Neural network based reliable transport layer protocol for MANET
Pravin Kumar, Sachin Tripathi, Pankaj Pal · 2018
Use of traditional transport layer protocol to achieve reliability, such as TCP, over mobile ad-hoc network (MANET) is a challenging task due to unique features. The unique features includes absence of a base station, as it is in the case of cellular networks, node mobility, multi-hop communication over lossy and non-deterministic wireless mediums, similarity in traffic pattern experienced by neighboring nodes, etc. One of the main reasons for the poor performance of formal TCP version used in wired networks over wireless networks is that the, packet loss in wireless networks is not only due to congestion there are other problem inherent to wireless network which can cause packet loss. To address this issue, this paper proposes a neural network based congestion control technique for reliable data transfer over MANET, which recognizes and capture the mobility behavior of node. The captured mobility behavior is used to identify the cause of packet loss, in order to take action which increases the reliability of underlying MANET. We evaluate the performance of proposed protocol based on the simulation result, obtained using QualNet 7.4 network simulator. Our evaluation results show that proposed congestion control technique shows improvement in MANET with high degree of node mobility in terms of reliability, bandwidth usage and energy efficiency.