Repository for wireless sensor data measurement system.
-
Goal: Implementation of a standalone wireless sensor data measurement system. Measurement system will do FSA and HHT + HSA and other time series analysis on raspberry Pi for processing real time sensor data (Updating graphs every 1-5 minutes).
-
For first phase of the measurement system, a temperature-pressure sensor (Bosch BMP280) with a rpi PicoW board will be used for wirelessly transmitting data to a server (using MQTT/HTTP).
-
Initial data logging system will be based on CSV files, later logging systems will move from CSV to SQL/Time Series DB (Postgre SQL / Influx DB)
-
Backend logic for FSA, HSA and other Time-series data analysis functions will be implemented in Python/JavaScript.
-
Initial data manipulation and visualization will be done using Python and Jupyter Notebooks
-
GUI dashboard implementation will be done on Rpi400 for interaction with the measurement system (Grafana based) in the later prototypes 2,3 ....
-
Experiments with system architechture will be done by trying and changing communication protocols- (MQTT, HTTP Restful APIs), sensor data logging system (CSV, PostgreSQL, InfluxDB), Data Visualization System (Jupyter notebooks, Grafana Dashboards, PowerBI, HTML/React based GUI)
- Implementation of the measurement system backend logic (With reduced functionalities) on inexpensive $10-20 computer like ESP32 / RP PicoW / RPzeroW2 in MicroPython/Arduino framework. (Edge Computing)
- Running the board on battery source and visualization of results via Mobile device using a Wifi Connection
- Implementation of the data visualization system using React (Native)
- Implementation of the whole Measurement system using containerization with Docker.
- Assigning Static IP address to the MQTT/HTTP Server
- Building a custom PCB for sensor data acquisition, logging, computing algorithms, and visualization of sensor data using Mobile device with WiFi connection.
- Cyber Security -- Password protection, Hardware encryption, Sensor data encryption using ECC(Elliptic Curve Cryptography) / other suitable crytographic algorithms for IoT devices.
- Upload of telemetry data to the cloud account/remote server.
Prtotype Hardware Device I
- PCB with ESP32/RP2040 | (Wroom/Wrover module), ESP32 S/C/P series, PicoW board
- UART-USB bridge | (CP2102)
- Inbuilt accelerometer/IMU unit | (SPI/I2C)
- Inbuilt temperature, humidity, pressure sensor | (Bosch BME280, BMP280 series) | (I2C)
- Inbuilt LDR
- Inbuilt Flash/EPROM storage + removable memory card support | (SPI)
- Option for programming the chip (RP2040) via SWD and USB| (USB to UART bridge) | CP2102
- Native USB support, USB OTG, WiFi and BLE capabilities | ESP32 S series chips
- Power supply via USB and 3.3V cheap Li-ion battery.
- Charging circuit for Li-ion battery
- Programming via USB interface using MicroPython/Arduino framework/ Embedded C
- Future boards to have LoRaWAN communication capabilities by adding the module/SOC | Semtech Sx1276 series chips
- In depth study of the variants of EMD/HHT
- In depth study of time and memory complexity of HHT and its variants
- Implementation of advance variants of HHT/EMD/HSA algorithms in a) Python b) C++
- First sandbox to be in Jupyter Notebooks --> Python Modules
- Developing Python libraries for implementation of HHT and its variants -- EEMD, MEMD, HHSA