The AMT platform is a multi-tenant PaaS and SaaS IOT system. The platform includes a gateway, a sensor network and a rule-based reasoning system. The platform was designed for Digital Asset Management but it can be used in other areas. Things and Events are described through a standards-based DAP (Domain Specific Protocol), which enables automatic discovery and assembly of new elements in the system.
At run-time the cloud side of AMTech runs in several machine clusters, leveraging multiple technologies, including Storm Analytics, Titan/Cassandra GraphDB, and Kafka Queues. The gateway can run in many locations, including a Raspberry Pi3. AMT is highly scalable and maintains end-to-end security.
Events are called Observations and are organized in Topics (akin to MQTT topics)
Things have properties. A few properties are special, either managed by the system (detection time) or interpreted by the system (location).
The system has built-in knowledge of location and proximity. Rules can ask for conditions based on them, and reverse geolocation provides data on addresses, et al.
Rules are called Reasoners. A Reasoner is defined by an “Observer” that analyzes Observations and Collections of Things, and can execute a Sequence of Actions based on the result of invoking the observer (e.g. Existence, Absence, etc). The actions can perform CRUD operations on things, send new observations (back into the system), send notifications (for external use), or send Commands (to things).
Rules are organized into Activites. Activities are the unit of deploy and un-deploy and of access control. Access control roles are called Actors.
A typical arrangement for AMT configuration is to have a Creator tenant that will define Activites (the PaaS side of AMT) and a Follower tenant that can be used to deploy the system (the SaaS side of AMT). Each tenant can have multiple users, with different Actor capabilities.
AMTech’s Web-based UX can be used to configure tenants, define users, define and test rules and things. Most of the concepts of the system have corresponding URIs, and many actions can be done through secure REST operations (e.g. through Postman or CURL).
Observations are consumed by the system through REST and WebSockets and can originate from different sources. AMTech’s gateway, the M2MBridge, is a node.js based software gateway with a flexible plugin-sytem that supports a variety of protocols, from iBeacons to RFID.
AMTech provides end-to-end security. Actors are used to define the capabilities of different participants, including sensors, and all Things and Observations are typed, using RDF.