Advanced Meta heuristic Strategies for Load Balancing in 5G C-RAN Environments

With the rapid expansion of internet usage and continuous technological progress, mobile network operators are compelled to enhance their investments in network infrastructure. Emerging technologies such as Cloud Radio Access Networks (C-RAN) and Software Defined Networking (SDN) are increasingly viewed as viable solutions to lower operational costs and improve scalability in fifth-generation (5G) mobile networks. A typical base station comprises two critical components: the baseband unit (BBU) and the remote radio head (RRH). Variations in data traffic can lead to network inefficiencies, including issues like call drops and call blocking. As traffic patterns fluctuate, system performance may degrade if not properly managed. To address this, self-optimizing network strategies are essential for redistributing the load from heavily burdened eNodeBs, which experience high call blocking rates, to underutilized ones with spare capacity. The primary goal of these self-organizing networks is to balance the traffic load and minimize call blocking incidents. In this work, an improved version of the Cat Swarm Optimization (CSO) algorithm—referred to as Enhanced Cat Swarm Optimization (ECSO)—is introduced. Managed by the host controller, ECSO identifies the optimal BBU-RRH pairings by assessing quality-of-service (QoS) metrics from various configurations. The optimization process evaluates each user connection by analyzing QoS data for every potential BBU-RRH combination. Simulation outcomes indicate that ECSO outperforms existing Particle Swarm Optimization (PSO) and standard CSO methods by reducing blocking probability by 10%, increasing throughput by 8%, and decreasing response time by 7%.