An Efficient and Most Reliable Novel Route Selection Algorithm (ERRSAE) to Improve QOS in Manet Environment

Mobile ad-hoc network (MANET) has got more focus because of its pragmatic applications and inevitability of communication in mobile devices. MANETs are collection of mobile hosts dynamically forming a temporary network without the aid of any existing infrastructure or centralized control. Quality of Service (QoS) support for MANET is a challenging task due to the dynamic topology and limited resources. The existing QoS based routing solutions for MANET involves with single metric or two metrics. It is important that MANETs should provide QoS support routing such as end-to-end-delay, energy consumption, cluster stability, detection rate, and link failure prediction analysis, etc. This paper proposed a QoS enabled Routing Algorithm called Efficient and most Reliable Route Selection Algorithm (ERRSAE) to improve QoS performance using Swarm Intelligence. The proposed algorithm has three parts. First of all, a clustering algorithm named Honey Bee Nest (HBNEST) is proposed which makes it possible to build efficient collective and self-organized clusters. Second, the Trusted Artificial Ant-Based Routing Protocol (TAABRP) is proposed based on Swarm Intelligence algorithms such as ACO to choose an optimal path from source to destination to transmit data packets. Third, proposed a Regression Based Link Failure Prediction Algorithm (RBLFPA) and Fuzzy Assisted Ant Colony Optimization Algorithm (FAACOA). The Regression Based Link Failure Prediction Algorithm (RBLFPA) for multipath routing in MANET is used to predict the link failure and forward packets whenever a link failure occurs and increases the quality of service. The Fuzzy Assisted Ant Colony Optimization Algorithm (FAACOA) uses a distributed fuzzy logic module to identify misbehaving nodes. The performance of the proposed algorithm ERRSAE with our other proposed algorithms SRABDE, SIBCSRP with pre-existing algorithms like TPACO, DE_AOMDV, FHACO, and SAFEACO was evaluated using Network simulator (NS). The comparison was done to evaluate performance metrics such as packet delivery ratio, routing overhead, end-to-end-delay, energy consumption, cluster stability, throughput, detection rate, and link failure prediction analysis. The results clearly demonstrate that the proposed ERRSAE algorithm outperformed under majority of the network performance metrics and QoS parameters.