Back close

Communication Capability Enhancement in Ocean Networks with Predictive Learning Models  

Project Incharge:Dr. Simi Surendran
Co-Project Incharge:Dr. Maneesha Vinodini Ramesh
Co-Project Incharge:Dr. Alberto Montresor, University of Trento, Italy
Communication Capability Enhancement in Ocean Networks with Predictive Learning Models  

The lack of low-cost communication facilities to the shore remains a fundamental problem for fishermen engaged in deep-sea fishing. The Offshore Communication Network (OCN) aims to resolve this connectivity issue by building a wireless network of fishing vessels to provide Internet over the ocean. The connectivity of fishing vessels is essential to disseminate messages, monitor emergency management, and provide information services to enhance the blue economy. Although OCN shares some of the characteristics of terrestrial networks, they exhibit unique features and research challenges. The influence of extreme weather conditions 
on wireless signals, the inability to deploy additional infrastructure, the movements induced by sea waves, the expanded mobility freedom at sea, and the misalignment of  directional antenna links cause frequent connectivity breakages in OCN. Hence, maintaining continuous connectivity is challenging in OCN. This project discusses methods to estimate and enhance the communication capabilities of nodes to provide uninterrupted Internet at sea. To evaluate the communication capability, we propose a hybrid machine learning framework leveraging real-time data collected from extensive marine experiments on multiple fishing vessels. In addition, a metric to quantify the node’s communication capability to the shore and the peer 
nodes is proposed. A three-level connectivity enhancement scheme is applied across the physical, link, and network layers to improve communication capability. At the physical level, the positions of the nodes are re-oriented to higher connectivity zones depending on the user requirement at distinct stages of fishing. For link-level optimization, the transmission queue size is estimated for multiple priority messages, and the traffic is scheduled adaptively to minimize queuing delay. An intelligent routing scheme with an OCN-specific reward function in a reinforcement learning model is employed for improving network-level packet delivery. The model suggests locations with high connectivity at different fishing stages and provides better connectivity at sea.

Outcome

  • Surendran, Simi, Maneesha Vinodini Ramesh, Martin J. Montag, and Alberto Montresor. “Modelling communication capability and node reorientation in offshore communication network.” Elsevier Computers & Electrical Engineering 87 (2020): 106781. 
  • Simi Surendran, Maneesha V Ramesh, Alberto Montresor, Martin Montag, Link Characterization and Edge-centric Predictive Modelling in an Ocean Network, IEEE Access  
  • Surendran, Simi, Maneesha Vinodini Ramesh, and Alberto Montresor. “Predictive Analytics Integrated Multi-level Optimization of Offshore Connectivity in Ocean Network.” 2021 IEEE 46th Conference on Local Computer Networks (LCN). IEEE, 2021. 
  • Surendran, S., Ramesh, M. V., Montresor, A., & Casari, P. (2022). Reinforcement learning-based connectivity restoration in an ocean network of fishing vessels. In 2022international conference on advanced networks and telecommunications systems (ants). IEEE

Related Projects

Comparison of Anti-microbial Activity of Anti-acne Agents
Comparison of Anti-microbial Activity of Anti-acne Agents
Malware detection using FPGA, Sandboxing and Machine Learning
Malware detection using FPGA, Sandboxing and Machine Learning
AI for Early Detection of Sepsis & Infections
AI for Early Detection of Sepsis & Infections
Marine Algae – An Important Source for Lectins
Marine Algae – An Important Source for Lectins
MedSIM Overview
MedSIM Overview
Admissions Apply Now