Nowadays, routing and the improvement of reliability along with the load balancing and congestion control are the most important and challenging concepts in ad-hoc networks. However, there are some constraints in the bandwidth and the communication range due to the different utility of routing in the Mobile Ad Hoc Network (MANET). These restrictions, in addition to the dynamic network topology, are causing more complexity of routing and data transmission. These problems make the use of conventional methods inefficient, especially in the congestion and traffic control. So, there is a need to propose new solutions for these kinds of networks. This study has used the multipath routing based on the original protocol Ad-Hoc On-Demand Multipath Distance Vector (AOMDV). It proposes an appropriate protocol for the congestion control and balances the favorable loads in the routes. The proposed algorithm is based on different parameters that prepare the fittest multi-paths based on the most efficient routes from the source to the destination, and then it makes a proper traffic distribution pattern in the middle routes based on the congestion analysis of the routes. Simulation results with NS2 shows improvements in the throughput, packet delivery ratio and delay in comparison to the Load Balancing Maximal Minimal Nodal Residual Energy (LBMMRE), Multipath Load Balancing Technique for Congestion Control (MLBCC), and AOMDV routing protocols.

Recently Mobile Ad-hoc Networks(MANETs) are deployed extensively in many areas. To obtain the important information in difficult terrain the feature of the mobility of nodes is used. The effectiveness of information collected by these nodes may be improved by setting up communication among nodes. It is very much important to consider energy conservation among nodes since these mobile nodes operate on limited battery power. The Reactive routing protocols are mostly used over proactive routing protocols due to the nodes mobility and limited energy. The Reactive routing protocols update routing information based on demand but proactive routing protocols update routing information when there is a variation in the network. The Reactive routing protocols collect a huge amount of data and to route these huge data, energy may drain out easily. So it is very much important for the QoS mechanism to be made more energy-efficient. In this paper, an energy-efficient routing mechanism is proposed to achieve this goal. The proposed mechanism achieves by selecting a suitable neighbour node in its routing path to route the information load with efficient energy conservation.