Platform Design Goals
The Infuse-IoT platform is designed in alignment with the following goals.
Machine learning support
Sending all data that a device generates to the cloud is typically undesirable due to power, bandwidth and cloud-compute cost limitations. By enabling machine learning algorithms to be dynamically loaded into a base application, higher-level context information can be extracted from the raw data streams and sent directly, saving both cost and energy.
Non-IP connectivity
Solutions limited by size, weight or cost constraints may choose to develop hardware without built-in Internet-Protocol support. The Infuse platform enables these solutions through multi-hop communications across short range radios, primarily Bluetooth Low Energy.
Intermittent connectivity
Connectivity through your preferred backhaul interface is not always guaranteed to exist, but that doesn’t mean that your devices should stop working. Infuse is designed to gracefully handle intermittent connectivity situations, whether networking is lost for an hour or a year.
Ultra low-power operation
The Infuse platform is focused on ultra low-power devices. By reaching down into the application layer, Infuse can help ensure that your hardware lasts as long as possible between recharges or battery swaps. By providing rich abstractions that handle the most common IoT use cases we minimise the development effort required. And because these abstractions have been battle-tested across multiple applications, you can be confident in software stability from day one.
Device-to-Device communications
Communicating with other local Infuse devices enables another level of data collection and context awareness. Direct links over Bluetooth GATT can enable a higher power device to record data from a more limited peer, or uplink data through an IP-enabled neighbour. Context awareness from nearby devices broadcasting can enable devices without IP connectivity to determine their own location without GNSS, greater temporal resolution in location data, and more.
Hardware Agnostic
The Infuse-IoT platform is intended to be reasonably agnostic to the choice of microcontroller and sensors on the board. Infuse leverages the Zephyr RTOS project to provide the required abstractions that makes this possible.
Observability
Debugging embedded devices can be a challenging endevour. Infuse-IoT aims to provide as much visibility as possible into both the embedded SDK code and the flow of data through the cloud services to simplify this process and monitor standard operations.
Scalable
The Infuse platform is designed to scale from 10 to 10 million devices.
Secure by design
The Infuse platform is committed to enabling organisations to meet security standards such as ETSI EN 303 645 and NIST IR 8425. By integrating industry standard encryption protocols over all communications, we help secure your IoT deployments without compromising on functionality.