Convex Optimized Lagrange Multiplier Based Algebraic Congestion Likelihood for Improved TCP Performance in MANET

In Transmission Control Protocol congestion control method improves throughput fairness. In this paper, A Convex Optimized Lagrange Multiplier-based Algebraic Congestion Likelihood (COLMACL) technique is designed to improve the throughput and to address the congestion control problem in MANET. Initially, Convex Optimized Lagrange Multiplier (COLM) of TCP is used to solve the congestion control problem. The convex optimization is applied in MANET to control the data flow rate. Therefore, an optimal rate of a mobile node depends on improving dynamics on Lagrange Multipliers of its route with queue factor. With the help of queuing factor, inter arrival time is calculated to maintain the data flow in MANET. According to the inter arrival time, the amount of data packet arrived in queue is controlled to minimize the congestion rate. By identifying convex optimized Lagrange Multiplier of a link; if the congestion of a link is positive precisely then the link is fully utilized. Otherwise, congestion of a link is to be zero when the link is underutilized. Therefore, COLM performs packet transmission using proportional fairness based on the queue factor. After that, Algebraic Congestion Likelihood (ACL) measure is performed in COLM-ACL technique to obtain throughput fairness by achieving congestion threshold factor and likelihood factor. This helps to improve throughput fairness in MANET. As a result, the COLM-ACL technique improves network throughput and congestion control rate.