Why Herald exists, how to get the code, and how to integrate Herald in to your mobile app.
VMware Herald is an open source Bluetooth Low Energy (BLE) based protocol licensed under the Apache-2.0 license for the reliable exchange of information between a range of mobile phones. Herald consists of a low-level BLE protocol, with a choice of data payloads. The Herald Envelope and Herald Secured payloads are examples for contact tracing applications that provide epidemiological information, individual privacy, ensure security, and allow international interoperability - even between ‘centralised’ and ‘decentralised’ contact tracing applications.
The aim of this research is to provide a much more reliable way to perform local information interchange between mobile phones. This has many potential applications, including commercial use, but the key one we’re concerned with is the enable more reliable Bluetooth Proximity Detection to improve Contact Tracing applications worldwide in order to help stop the spread of COVID-19.
Our research has been split into several parts. We’ve also provided an independent introduction to mobile app based contact tracing and provide explanations of the key terminology of such a system. This aims to demystify the technology and increase public trust.
The remainder of this page breaks down the key areas of the documentation on this site.
To see information on integrating Herald with your own application, please see the developer & integration guide.
You can reach out to us and log general project and research data requests and issues on our GitHub issues site. We are also keen to receive contributions to all of our open source repositories and research areas:-
This project is maintained primarily by VMware personnel and some external contributors in our spare time. We aim to respond to logged issues and emails within 24-28 hours during week days (Monday-Friday) in the UK time zone.
Below are the latest API documentation for each platform:-
A layman’s introduction to contact tracing and mobile app proximity detection.
A scientific paper suggestion a fair and independent approach to measuring the efficacy of a mobile phone based contact tracing protocol, providing a direct calculation of efficacy that can be used with the Oxford Risk Model to calculate the extent to which such an approach can control the spread of a virtus in a given population.
During our research on creating an efficacy measure we created the Herald Bluetooth LE protocol implementation to test the ideas. We found a welath of issues in Bluetooth on mobile phones, identified the issues, and workaround them in order to provide a robust communication protocol.
This low level protocol is independent of any information you pass over it. You could use it for contact tracing proximity detection, but also other uses within your own applications. This low-level protocol provides reliable communications between proximite mobile devices.
We have created a design for a payload that not only preserves privacy, but also provides the right levels of epidemiological insight to prevent virus spread. We show this can be done in a way that ensures security of the contact tracing system, and can be trusted by people worldwide as they go about their lives. The protocol has international interoperability built in and prevents attack from hostile groups or state actors.
The protocol also supports other sample payloads. You can plug in your own for your specific application needs. See Custom Outer Payloads for details.
We found a lot of issues in mobile phone Bluetooth implementations and have worked around a great many of them. You can read about our Bluetooth discoveries and workarounds here.