Smart buildings represent a major development in the field of construction, revolutionizing the way we both live and work in these spaces. These structures integrate intelligent management systems that collect and analyze data in real time, reducing energy consumption and facilitating preventive maintenance, among other things. They also promote safety, for example by monitoring the operating status of elevators and defibrillators and detecting when people enter and exit the premises.
These buildings also offer many other features such as automated heating, ventilation, air conditioning, lighting and home automation and Internet of things (IoT) applications. The convergence of these advanced construction, communication and automation technologies help create a flexible, efficient, comfortable and safe environment for occupants.
Yet implementing this “intelligence” requires the deployment of a wide range of sensors, actuators and software programs, which in turn introduces a burden in terms of equipment, safety and cost issues. In response to these challenges, researchers Daqing Zhang and Xujun Ma of Telecom SudParis are studying the exploitation of wireless signals to power the existing radiofrequency (RF) devices with a greater capacity to sense the environment and to provide the customized responses. Their work is part of SUST(AI)N, a European project aimed at creating an “aware” work building, using connected and sustainable RF devices.
“Sensitive” buildings, focused on human needs
SUST(AI)N is an acronym for “Smart bUilding Sensitive To dAily sentiment.” While the name highlights the project’s sustainable and artificial intelligence (AI) aspects, its key mission is to create systems that achieve self-awareness inside buildings. “Smart buildings currently process ‘unconscious’ information from their sensor network,” Daqing Zhang explains. “SUST(AI)N aims to turn these buildings into organisms with self-awareness based on an overall understanding of everything that is happening within them and on automation control.” To accomplish this, research teams rely on connected devices, combining highly accurate – near omniscient – detection, distributed intelligence and probabilities.
Simply put, these systems will exploit wireless or RF signals, which are already emitted from existing commodity devices, for use in future aware buildings. “Nowadays, every office room has access to Wi-Fi signal, and everyone has a 4G or 5G smartphone,” says Daqing Zhang. “Until now these signals had been used solely for wireless communication, but in this project, we want to use them for sensing as well. Each signal type has different frequency and characteristics. They can therefore be used to sense different types of information.”
The SUST(AI)N project will investigate those wireless signals mainly for human activity sensing. Some will be used to locate people in the building, for example to ensure the security of sensitive rooms, or to detect the occupancy rate of a room to manage lighting and heating. Others recognize movement and posture or monitor physiological signs, including heart rate and breathing, to detect incidents such as fainting.
Wireless signals with a wide range of potential uses
Stephan Sigg, coordinator of the project and researcher at Aalto University (Finland) and Daqing Zhang are not beginners in European projects. This is the third time the two scientists from the same ubiquitous computing community have worked together. The SUST(AI)N project, which has already reached an advanced phase since its launch on October 1st, 2022, has benefited from their experience. Among the different types of RF signals that exist, scientists from Aalto and Telecom SudParis have already explored possibilities involving RFID and LoRa.
RFID can be used to locate or track physiological data by passively reading what is referred to as a tag. Yet this use requires that each person moving through the building wear a tag. In addition, due to the signal’s short transmission range RFID readers must be installed throughout the premises. The teams have therefore prioritized exploitation that recognizes hand gestures and monitors the breathing of several subjects in the same room. On the other hand, LoRa is a signal resulting from the LoRaWAN (Long-Range Wide-Area Network) communication protocol and is used for IoT. It requires little energy and has a very long sensing range – ten times that of Wi-Fi – making it valuable for locating people or detecting a presence in a hallway or parking lot.
Research on the use of Wi-Fi is a key future step in this project. “We haven’t yet started our work on this task but we expect to do so this year,” says Daqing Zhang. Unlike the RFID signal, Wi-Fi does not need tags or signal readers, just a phone. It is expected to be used to capture physiological data such as breathing, heart rate and other human activities.
Plans for a demonstration building
The exploitation of wireless signals for sensing is just one of the project’s technological blocks. It also relies on the expertise of three other academic partners based in Spain, Italy and Turkey. One area of study is the implementation of a distributed intelligent network to connect all the sensors and statistically process the data using AI algorithms. Cryptographic protocols will be developed to protect the collected data.
In addition to designing these technologies to be as energy efficient as possible, a specific research area will focus on the harvesting of ambient energy to operate them. “Our partners at the Polytechnic University of Catalonia in Spain are studying different possible energy sources, such as solar and thermal energy as well as harvesting energy from radio frequency signals,” says Daqing Zhang.
The various developed technologies will be gradually incorporated in a pilot building, on the campus of Aalto University in Finland. It will serve as a showcase for all the innovations. Telecom SudParis plans to send a member of its team there for a few months this year, to discuss the integration of developments, applications and use cases with the local partners. One cannot say that all the features developed, some of which could potentially raise ethical issues, will be deployed in the end. “We will first see where all our efforts on the various technological innovations lead us over the next two or three years,” concludes Daqing Zhang.