Skip to content
This repository has been archived by the owner on Oct 11, 2023. It is now read-only.

Device Models

Peter Felts edited this page Mar 28, 2019 · 5 revisions

Each Simulated Device belongs to a specific Device Model, which defines the simulation behavior, for example how frequently telemetry is sent, what kind of messages to send, which methods are supported, etc.

Device models are defined using a set of JSON configuration files, (one file for each device model), and a set of **Javascript files defining the simulation behavior, e.g. the random telemetry, the method logic, etc. A typical simulation service should have:

  • One JSON file for each device model (e.g. elevator.json)
  • One Javascript file for each device model (e.g. elevator-state.js)
  • One Javascript file for each device method (e.g. elevator-go-down.js)

The DeviceModels folder contains some examples of these files, showing how to define a device model and how to simulate a device behavior.

Device Model JSON example

{
    "SchemaVersion": "1.0.0",
    "Id": "elevator-01",
    "Version": "0.0.1",
    "Name": "Elevator",
    "Description": "Elevator with floor, vibration and temperature sensors.",
    "Protocol": "AMQP",
    "Simulation": {
        "InitialState": {
            "online": true,
            "floor": 1,
            "vibration": 10.0,
            "vibration_unit": "mm",
            "temperature": 75.0,
            "temperature_unit": "F"
        },
        "Interval": "00:00:10",
        "Scripts": [
            {
            "Type": "javascript",
            "Path": "elevator-01-state.js"
            }
        ]
    },
    "Properties": {
        "Type": "Elevator",
        "Location": "Building 40",
        "Latitude": 47.636369,
        "Longitude": -122.133132
    },
    "Telemetry": [
        {
            "Interval": "00:00:10",
            "MessageTemplate": "{\"floor\":${floor},\"vibration\":${vibration},\"vibration_unit\":\"${vibration_unit}\",\"temperature\":${temperature},\"temperature_unit\":\"${temperature_unit}\"}",
            "MessageSchema": {
                "Name": "elevator-floor;v1",
                "Format": "JSON",
                "Fields": {
                    "floor": "integer"
                }
            }
        }
    ],
    "CloudToDeviceMethods": {
        "StopElevator": {
            "Type": "javascript",
            "Path": "TBD.js"
        },
        "StartElevator": {
            "Type": "javascript",
            "Path": "TBD.js"
        }
    }
}

Device Model Javascript simulation example

/*global log*/
/*global updateState*/
/*global updateProperty*/
/*jslint node: true*/

"use strict";

var floors = 15;

// Default state
var state = {
    online: true,
    floor: 4,
    vibration: 10.0,
    vibration_unit: "mm",
    temperature: 75.0,
    temperature_unit: "F",
    moving: false
};

// Default properties
var properties = {};

/**
 * Restore the global state using data from the previous iteration.
 *
 * @param previousState device state from the previous iteration
 * @param previousProperties device properties from the previous iteration
 */
function restoreSimulation(previousState, previousProperties) {
    // If the previous state is null, force a default state
    if (previousState) {
        state = previousState;
    } else {
        log("Using default state");
    }

    if (previousProperties) {
        properties = previousProperties;
    } else {
        log("Using default properties");
    }
}

/**
 * Simple formula generating a random value around the average
 * in between min and max
 */
function vary(avg, percentage, min, max) {
    var value =  avg * (1 + ((percentage / 100) * (2 * Math.random() - 1)));
    value = Math.max(value, min);
    value = Math.min(value, max);
    return value;
}

function varyfloor(current, min, max) {
    if (current === min) {
        return current + 1;
    }
    if (current === max) {
        return current - 1;
    }
    if (Math.random() < 0.5) {
        return current - 1;
    }
    return current + 1;
}

/**
 * Entry point function called by the simulation engine.
 * Returns updated simulation state.
 * Device property updates must call updateProperties() to persist.
 *
 * @param context             The context contains current time, device model and id
 * @param previousState       The device state since the last iteration
 * @param previousProperties  The device properties since the last iteration
 */
/*jslint unparam: true*/
function main(context, previousState, previousProperties) {

    // Restore the global device properties and the global state before
    // generating the new telemetry, so that the telemetry can apply changes
    // using the previous function state.
    restoreSimulation(previousState, previousProperties);

    if (state.moving) {
        state.floor = varyfloor(state.floor, 1, floors);
        // 10 +/- 5%,  Min 0, Max 20
        state.vibration = vary(10, 5, 0, 20);
    } else {
        state.vibration = 0;
    }

    // 75 +/- 1%,  Min 25, Max 100
    state.temperature = vary(75, 1, 25, 100);

    updateState(state);
}

Thanks to these files, it is possible to add new device models and customize the simulation behavior, without rebuilding the service. To add new types or to change the simulated behavior, edit the relevant *.json and *.js files and redeploy the device simulation microservice.

Each device model file contains the definition of the simulated device model, including the following information:

  • Device model name: string
  • Protocol: AMQP | MQTT | HTTP
  • The initial device state
  • How often to refresh the device state
  • Which Javascript file to use to refresh the device state
  • A list of telemetry messages to send, each with a specific frequency
  • The schema of the telemetry messages, used by backend application to parse the telemetry received
  • A list of supported methods and the Javascript file to use to simulate each method.

Check the DeviceModels folder for some examples.

