Imagine you step back into your home on a cold evening and the house gently lights itself up to a warm, cosy ambience. This is probably a quintessential example of a smart home, made possible thanks to many ubiquitous next-generation smart sensors.
In the invisible background, a WiFi signal flies in our house, bending and reflecting on our bodies and furniture. By studying the physical shape of the wireless signal, abstractly indicated by Channel State Information (CSI), a WiFi smart sensing system could ‘see’ the environment.
Why WiFi? WiFi is readily accessible in terms of the ubiquitous presence of the infrastructure and its use of licence-exempt spectrum. In other words, WiFi-based smart sensing could be implemented on existing WiFi hardware and could face relatively fewer regulatory hassles.
On top of that, WiFi-based sensing is much less intrusive than cameras, making public acceptance easier. Most importantly, WiFi’s versatility has been demonstrated in various studies, ranging from human activity monitoring to detecting fine-grained environmental changes, such as temperature and fire, according to research conducted at the University of New South Wales (UNSW).
Until recently, such applications have been primarily confined to laboratories due to the lack of hardware and software support for CSI extraction in consumer devices, which is essential to WiFi sensing.
Tackling this challenge, IEEE, the WiFi governing body, proposed the IEEE802.11bf WiFi SENS standard. Scheduled for approval in 2024, the standard aims to incorporate WiFi sensing capability into common WiFi hardware, like your home router and mobile phone.
According to the SENS standard, commodity WiFi devices will be able to:
- broadcast their WiFi sensing capabilities;
- request and set up WiFi sensing; and
- exchange WiFi sensing feedback and information.
As a result, we could see our homes and old-school shopping centres turned into smart spaces, capable of monitoring the people and the environment from a web of connected, smart WiFi networks.
Although this could make our life much more convenient, it is still understandable that some people might want to opt-out of such WiFi sensing systems due to privacy concerns, such as one implemented by retail shops to track customer location and movement.
To this end, possible methods include a default ‘zero-sensing’ opt-in scheme to be part of WiFi sensing systems, or a zero-sensing signal implemented by a user-wearable device to exclude the user from WiFi sensing systems. Our separate story dives into aspects of protecting user privacy in such public WiFi sensing systems.
In this article, we argue that even in such circumstances, WiFi sensing for critical applications such as fire emergencies should be mandated to be operational by the policymakers.
As WiFi sensing runs on Artificial Intelligence (AI) enabled models to identify real-world events, the mandate should target to enforce the operation of models for critical application sensing.
Specifically, the critical sensing application should be enabled regardless of whether the WiFi sensing system is in zero-sensing mode.
For example, in the context of Covid-19, WiFi sensing systems should be mandated to accurately monitor the number of customers in a shop to ensure social distancing compliance, regardless of whether any customer has opted for the zero-sensing.
Moving forward, we firstly propose the regulatory body to consult with government departments that might have an interest in public space monitoring and management to create a list of critical applications to be mandated.
After that, the government should enlist universities, research institutions and technical groups to establish an effective, practical, and yet, a privacy-preserving approach to implementing such mandates for WiFi sensing for critical applications.
This should ideally be conducted in direct collaboration with IEEE, to ensure the SENS WiFi standard incorporates such mandates.
Junye Li is a PhD candidate at the University of New South Wales, and finished his Bachelor of Electrical Engineering in 2019. His research focuses on WiFi sensing applications for network and environment awareness. This paper was co-authored by Aryan Sharma, Eranga Perera, Deepak Mishra, Joseph Davis & Aruna Seneviratne
Do you know more? Contact James Riley via Email.