Cloud Post-Processing

Problem Statement

Calibration is to check, adjust, or determine by comparison with a standard value. Sensor calibration adjustment is necessary for the Dissolved Oxygen, Turbidity, and TDS sensors. Introducing this adjustment introduces out of scope complexities to the microcontroller (Arduino Zero).

Suggested Solution

Considering the scope of the current project, our solution involves post-processing raw sensor values on the cloud. The microcontroller would send raw values from the sensors and the cloud would post-process those values.

Microcontroller Complexities

To detail the aforementioned complexities, the smart device team has strived to create a uniform, reusable C environment. In C, we have created a sensor object that encapsulates all sensors.

struct sensor
{
    int pin;
    float value_adc;
    unsigned int type;
    float value_output;
    float * temperature;

    Measure measure;
    Convert convert;
};

Introducing unique calibration functions expands the code base by file size and required understanding by future teams. Furthermore, if YVR intends to add a greenhouse gas sensor, we have decided adding a sensor would be easier by separating the sensor's microcontroller implementation and the post-processing.

Action Items

• Pass on post-processing functions to cloud team from current microcontroller C environment.
• Implement post-processing for AWS.