Dynamic channel assignment for large-scale cellular networks using noisy chaotic neural network.

Area of Science:

• Electrical Engineering
• Computer Science
• Telecommunications

Background:

• Large-scale cellular networks (LCNs) face challenges in dynamic channel assignment (DCA) due to signaling overhead and computational load.
• Efficient spectrum utilization is critical for managing increasing mobile data demands.

Purpose of the Study:

• To propose a novel DCA technique for LCNs using a noisy chaotic neural network.
• To enhance spectrum utilization and reduce interference in cellular networks.

Main Methods:

• Decomposition of LCNs into smaller, manageable decomposed cellular subnets (DCSs).
• Independent DCA within each DCS to reduce signaling and computational complexity.
• Formulation of a novel energy function to prevent inter-subnet interference using a real-time interference channel table.
• Optimization of channel assignments within each DCS to satisfy interference constraints and minimize total channels used.

Main Results:

• Demonstrated validity of the proposed DCA technique in a 441-cell LCN with 70 channels, decomposed into nine 49-cell DCSs.
• Analysis of blocking probability under uniform and hot spot traffic patterns.
• Significant improvement in spectrum utilization through minimized channel assignments.

Conclusions:

• The proposed noisy chaotic neural network-based DCA technique effectively manages channel assignment in LCNs.
• Decomposition into DCSs alleviates signaling overhead and computational load.
• The novel energy function successfully minimizes interference and optimizes spectrum usage.