Journal Articles
Permanent URI for this collectionhttps://idr.nitk.ac.in/handle/123456789/19884
Browse
6 results
Search Results
Item Integrating artificial intelligence in aquaculture: opportunities, risks, and systemic challenges(Springer Science and Business Media Deutschland GmbH, 2025) M R, D.; Sanshi, S.; Singh, M.P.; Gupta, M.Aquaculture plays a significant role in the food chain and in rural economies. Continuous water quality monitoring, health management, real-time growth monitoring, and biomass estimation are critical aquaculture activities. Recent advances in computer vision, image processing, and Artificial Intelligence (AI), particularly in Machine Learning (ML) and Deep Learning (DL), enable the control, drive, and solve the problems related to daily real-time aquaculture activities. The performance of various ML and DL models and the quality and availability of public datasets in this domain remain underexplored, and the redefinition of the role of AI in aquaculture is the motivation for this study. This survey aims to analyse and evaluate various methods and datasets for monitoring water quality, estimating fish biomass, disease prediction, and behavioural analysis, to highlight recent developments and to seek the attention of researchers to address the challenges and concerns of aquaculture systems. Currently, there is no single, generalised AI model capable of performing all essential aquaculture activities using continuous time-series data generated from diverse, heterogeneous ponds across a wide geographical area. To address this gap, a 5G-enabled Federated Learning (FL) framework utilising Unmanned Aerial Vehicles (UAV) for cooperative data collection and model training is highly recommended. © The Author(s), under exclusive licence to Springer Nature Switzerland AG 2025.Item Enhanced mobility aware routing protocol for Low Power and Lossy Networks(Springer New York LLC barbara.b.bertram@gsk.com, 2019) Sanshi, S.; Jaidhar, C.D.Due to the technological advancement in Low Power and Lossy Networks (LLNs), sensor node mobility becomes a basic requirement for many extensive applications. Routing protocol designed for LLNs must ensure real-time data transmission with minimum power consumption. However, the existing mobility support protocols cannot work efficiently in LLNs as they are unable to adapt to the change in the network topology quickly. Therefore, we propose an Enhanced Routing Protocol for LLNs (ERPL), which updates the Preferred Parent (PP) of the Mobile Node (MN) quickly whenever the MN moves away from the already selected PP. Further, a new objective function that takes the mobility of the node into an account while selecting a PP is proposed. Performance of the ERPL has been evaluated with the varying system and traffic parameters under different topologies similar to most of the real-life networks. The simulation results showed that the proposed ERPL reduced the power consumption, packet overhead, latency and increased the packet delivery ratio as compared to other existing works. © 2017, Springer Science+Business Media, LLC, part of Springer Nature.Item Fuzzy optimised routing metric with mobility support for RPL(Institution of Engineering and Technology JBristow@theiet.org, 2019) Sanshi, S.; Jaidhar, C.D.Recently, many Internet of Things (IoT) applications have emerged with mobility as a fundamental requirement. The presence of a mobile node that changes location around the application domain affects the performance of the Routing Protocol for Low Power Lossy Network (RPL) designed for IoT, leading to repeated disruptions that cause data loss and more power dissipation. In this study, a fuzzy optimised routing metric with mobility support (FL-RPL) has been proposed to enhance the performance of the RPL. The fuzzy inference system considers various routing metrics to pick a suitable candidate parent as the preferred parent node to forward the data to the sink node. Further, timer functions have been added to maintain consistent neighbours to support mobility and seamless connectivity. The FL-RPL has been implemented and tested with different parameter settings for a practical scenario. The obtained simulation results clearly demonstrated that the proposed solution increased packet delivery ratio by approximately 12%, and reduced power consumption by 20% compared with the standard RPL. © 2019 The Institution of Engineering and Technology.Item Enhanced mobility routing protocol for wireless sensor network(Springer, 2020) Sanshi, S.; Jaidhar, C.D.Recently, the routing protocol for low power and lossy networks (RPL) was standardized and is considered as the default standard for routing over