Table 1 Summary of key studies in smart parking systems.
Ref. and approach | Proposed architecture | Key technologies | Limitations/gaps |
|---|---|---|---|
Multi-tier architectures for pricing structures and contract management | Conceptual pricing models, usability frameworks | Primarily conceptual with limited real-time implementation details | |
Models for forecasting parking availability and optimal resource allocation | Predictive analytics, usability approaches | Focused on urban environments; lack localised contextual testing | |
Broader IoT Applications31 | IoT-based system for intelligent transportation (dynamic ride-sharing, data security) | IoT integration, edge/fog computing | Designed for large-scale urban networks; not tailored for smaller, controlled contexts |
Middleware-Based Integrated Approach32 | Loosely coupled middleware integrating AVP and smart parking systems for both automatic and manual vehicles | Middleware integration, Markov Chain model for usability evaluation | Not evaluated in localised campus settings; focused on overall cost, space availability, and traffic reduction |
Three-tier system (Awaisi et al.) with LED displays and multi-area fog nodes. Five-tier fog computing model (Thangam et al.) | Image processing, sensor networks, fog computing | Limited evaluation in non-urban, localised environments | |
Real-Time Tracking & Dynamic Pricing36 | Multi-layered intelligent parking system leveraging resilient edge and cloud computing with dynamic pricing | Real-time tracking, low-power wireless communications, dynamic pricing | Primarily urban focus; challenges in applying to smaller, controlled settings |
IoT-based smart parking systems with sensor nodes, fog computing, reservations, and camera-based real-time image processing | Sensor networks, fog computing, real-time image processing | Evaluated in urban scenarios; scalability and adaptation to university environments remain unverified | |
Our Study | Fog-enabled IoT smart parking system tailored for SDU University | Fog computing, local real-time decision-making, reduced reliance on centralised cloud servers | Addresses irregular campus parking demand and limited data centre capacity |
