PORTRAIT of Alain-Serge Porret
Alain-Serge was born in La Béroche. He studied at the engineering school in Le Locle before moving to Lausanne to conduct doctoral research in the field of very-low-consumption integrated systems. In 2000, he left Switzerland to join a start-up in Silicon Valley. He spent thirteen years in the United States and worked for a series of start-ups, including digital television specialists Xceive. The company had up to 70 employees and some major clients, such as LG, Samsung and Panasonic.
In 2013, he joined the CSEM (Swiss Centre for Microtechnology), and specifically the integrated systems department, which was experiencing rapid growth. Indeed, low-consumption wireless technologies had already begun imposing themselves as the future standard.
His department currently has a central role in tackling the challenges associated with the Internet of Things (IoT). Their objective: to produce energetically autonomous wireless sensors capable of collecting data and processing it locally. Interview
What are the challenges posed by a world structured by the IoT ?
In 2018, information is relayed from the sensor to the cloud without necessarily being sorted first. A lot of data is therefore stored automatically, with no prior thought being given to its practical use. This presents a number of challenges:
- The greater the amount of data to be transmitted, the greater the energy requirements. So it is important that only pertinent data be transmitted. This requires the sensor to understand the data it receives, as well as the environment in which it is situated. In concrete terms, where vision sensors are concerned, video recording generates gigabytes of data. If used to regulate the heating or air-conditioning of a room for example, the system will only need to know the number of people present in the room. So we need to be able to tweak the “intelligence” of the sensor mounted on the camera so that it selects this information and relays it to a server. With this approach, it is possible to obtain a lightweight wireless installation that consumes little energy.
- The protection of personal data is another challenge. In the preceding example, it is hard to imagine that a camera would be permitted to film people all the time. It is essential that the information collected be broken down and that only the fragment of information that will serve to trigger the system be sent to the server.
- Finally, in the industrial world, confidentiality issues are extremely important. In the case of predictive maintenance on an automated production line, the robot manufacturer only requires access to the data that will allow its maintenance. Data providing information on production rates or quality should be excluded. Local processing of information serves not only to limit the volume transferred but also to avoid the GAFA networks, which currently collect much more data than they actually require.
All in all, what is fascinating in IoT projects is the need to combine different technologies: telecommunications systems, low-consumption processors, artificial intelligence algorithms requiring minimal resources and sensor portability.
What ultimate technological objective are you seeking to achieve?
20 years ago a term was coined that has always inspired me: smart dust.
It relates to the production of components that are so tiny, so cheap and that consume so little energy that they resemble dust. Technology is making greater and greater advances in this direction. We could imagine an intelligent camera in the form of a sticker, powered by the ambient light. It could monitor an elderly person in their flat and send an alert when no movement is perceived over several hours. Once again, no picture would leave the sensor, just the alert.
I believe that these kinds of technology will soon be in place because we are already capable of creating intelligent systems that integrate certain parameters of smart dust. We have recently been working on intelligent fire hydrants with wireless interconnections. In certain countries, up to 60% of the water is lost due to leaks in the networks. It takes a long time for such a loss to be noticed. These intelligent hydrants manage to detect the leaks by analysing the sound of the water. This allows excavations to be made at the right spot. In this project, the size of the sensors was of little importance. However, for medical implants, miniaturisation is of paramount importance, while the cost is less significant.
What do you think of the innovation ecosystem in the Neuchâtel region ?
We are lucky to have microtechnological expertise that finds application in the region’s “traditional” industry, but this could be extended. Neuchâtel could strengthen its positioning in the medical field, where miniaturisation, precision and reliability are vital. In IoT, I would say that the possibilities for innovation are limitless. You just have to have a good idea. On the other hand, our approach to innovation needs to become more vertical if we are to have access to the entire value chain, from the sensor to the server. This is also indispensable for improving products through the processing of data collected by sensors.
Lastly, I think that efforts to support SMEs in their innovation projects still need to be stepped up at all levels, including in the canton. They also need guidance in digitalisation and the setting up of digital processes that provide them with added value. The CSEM’s “Digital Journey” challenge has paved the way for dialogue with SMEs on this subject. And in 2019, Microcity SA will be a key structure involved in fulfilling this function.