Using privacy-centric Bluetooth bracelets for COVID-19 contact tracing

Article By : Asem Elshimi

Bluetooth bracelets can expedite coronavirus tracing without compromising user privacy...

How can Bluetooth direction-finding be used in digital contact tracing? For a start, the technology requires additional hardware and software infrastructure to be fully functional. For example, more antennas are required on the receiver or transmitter side.

The Bluetooth direction-finding-based contact tracing system comprises two parts: first, attaching Bluetooth tags to individuals, and second, gateway infrastructure, the system’s backbone. It’s analogous to how computers and servers work together to form the Internet.

Installing the infrastructure is a sophisticated yet portable operation that must happen first. Once done, how do designers attach Bluetooth tags to individuals? As mentioned earlier, existing smartphones, while enabled with Bluetooth radios, cannot perform Bluetooth direction-finding-based tracking. Smartphones available in the market today are best suited for received signal strength indicator (RSSI) solutions.

The alternative to smartphone solutions is wearable Bluetooth tags or bracelets. Bluetooth bracelets can be developed at a cost of $1 or $2 each, and they could operate for up to 10 years on a coin cell battery. Bluetooth bracelets can be manufactured using Bluetooth direction-finding technology as well as other more accurate Bluetooth location measurement technologies. This ready-for-deployment solution can be used to tackle densely populated urban environments where smartphone RSSI distance resolution is insufficient.

In the foreseeable future, highly-dense commercial environments, such as hospitals, offices, and retail spaces, could be populated with Bluetooth direction-finding gateways. People would be asked to wear Bluetooth bracelets to enter buildings and facilities. Every individual would actively check whether he or she is exposed to a confirmed case or not.

photo of a Bluetooth bracelet that can be used for digital tracingFigure 1 Bluetooth bracelets can be developed at a cost of $1 or $2 each.

In this Bluetooth bracelet scenario, competing solutions need to work together. As mentioned in the first article in this series, engineers have two attractive Bluetooth contact-tracing technologies. RSSI, which is simple and inexpensive, also incurs a high false-positive and false-negative rate. It can be deployed in the near term for high-level Bluetooth tracing. Second, Bluetooth direction-finding can provide sub-one-meter accuracy, and is best suited for dense urban environments. However, it requires an initial installation of infrastructure, and thus, it will be deployed in the long term.

Technology RSSI Bluetooth direction finding
Granularity 3-5 meters Sub-meter
Deployment timeframe Short term Long-term (requires infrastructure)
Cost Inexpensive Initial cost of infrastructure
Use case Suburban and rural areas, as well as open public spaces Retail, hospitals, smart buildings, and other facilities

Figure 2 The table illustrates the advantages and limitations of each Bluetooth location technology.

From the implementation standpoint, there are two viable solutions: a smartphone app and Bluetooth bracelets. As discussed earlier, Bluetooth direction finding is not supported by smartphone software, which means it would more likely be implemented using Bluetooth bracelets. What about RSSI? An RSSI smartphone app can be deployed in the near term, however, the caveat to RSSI phone apps is that not everyone in the world has a smartphone.

Approximately half the world’s population (3.8 billion people) own and use smartphones. Besides, political establishments across the globe have varying requirements, especially regarding privacy and human rights. Some governments might completely reject a smartphone solution for political and economic reasons. So, how do we provide RSSI-like applications to the rest of the world? Cost-effective Bluetooth bracelets will enable broader access to contact tracing. Bluetooth bracelets and smartphones can complement each other in facilitating an effective global digital-tracing solution.

Implementation BLE bracelets Smartphone app
Cost $1 each App development only
Timeframe Medium to long term Near term
RSSI Yes Yes
Bluetooth direction finding Yes No
Identity blind Yes No
Global coverage Yes No
Add-ons Yes No

Figure 3 Bluetooth bracelet and RSSI-based smartphone apps can complement each other for providing an effective digital contact-tracing solution.

While the smartphone app solution has the advantage of near-term deployment, Bluetooth bracelets offer the possibilities of customization. The same system-on-chip (SoC) that enables Bluetooth connectivity on a bracelet can also take medical diagnostic readings such as temperature, heart rate, and blood pressure from sensors integrated into the bracelet. The bracelet can then notify its user of symptoms.

Privacy-based solution

Mass monitoring of people’s locations poses legitimate privacy concerns. Although the COVID-19 pandemic is a global emergency and not an everyday situation, we cannot afford the risk of exposing everyone’s location. While some parts of the world are more tolerant of privacy concerns than others, most of us would prefer privacy-based solutions.

Therefore, digital tracing algorithms must be built with data protection as a primary goal. Transparent algorithms and ethical data handling are non-negotiables. Such concerns must be addressed whether the implementation is a smartphone app, a Bluetooth bracelet, or both.

Unlike smartphone apps built around user data collection, making it more difficult for people and political establishments to trust a smartphone solution, a Bluetooth bracelet offers a privacy-centric alternative. Your smartphone knows who you are; it’s built to do so. On the other hand, the bracelet doesn’t need to know who you are.

The concept is simple: the bracelet can be designed in an identity-blind way, with each device being assigned a unique identification number (UIN). End users wear the bracelet while wandering grocery stores and other public spaces. The Bluetooth tracking magic—RSSI, direction-finding, or any other Bluetooth technology—works on the bracelets. And the bracelets will exchange short random—but identifiable—messages.

Here is a simplified scenario of how this can work: Say Shaina and Ryan come in contact at a store. Shaina’s bracelet would record Ryan’s bracelet’s encrypted message but won’t know anything about Ryan. Ryan’s bracelet would do the same with Shaina’s bracelet. Both bracelets communicate with a nearby server.

Days later, Ryan gets diagnosed with COVID-19 infection. The medical team goes online and flags Ryan’s bracelet as a “confirmed case.” The server then flags Shaina’s bracelet as a “direct contact subject.” When Shaina logs in to the database using her UIN—not her personal identity—she would see that her bracelet is flagged as a direct contact subject. Now Shaina can take immediate action to isolate herself, seeking testing and care if needed.

While this is a simple scenario, it showcases the potential for an identity-blind digital tracing implementation. It is the individual’s responsibility to check and confirm his or her status on the database, and the database doesn’t know anything about the users’ identities.

Bluetooth ready for deployment

Considering the urgency of the global pandemic, design engineers need to expedite the identify-and-trace action plan. Bluetooth digital tracing has the advantage of being a proven and mature technology over alternative radio technologies.

Based on decades of deployment, there is a sense of confidence in Bluetooth software because of its simplicity and ultra-low power consumption. Engineers know what to expect from Bluetooth in terms of performance and technical features, and most importantly, the ubiquity of Bluetooth means that it’s ready for deployment now.

Editor’s Note: This article is for general information purposes only and provides an overview of a specific developing situation that continually evolves. It is not intended to, and should not be construed, for public health guidance. —Majeed Ahmad

— Asem Elshimi is an RFIC design engineer for IoT wireless solutions at Silicon Labs.

Related articles:

Leave a comment