the low power and lossy networks. However, it has not been optimized to work effectively, especially under mobility, and suffers from frequent disconnections that result in packet loss and increased energy consumption. In this paper, an enhanced mobility routing protocol for wireless sensor network (EM-RPL) that incorporates modules to support the mobility of nodes has been proposed. The main goal of the EM-RPL is to increase network reliability and efficiency by selecting a route that is more stable and reduces the frequency of route discovery process. The performance of the proposed EM-RPL has been evaluated in the Contiki-based Cooja simulator and compared with the performance of other protocols that support mobility in the RPL. The simulation results demonstrated that the EM-RPL improves the packet delivery ratio and minimizes power consumption by allowing the mobile nodes to select a more stable path. © 2018, Springer Science+Business Media, LLC, part of Springer Nature.Item IoT energy efficiency routing protocol using FHO-based clustering and improved CSO model-based routing in MANET(John Wiley and Sons Ltd, 2024) Sanshi, S.; Karthik, N.; Vatambeti, R.Many protocols, services, and electrical devices with built-in sensors have been developed in response to the rapid expansion of the Internet of Things. Mobile ad hoc networks (MANETs) consist of a collection of autonomous mobile nodes that can form an ad hoc network in the absence of any pre-existing infrastructure. System performance may suffer due to the changeable topology of MANETs. Since most mobile hosts operate on limited battery power, energy consumption poses the biggest challenge for MANETs. Both network lifetime and throughput improve when energy usage is reduced. However, existing approaches perform poorly in terms of energy efficiency. Scalability becomes a significant issue in large-scale networks as they grow, leading to overhead associated with routing updates and maintenance that can become unmanageable. This article employs a MANET routing protocol combined with an energy conservation strategy. The clustering hierarchy is used in MANETs to maximize the network's lifespan, considering its limited energy resources. In the MANET communication process, the cluster head (CH) is selected using Fire Hawk Optimization (FHO). When choosing nodes to act as a cluster for an extended period, CH election factors in connectivity, mobility, and remaining energy. This process is achieved using an optimized version of the Ad hoc On-Demand Distance Vector (AODV) routing protocol, utilizing Improved Chicken Swarm Optimization (ICSO). In comparison to existing protocols and optimization techniques, the proposed method offers an extended network lifespan ranging from 90 to 160 h and reduced energy consumption of 80 to 110 J, as indicated by the implementation results. © 2024 John Wiley & Sons Ltd.Item An Efficient Optimized Neural Network System for Intrusion Detection in Wireless Sensor Networks(Modern Education and Computer Science Press, 2024) Sanshi, S.; Vatambeti, R.; Revathi, V.; Rahman, S.Z.In the realm of wireless network security, the role of intrusion detection cannot be overstated in identifying and thwarting malicious activities within communication channels. Despite the existence of various intrusion detection system (IDS) approaches, challenges persist in terms of accurate classification and specification. Consequently, this article introduces a novel and innovative approach, the African Vulture-based Modular Neural System (AVbMNS), to address these issues. This research aims to detect and categorize malicious events in wireless networks effectively. The methodology begins with preprocessing the dataset and extracting relevant features. These extracted features are then subjected to a novel training technique to enhance the detection and classification of network attacks. The integration of African Vulture optimization significantly enhances the detection rate, leading to more precise attack identification. The research's effectiveness is demonstrated through validation using the NSL-KDD dataset, with impressive results. The performance analysis reveals that the developed model achieves a remarkable 99.87% detection rate and 99.92% accuracy when applied to the NSL-KDD dataset. Furthermore, the outcomes of this novel model are compared with existing approaches to gauge the extent of improvement. The comparative assessment affirms that the developed model outperforms its counterparts, underscoring its effectiveness in addressing the challenges of intrusion detection in wireless networks. © 2024, Modern Education and Computer Science Press. All rights reserved.