Device Model files

File schema

The schema version is always "1.0.0" and is specific to the format of this file:

"SchemaVersion": "1.0.0"

Device model description

The following properties describe the device model. Each type has a unique identifier, a semantic version, a name and a description:

"Id": "chiller-01",
"Version": "0.0.1",
"Name": "Chiller",
"Description": "Chiller with external temperature, humidity and pressure sensors."

IoT Protocol

IoT devices can connect using different protocols. The simulation allows to use either AMQP, MQTT or HTTP:

"Protocol": "AMQP"

Simulated device state

Each simulated device has an internal state, which needs to be defined. The state also defines the properties that can be reported in telemetry. For example an elevator might have an initial state like:

"InitialState": {
    "floor": 1
},

while a moving device with multiple sensors might have more properties, like:

"InitialState": {
    "latitude": 47.445301,
    "longitude": -122.296307,
    "speed": 30.0,
    "speed_unit": "mph",
    "cargotemperature": 38.0,
    "cargotemperature_unit": "F"
}

The device state is kept in memory by the simulation service, and provided in input to the Javascript function. The Javascript function can decide to ignore the state and generate some random data, or to update the device status in some realistic way, given a desired scenario.

The function generating the state receives in input also the Device Id, the Device Model and the Current Time, so it is possible to generate different data by device and by time if required.

Generating telemetry messages

The simulation service can send multiple messages for each device, and each message can be sent at a different frequency. Typically, a telemetry will send a message including some data from the device state. For example, a simulated room might send information about temperature and humidity every 10 seconds, and lights status once per minute. Note the placeholders, which are automatically replaced with values from the device state:

"Telemetry": [
    {
        "Interval": "00:00:10",
        "MessageTemplate":
            "{\"temperature\":${temperature},\"temperature_unit\":\"${temperature_unit}\",\"humidity\":\"${humidity}\"}",
        "MessageSchema": {
            "Name": "RoomComfort;v1",
            "Format": "JSON",
            "Fields": {
                "temperature": "double",
                "temperature_unit": "text",
                "humidity": "integer"
            }
        }
    },
    {
        "Interval": "00:01:00",
        "MessageTemplate": "{\"lights\":${lights_on}}",
        "MessageSchema": {
            "Name": "RoomLights;v1",
            "Format": "JSON",
            "Fields": {
                "lights": "boolean"
            }
        }
    }
],

The placeholders use a special syntax ${NAME} where NAME is a key from the device state object returned by the Javascript main() function. Note that strings should be quoted properly, while numbers should not.

Message schema

Each message type must have a well defined schema. The message schema is also published to IoT Hub, so that backend applications can reuse the information to interpret the incoming telemetry.

The schema supports JSON format, which allows for easy parsing, transformation and analytics, across several systems and services.

The fields listed in the schema can be of the following types:

  • Object - serialized using JSON
  • Binary - serialized using base64
  • Text
  • Boolean
  • Integer
  • Double
  • DateTime

Supported methods

Simulated devices can also react to method calls, in which case they will execute some logic and provide some response. Similarly to the simulation, the method logic is stored in a Javascript file, and can interact with the device state. For example:

"CloudToDeviceMethods": {
    "Start": {
        "Type": "javascript",
        "Path": "truck-start.js"
    }
}

Function script files

Functions are stored in Javascript files, which are loaded and executed at runtime, using Jint, a Javascript interpreter for .NET.

The Javascript files must have a main function, and accept three parameters:

  1. a context object which contains the following properties:
    1. currentTime as a string with format yyyy-MM-dd'T'HH:mm:sszzz
    2. deviceId, e.g. "Simulated.Elevator.123"
    3. deviceModel, e.g. "Elevator"
  2. a previousState object, which is the value returned by the function in the previous call. This device state is maintained by the simulation service, and used to generate telemetry messages.
  3. a previousProperties object, which is the value that the Simulation service is storing at the time of the function call. This object is stored by the Simulation service and is sent to the IoT Hub as a device-twin update.

The main function can update both state and properties via calls to updateState and updateProperty, respectively. Example:

/*global log*/
/*global updateState*/
/*global updateProperty*/
/*global sleep*/
/*jslint node: true*/

"use strict";

/**
 * Entry point function called by the simulation engine.
 *
 * @param context        The context contains current time, device model and id
 * @param previousState  The device state since the last iteration
 * @param previousProperties  The device properties since the last iteration
 */
/*jslint unparam: true*/
function main(context, previousState, previousProperties) {

    log("Starting 'EmergencyValveRelease' JavaScript method simulation.");

    var state = {   
        simulation_state: "normal_pressure",
        pressure: 150
    };
    updateState(state);

    log("'Emergency Valve Release' JavaScript method simulation completed.");
}

Debugging script files

While it's not possible to attach a debugger to the Javascript interpreter, it's possible to log information in the service log. For convenience, the application provides a log() function which can be used to save information useful to track and debug the function execution. In cases of syntax errors, the interpreter will fail, and the service log will contain some information about the Jint.Runtime.JavaScriptException exception occurred.

Logging example:

function main(context, previousState, previousProperties) {

    log("This message will appear in the service logs.");

    log(context.deviceId);

    if (typeof(state) !== "undefined" && state !== null) {
        log("Previous value: " + state.temperature);
    }

    // ...

    updateState(state);
}