Thanks to visit codestin.com
Credit goes to www.scribd.com

Scribd
Upload
28 views1 page

ML4IoT HW3 Report

The document outlines a project involving the collection, communication, and storage of temperature and humidity data using a DHT-11 sensor and Raspberry Pi, utilizing MQTT for real-time data transmission and Redis for data storage. A REST API was also developed for retrieving historical data, emphasizing the advantages of MQTT over REST for IoT applications due to its efficiency and lightweight nature. The project successfully demonstrated effective data management and visualization techniques, meeting the outlined requirements.

Uploaded by

marcomungaicoppolino
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
28 views1 page

ML4IoT HW3 Report

The document outlines a project involving the collection, communication, and storage of temperature and humidity data using a DHT-11 sensor and Raspberry Pi, utilizing MQTT for real-time data transmission and Redis for data storage. A REST API was also developed for retrieving historical data, emphasizing the advantages of MQTT over REST for IoT applications due to its efficiency and lightweight nature. The project successfully demonstrated effective data management and visualization techniques, meeting the outlined requirements.

Uploaded by

marcomungaicoppolino
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 1

Machine Learning for IoT - Report Homework 3

Giuseppe Acquaviva, Mario Capobianco, Marco Mungai Coppolino


Politecnico di Torino, Italy

I. E XERCISE 1 - DATA C OLLECTION , C OMMUNICATION , B. REST Client for Data Visualization


AND S TORAGE A REST client was developed in rest_client.ipynb
A. Implementation of Data Collection using MQTT to interact with the API:
• Server status was verified using GET /status.
The primary goal of this task was to collect temperature and
• Sensor node was added using POST /sensors.
humidity data using a DHT-11 sensor connected to a Raspberry
• Data was retrieved and visualized with Deepnote chart
Pi and publish the data to an MQTT broker. The following
steps were performed: blocks.
• The script publisher.py was developed to measure C. HTTP Method Justification
and publish data every 2 seconds. The most suitable HTTP method for implementing the
• JSON messages were structured with fields: MAC ad- /data/{mac_address} endpoint is GET because:
dress, timestamp, temperature, and humidity. • Read-only operation: GET is designed for retrieving
• The Eclipse MQTT broker was used as a communication
data without altering server resources, aligning with the
endpoint. endpoint’s purpose.
• Idempotency: Multiple GET requests with the same pa-
B. Data Storage using Redis
rameters will yield the same response, ensuring consistent
To store the data: and predictable behavior.
• An MQTT subscriber was implemented in • Unsuitability of other methods: POST is typically used
subscriber.ipynb. to create resources, while PUT and DELETE involve mod-
• Data was stored in Redis TimeSeries for efficient retrieval ifying or deleting resources, making them inappropriate
and management. for data retrieval.

C. MQTT vs REST for IoT Applications III. C ONCLUSIONS


The tasks highlighted the importance of MQTT for real-time
MQTT is a better choice than REST for this application
IoT communication and REST for structured data retrieval.
because:
The implemented solutions met the project requirements and
• Efficient for frequent updates: MQTT’s demonstrated efficient data management and visualization.
publish/subscribe model eliminates the need for
continuous client polling, reducing communication
overhead.
• Lightweight: MQTT requires minimal resources, making
it ideal for constrained IoT devices like the Raspberry Pi
and DHT-11 sensor.
• Real-time performance: MQTT ensures faster delivery
of data compared to REST, which involves higher latency
due to its request-response model.
In contrast, REST is better suited for less frequent interactions
and when resources are not constrained.

II. E XERCISE 2 - DATA M ANAGEMENT & V ISUALIZATION


A. REST API Design
A REST API was implemented in rest_server.ipynb
to enable retrieval of historical temperature and humidity data.
The endpoint /data/{mac_address} was designed with:
• HTTP GET method for data retrieval.
• Query parameters to specify a date range.
• Response codes for error handling and data availability.

You might also like