Hybrid GA-PSO Optimization for Controller Placement in Large-Scale Smart City IoT Networks

Area of Science:

• Computer Science
• Network Engineering
• Smart City Technology

Background:

• The Internet of Things (IoT) is integral to smart city development.
• Optimal controller placement is critical for IoT network performance and resilience.
• Existing placement strategies may not adequately address the complexities of large-scale smart city networks.

Purpose of the Study:

• To propose a hybrid optimization approach for strategic controller placement in IoT networks.
• To enhance network performance metrics including latency, load balancing, energy efficiency, and redundancy.
• To validate the proposed algorithm in a real-world smart city model.

Main Methods:

• A hybrid optimization algorithm combining Genetic Algorithm (GA) and Particle Swarm Optimization (PSO) was developed.
• A large-scale Narrowband Internet of Things (NB-IoT) network simulation with 2000 nodes was conducted.
• Controller placement was optimized to minimize latency and balance load, focusing on energy efficiency and redundancy.

Main Results:

• The hybrid GA-PSO algorithm demonstrated superior performance compared to random and K-Means clustering placements.
• Significant improvements were observed in load balancing, reduced packet loss, and enhanced energy efficiency.
• The algorithm proved robust and effective across normal operation, node failure, and traffic spike scenarios.

Conclusions:

• The proposed hybrid GA-PSO algorithm offers an effective solution for optimizing controller placement in smart city IoT networks.
• This approach enhances network resilience, scalability, and overall performance.
• The findings underscore the algorithm's potential for practical implementation in smart city infrastructure.