Database Schema

This document provides a detailed overview of the MongoDB database schema for the AirZen air quality monitoring system. The schema is defined using Mongoose, an Object Data Modeling (ODM) library for Node.js, which provides a straightforward, schema-based solution to model application data. The database is structured to store information about users, IoT devices, and their associated sensors.

The core collections in the database are User (or TaiKhoan), ThietBi (Device), and CamBien (Sensor). These collections work together to manage user accounts, track physical monitoring devices, and store data from individual sensors.

Entity Relationship Diagram

The diagram below illustrates the relationships between the main collections in the database. A ThietBi (Device) can have multiple CamBien (Sensors) associated with it.

Models

User / TaiKhoan Model

The User model, defined in backend/src/models/User.js, stores user account information, including credentials, roles, and status. There is a similar, less detailed schema named TaiKhoan in backend/src/models/TaiKhoan.js which is imported by some controllers. The User.js schema is more comprehensive and includes advanced features like password hashing and instance methods.

Sources: backend/src/models/User.js, backend/src/controllers/userController.js:5

Schema Definition

Sources: backend/src/models/User.js:10-53

Middleware and Methods

• Password Hashing: A pre('save') middleware automatically hashes the user's password using bcryptjs with a salt round of 12 before saving it to the database. This only runs if the password field has been modified.
  Sources: backend/src/models/User.js:70-84
• comparePassword(candidatePassword): An instance method to compare a plain-text password with the stored hash.
  Sources: backend/src/models/User.js:91-93
• updateLastLogin(): An instance method to update the lastLogin field to the current time.
  Sources: backend/src/models/User.js:96-99
• findActive(): A static method to find all users where isActive is true.
  Sources: backend/src/models/User.js:105-107
• findByEmail(email): A static method to find a user by email, explicitly including the password field for authentication purposes.
  Sources: backend/src/models/User.js:110-112

Indexes

To optimize query performance, the following indexes are defined on the User collection:

• A unique index on email.
• An index on createdAt for sorting by creation date.
• An index on isActive for filtering active users.
  Sources: backend/src/models/User.js:118-120

ThietBi Model (Device)

The ThietBi model represents a physical IoT device. It is defined in the ThietBi collection.

Sources: backend/src/models/ThietBi.js

Schema Definition

Sources: backend/src/models/ThietBi.js:5-20

CamBien Model (Sensor)

The CamBien model represents a sensor that is part of a ThietBi. It is stored in the CamBien collection and is linked to a device via the ID_ThietBi field.

Sources: backend/src/models/CamBien.js

Schema Definition

Sources: backend/src/models/CamBien.js:5-32

Indexes

• An index on ID_ThietBi to speed up lookups for sensors belonging to a specific device.
• An index on TrangThai to efficiently filter sensors by their status.
  Sources: backend/src/models/CamBien.js:34-35

Data Relationships and Usage

The primary relationship is between ThietBi and CamBien, where one device can have many sensors. This is implemented by storing the device's _id in the ID_ThietBi field of each associated sensor document.

The GET /api/thietbi endpoint demonstrates how these two collections are joined using a MongoDB aggregation pipeline. The $lookup stage is used to fetch all camBien documents that match the _id of a ThietBi.

Sources: backend/src/routes/thietBi.js:6-59

The following diagram shows the aggregation flow for retrieving devices along with their sensors.

Sources: backend/src/routes/thietBi.js:11-50