Solar-Powered IoT Smart Bus Stops

• IoT-based solar-powered smart bus stops are autonomous nodes featuring photovoltaic panels, battery storage, and microcontroller-driven displays for real-time transit updates.
• They employ a robust energy management system—with MPPT controllers, Arduino Mega, and Wi-Fi modules—to ensure 24-hour off-grid operation and 100% uptime.
• The integrated communications use a star topology linking bus and bus-stop nodes to a cloud server, enabling efficient passenger tracking and low-latency data refresh.

IoT-based solar-powered smart bus stops are distributed, autonomous infrastructure nodes designed to provide real-time transit information while operating independently from the electrical grid. Architected to form a critical element in energy-efficient smart public transport management, these systems combine photovoltaic power generation, microcontroller-based information kiosks, and networked IoT components to achieve high reliability, low energy consumption, and seamless integration with vehicle telematics (Haque et al., 3 Jan 2026).

1. System Configuration and Hardware Architecture

Each smart bus stop consists of three principal hardware subsystems: power generation/storage, computational/display units, and communications.

Photovoltaic (PV) energy is harvested with a 20 W monocrystalline panel (nominal 18 V), linked via a pulse-width modulation (PWM) maximum power point tracking (MPPT) charge controller to a 12 V, 5 Ah lead-acid or LiFePO₄ battery. This configuration ensures overnight and low-irradiance autonomy. Downstream power conditioning stabilizes and down-converts the voltage to 5 V rail for logic devices.

The local computational core is an Arduino Mega (ATmega2560), executing the display firmware and coordinating peripheral activity. Key peripherals include:

• 0.96″ I²C OLED (SSD1306, 128×64 px) for displaying live seat-count and system time
• DS3231 real-time clock (RTC) for timestamped updates
• ESP-01 (ESP8266) Wi-Fi module for communication

All digital logic and display components draw power from the regulated 5 V bus. Table 1 aggregates the main hardware elements:

Subsystem	Component	Typical Power
PV Generation	20 W monocrystalline panel	—
Energy Storage	12 V, 5 Ah battery (lead/LiFePO₄)	—
MCU & Display	Arduino Mega, OLED, RTC	<2.5 W
Communications	ESP-01 Wi-Fi module	<1.0 W

2. Energy Management and Sustainability Metrics

The energy model involves quantifying device-level consumption per bus stop and provisioning sufficient PV and battery capacity for 24-hour autonomous operation. The daily system load is calculated:

$$E_\text{load} = \sum_i P_i \times t_i = 0.48\,\mathrm{Wh} + 2.4\,\mathrm{Wh} + 0.48\,\mathrm{Wh} + 0.0024\,\mathrm{Wh} + 0.1\,\mathrm{Wh} \approx 3.46\,\mathrm{Wh}\ (\text{min})$$

scaling up to

$\approx 34.82\,\mathrm{Wh}\ (\text{max})$

Assuming 5 peak-sun hours, the panel sizing is: $$P_\text{PV} = \frac{E_\text{load}}{H_\text{sun}} \implies 0.69\,\mathrm{W}\ (\text{min}),\ 6.96\,\mathrm{W}\ (\text{max})$$ which motivates the standardization on a 10 W (or 20 W) panel.

Battery sizing for night operation yields: $$C_\text{bat} = \frac{E_\text{load}}{V_\text{bat} \times \mathrm{DoD}}$$ with $\mathrm{DoD} = 0.8$ (max. safe depth of discharge), translating, e.g., $E_\text{min}\rightarrow 0.36\,\mathrm{Ah}$ up to $2.89\,\mathrm{Ah}$.

The state-of-charge (SoC) is maintained within bounds: $$\mathrm{SoC} = \frac{E_\text{bat,remaining}}{E_\text{bat,max}}\in [\mathrm{SoC}_\text{min},\,1.0]$$

Annual grid offset, at maximal load, is: $$34.824\,\mathrm{Wh/day} \times 365 = 12710.76\,\mathrm{Wh/yr} \approx 12.71\,\mathrm{kWh/yr}$$ per stop.

3. IoT Communications Protocols and Data Topology

System connectivity implements a star topology wherein both buses and bus stops act as independent IoT nodes interfacing with a central cloud server.

• Bus Node: A Raspberry Pi 4 B runs the RFID-based passenger-counting system, gathering “card-in/card-out” events and geolocation streams. Event data and GPS coordinates are transmitted as JSON payloads via HTTP(S) POST to a Node.js back-end REST API.
• Cloud Server: Receives, persists, and processes updates. Data formats include:

  { "bus_id": ..., "timestamp": ..., "seat_count": ..., "latitude": ..., "longitude": ... }

  PostgreSQL or MongoDB stores time-series data.
• Bus-Stop Node: Upon receiving WebSocket or HTTP-push messages from the server, the ESP-01 updates the local microcontroller. The Arduino parses JSON and refreshes the OLED display for passenger visibility.

The React-based user frontend polls the API for live data (≈1 Hz) and visualizes bus positions via Google Maps API, while an admin portal presents passenger-flow analytics and system uptime.

4. Functional and Energy Performance

Field measurements during a 30-day continuous mission documented 100% uptime for the solar-backed bus stops, with no grid fallback events, substantiating the system’s off-grid sustainability claims.

The annualized maximum grid offset for a single stop is quantified as 12.71 kWh/yr, equivalent to approximately 1.5 L diesel offset per stop annually.

RFID-enabled passenger-count accuracy exceeded 99% across 1,000 board/alight events.

Integration with the vehicle-based YOLOv4-Tiny + SONAR blind-spot warning engine achieved:

• Detection precision: 98.8%
• Recall: 93.6%
• $F_1$: 96.1% at 15 fps on the Pi 4B platform.

The live location information system delivered 1 Hz refresh cycles with mean end-to-end latency of 200 ms. Bench tests established the HTTP server’s ability to serve 500 simultaneous connections with less than 5% CPU load on a 2-core VM.

5. Block-Level Schematics and Data Flow

The bus stop's energy and data-path architecture are delineated as follows:

[A] Energy Subsystem:

PV Panel ──> MPPT Charger ──> 12 V Battery ──┐ ├─> 12 V Bus ──> 5 V DC-DC ──> Arduino Mega │ • OLED Display │ • RTC DS3231 │ • 5 mm Status LEDs └─> ESP-01 (Wi-Fi)

[B] IoT Data Flow:

Raspberry Pi (Bus) • RFID → seat_count • GPS → lat/lon • HTTP(S) POST ↘ Central Server (Node.js + DB) • WebSocket Push ↗ ESP-01 (Bus-Stop) → Arduino → OLED

This architecture enables zero grid dependency, real-time passenger availability, and vehicular tracking for waiting passengers, with all core subsystems drawing under 3.5 W average load from solar input.

6. Contextual and Implementation Considerations

The coordinated hardware and data stack outlined supports seamless integration into existing public transit systems, particularly in settings with unreliable electrical grids. The empirical demonstration of 12.71 kWh/year energy savings, >99% RFID read accuracy, and sub-second end-to-end user information update substantiates its operational suitability (Haque et al., 3 Jan 2026).

The bus stop platforms are engineered for modular scalability; each node operates independently, allowing for incremental network expansion or redundancy. These capabilities position such systems as foundational components in sustainable urban mobility infrastructures.