Sensor network mimics brain function

Researchers at Brown University in the United States have presented a sensor network designed so that its chips can be implanted in the body or integrated into wearable devices. The communication network can efficiently transmit, receive and decode data.

According to the study, each submillimetre-sized sensor mimics the brain’s neurons and communicates via spikes in electrical activity. It can detect specific events and transmit data in real time using radio waves. The events that the sensors identify and transmit can be specific occurrences, such as changes in the environment they are monitoring, such as temperature fluctuations or the presence of certain substances.

“Our brains work very sparsely. Neurons don’t fire energy all the time. They compact information and distribute it sparsely, which makes them very efficient. We’re mimicking this structure in our approach to wireless telecommunication,” explains Jihun Lee, a postdoctoral researcher at Brown University and lead author of the study.

According to Lee, the sensors don’t send data all the time, but only relevant data as needed, such as small bursts of electrical spikes, and are able to do so independently of the other sensors and without coordination by a central receiver. By doing this, it is possible to save energy and not flood the receiving hub with less important data.

The sensors are capable of using minimal energy and, in addition, the external transceivers supply them with energy as they transmit data. However, this means that they need to be within range of the waves sent by the transceivers in order to be charged with energy. This ability to operate without the need to be connected to a power source or battery makes the sensors convenient and more versatile for use in different situations.

This radio frequency transmission scheme also makes the system scalable and solves a common problem in today’s sensor communication networks – perfect synchronisation to work well. The creators of the sensors say that this advance could in the future help shape the way information is collected and interpreted from these chips, especially as sensors are becoming ubiquitous.

“We live in a world of sensors. They are everywhere, certainly in our cars, in many workplaces and increasingly in our homes. The most demanding environment for these sensors will be inside our human bodies,” emphasises Arto Nurmikko, professor at Brown University’s School of Engineering and senior author of the study.

The researchers believe that the solution could help lay the foundations for a new generation of implantable and wearable biomedical sensors.

How the tests were carried out

The team designed and simulated the electronics on a computer and worked through various manufacturing stages to produce the sensors. The work builds on previous research by Nurmikko’s laboratory at the university, which introduced a new type of neural interface system called “neurograins”.

The efficiency of the system and how much it could be expanded were evaluated. In these tests, 78 sensors were assembled in the laboratory and proved capable of collecting and sending data with few errors, even when transmitting at different times. Through simulations, it was possible to demonstrate how to decode data collected from the brains of primates using around 8,000 hypothetically implanted sensors.

The next step is to optimise the system to reduce energy consumption and explore wider applications beyond neurotechnology. The work was published in Nature Electronics.

Skin-like sensors

Another article also published in Nature recently reported on work at Stanford University focussed on elastic skin-like electronics carried out over a decade ago. Now a new design and manufacturing process has been presented for these circuits, which are five times smaller and operate at speeds a thousand times higher than previous versions. The researchers have demonstrated that it is possible to trigger a micro-LED screen and detect a Braille matrix more sensitively than human fingertips.

The core of the circuits are stretchable transistors made of carbon nanotube semiconductors and soft elastic electronic materials developed in the university’s laboratory. Unlike silicon, which is hard and brittle, carbon nanotubes sandwiched between elastic materials have a network-like structure that allows them to continue working when stretched and deformed.

“It’s been many years of materials development and engineering. We didn’t just have to develop new materials, but also the circuit design and the manufacturing process. There are many layers stacked up and if one layer doesn’t work, we have to restart everything from scratch,” explains Zhenan Bao, professor of chemical engineering at Stanford and lead author of the article.

In a demonstration of the new extendable electronic design, it was possible to accommodate more than 2,500 sensors and transistors in one square centimetre, creating an active tactile array 10 times more sensitive than human fingertips. The researchers showed that the sensor array can detect the location and orientation of tiny shapes or recognise entire words in Braille.