E-CARP: efficient-Cross Layer Congestion Control Adaptive Routing Protocol for Mobile Ad hoc Networks in Constrained Environment

Abstract

Mobile Ad hoc Networks (MANETs) shows unexpected behavior when multiple data streams with heavy traffic load such as multimedia data and data is sent to common destination. The main reason for packet loss in mobile ad hoc networks is due to congestion. In the current design, the routing protocols are not congestion adaptive. The way in which the congestion is handled results in longer delay and more packet loss. When a new route is needed the routing protocols take significant overhead in finding it. In this paper we propose an adaptive congestion control algorithm using cross layer design, which out-performs even during constrained situation. For analyzing the performance we have chosen four popular routing protocols such as AODV, DSR, DSDV and TORA. We also observe through simulation in NS-2 that AODV out-performs other routing protocols in normal situation and DSR out-performs in constraint environment. We strongly argue that routing should not only be aware of but also be adaptive to network congestion.