IoT Smart Locker System | Remote Access & Monitoring

The Smart Locker Box Management System is an IoT-based platform designed to manage small cabinet lockers through a connected ecosystem of hardware sensors, embedded firmware, and a cloud-based management dashboard. The system enables remote monitoring, access control, and real-time status tracking for locker units deployed across multiple locations.

Built for facility managers, coworking spaces, logistics hubs, and parcel delivery networks, the platform replaces manual key-based locker management with a fully digital, sensor-driven workflow.

Traditional locker systems rely on physical keys or static PIN codes, creating several operational pain points:

No visibility into locker occupancy or usage patterns across locations

Manual intervention required for every lock/unlock event, lost key, or code reset

Zero audit trail for access events, making accountability difficult

No remote management capability for facility operators managing multiple sites

Scaling friction when adding new locker units or locations to the network

The client needed a system that could bring real-time intelligence to their locker infrastructure while keeping the hardware footprint small enough for compact cabinet-style units.

Each locker unit is equipped with:

Electromagnetic lock module controlled via microcontroller (ESP32-based)

Proximity sensor for detecting occupancy status (empty/occupied)

NFC/RFID reader for contactless user authentication at the locker

Status LED indicators for visual feedback (available, occupied, locked, error)

Wi-Fi connectivity module for real-time communication with the cloud backend

The firmware handles local lock/unlock logic, sensor polling, heartbeat signals to the server, and graceful offline fallback when connectivity drops.

The server-side architecture powers all remote operations:

RESTful API built with Node.js and Express for device registration, access control commands, and telemetry ingestion

MQTT broker for lightweight, real-time device-to-cloud messaging (lock events, sensor readings, heartbeats)

PostgreSQL database for structured storage of locker configurations, user access logs, and location hierarchies

Redis caching layer for fast locker status lookups and session management

WebSocket connections for pushing live status updates to the admin dashboard

A responsive web dashboard built with React.js gives facility managers full control:

Real-time locker grid view showing occupancy, lock state, and health status across all locations

Remote lock/unlock controls with confirmation dialogs and audit logging

User management with role-based access (admin, operator, viewer)

Access history timeline with filterable logs per locker, user, or location

Alert system for hardware faults, connectivity drops, unauthorized access attempts, and maintenance reminders

Analytics dashboard with occupancy heatmaps, peak usage hours, and utilization rates per location

Multi-location management with site grouping and bulk operations

End users interact with lockers through a companion mobile application:

QR code / NFC tap to unlock assigned lockers

Real-time notifications for locker assignments, access confirmations, and expiry reminders

Booking system for reserving lockers in advance with configurable time slots

Access history showing personal usage logs

[Locker Hardware (ESP32 + Sensors)] | [MQTT / Wi-Fi] | [MQTT Broker] ←→ [Node.js API Server] ←→ [PostgreSQL + Redis] | ┌───────────┼───────────┐ | | | [React Dashboard] [Mobile App] [Webhook Integrations]

MQTT over HTTP for device communication: MQTT's lightweight publish/subscribe model is ideal for constrained IoT devices. It reduces bandwidth, supports QoS levels for reliable delivery, and handles intermittent connectivity gracefully.

ESP32 as the microcontroller: Cost-effective, built-in Wi-Fi/Bluetooth, sufficient GPIO pins for lock control and sensor inputs, and strong community support for OTA firmware updates.

PostgreSQL for structured data: Access logs, user permissions, and location hierarchies benefit from relational integrity and complex query support.

Redis for real-time status: Locker state lookups need sub-millisecond response times for the dashboard grid view. Redis serves as the single source of truth for current device state.

WebSockets for dashboard updates: Polling hundreds of lockers would be wasteful. WebSocket connections push state changes to connected admin clients instantly.

Lockers deployed in basements or areas with spotty Wi-Fi needed to function even when disconnected from the cloud.

Solution: Implemented local access control logic on the ESP32 firmware. The device caches the last-known authorized user list and processes NFC/RFID authentication locally. When connectivity resumes, queued events sync to the server automatically.

Physical security systems demand tamper-resistant authentication to prevent unauthorized access.

Solution: Implemented end-to-end encrypted communication between devices and the server using TLS. NFC tokens use rolling codes rather than static IDs. The API enforces time-bound access tokens that expire after the reservation window closes.

The system needed to support hundreds of locker units across multiple physical sites without degrading dashboard performance.

Solution: Designed the backend with location-scoped data partitioning. MQTT topics are namespaced by site, and the dashboard lazy-loads locker grids per selected location. Redis handles the hot path for status queries, keeping PostgreSQL reserved for historical and analytical workloads.

100% remote management capability, eliminating on-site visits for routine lock/unlock operations

Real-time visibility across all deployed locker units from a single dashboard

Sub-second lock response time from user tap to electromagnetic lock release

Automated alerting reduced hardware downtime by catching connectivity and sensor faults early

Scalable architecture supporting multi-site deployments with independent location management

I led the full-stack development of this project, covering:

System architecture design spanning IoT hardware integration, cloud backend, and client applications

Backend API development with Node.js, MQTT broker configuration, and database schema design

Frontend dashboard development with React.js, including real-time data visualization

IoT device communication protocol design and firmware integration planning

Security implementation including encrypted device communication and token-based access control

UI/UX design for both the admin dashboard and the mobile companion app