wearing Google Glass, and that’s a concern. We’re going to be the first generation that has laptops next to us. What effects will that have on our health? What about having a cell phone in your pocket. People will have to get better about how to generate power because we also will be the first generation to see the effects of faulty designs.
Logvinov: It probably will take several generations.

SE: What are the top items missing in the Internet of Things?
Logvinov: A unified approach to multi-layer data abstraction that allows you to expose the same data from the same sensor and actuator to multiple consumers of the data.
Sherwani: There are a lot of good ideas when it comes to system-level solutions, but the people who are going to implement those ideas aren’t knowledgeable enough. You have to provide a complete solution. At the semiconductor level we have to figure out optimal design, not the over-design we have experienced.
Hernandez: It’s having people throughout the production chain actually involved. That’s a big problem. There are too many people in silos talking about this great thing that needs to interconnect, but they’re not interconnecting with anyone else.
Lund: There is a risk of an explosion of different standards, which means nothing talks to each other. That means it’s not actually connected, and therefore it’s not an ‘inter-net.’ There is a need for some alignment of how these devices can talk to each other. It can’t be a monolithic solution. On the other hand, it can’t be the Wild West because that means nothing will be connected.

SE: There are billions of things, which means there can’t be a single optimal design. Where do you draw the line so we get economies of scale to build silicon?
Logvinov: Someone said the beauty of standards is that you can pick the one you like. That’s a problem. Today we have many standards targeted at the same thing. We need to find a way for all of the stakeholders to get involved—semiconductors, systems, services, visualization, and consumer products. Even in the areas where we have multiple standards, why not reduce it down to one? If you look at a home today, there is an alarm that speaks its own language, a fire and safety system built on a different language, a Nest thermostat on a third language. Eventually there will be a benefit if they start talking the same language. Then you can take a visualization tool and spread it across multiple teams so they can control the data. We’re not there yet. Hopefully we’ll find a path there.
Hernandez: I would like to see the groups form to align the process. We see a lot of that in Web 3D standards. There are a lot of industries that create their own standards, and then it causes communications problems. It’s about people identifying best processes that leverage standards. At the end of the day people will choose whatever path they want, but if there’s a guide to tell them this standard works best and gives options that will work best.
Lund: When can you design products for the Internet of Things? It has to leverage a lot of the Internet technologies that exist today. I don’t think we’re going to find a new set of protocols. It won’t be orthogonal. But there are some missing pieces that will be added to solve the alarm system or low power or the light bulbs talking to each other. You don’t need an Ethernet cable running between them. It’s really connecting Internet technologies to the legacy, unconnected and ‘not smart’ world that is out there.

To read part one of this roundtable, click here.

To read part two of this roundtable, click here.