REDUCTION AND CONTROL CONGESTION IN MULTI-TIER 4G-LTE-A NETWORK USING HYBRID ARTIFICIAL INTELLIGENCE TECHNIQUE

Subject Area: Engineering

This paper presents the reduction and control of congestion in multi-tier 4G-LTE-A network using hybrid artificial intelligence technique. The study addressed the crucial requirement of maintaining Quality of Service (QoS) in 4G networks to support real-time multimedia applications and seamless packet streaming in wireless networks.The paper identified the need to address congestion, considering multiple congestion constraints, including throughput, loss, and load factor. To overcome this challenge, a Genetic Algorithm (GA) was integrated with the Datagram Congestion Control Protocol (DCCP) for congestion detection and control. The GA-DCCP approach effectively detected congestion and dynamically adjusted the data reception rate to prevent overload, thus mitigating congestion-related issues. The result of simulation and evaluation of the model presented that the effectiveness of the proposed GA-DCCP approach, and achieved average throughput of 89.03% during testing which showcased improved data transfer efficiency. Furthermore, the validation results indicated a significant 26.06% improvement compared to the network without GA-DCCP, confirming the efficacy of the proposed solution.