RSRani Sahu

Rani Sahu

*This is an algorithmically generated author record based on publicly available information.

Rani Sahu has a total of 19 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    A Novel RL-Driven Zone-Based Leader-Aware Energy-Efficient Routing Protocol for Dynamic MANET Environments

    by Rani Sahu, Babita Rathore

    Mobile Ad Hoc Networks (MANETs) face major routing problems because their network structure keeps changing, their energy levels are minimal, their nodes move between locations, and they lack any central network control systems. The traditional routing protocols Ad hoc On-Demand Distance Vector (AODV) and Ad hoc On-Demand Multipath Distance Vector (AOMDV) face major problems in dynamic environments because they use too much energy, their routes fail too often, and their network control demands become too high. The paper introduces RML-ZEREM to solve existing limitations, which functions as a Reinforcement Learning (RL) based Zone-Based Leader-Aware Energy-Efficient Routing Protocol for MANETs. The proposed approach partitions the network into multiple zones and employs energy-aware leader node selection to manage routing operations efficiently. The system uses Q-learning to create an adaptive routing system that chooses the best routing paths according to current network conditions, including residual energy levels, node movement, traffic intensity, and link reliability. The proposed protocol performance assessment uses the NS-2.35 simulator to test different simulation conditions, which include various simulation durations, node mobility rates, network capacity, and simulation area size. The simulation results show that RML-ZEREM achieves better performance than traditional AODV and AOMDV protocols through its ability to increase throughput while decreasing energy usage, improving packet delivery ratio, and reducing routing overhead. The zone-based hierarchical structure enhances network stability and scalability for MANET systems that operate in dynamic environments. The RML-ZEREM protocol functions as an intelligent routing system that adjusts its operations to achieve energy efficiency through its framework, which serves next-generation MANET applications.

  • Open Access

    Platform-Independent Performance Validation of NS-2.35: A Comparative Study of Native and Virtualized Environments

    by Vinay Sahu, Ankur Khare, Rani Sahu

    The paper presents a comprehensive experimental study that evaluates the Network Simulator 2 NS-2.35 on both native Linux systems and VMware-based virtualised environments. Researchers use NS-2.35 as a standard tool for testing Mobile Ad Hoc Networks MANETs. The researchers face challenges because they must install their software to achieve functionality across multiple operating systems. The research study provides a complete installation guide and troubleshooting instructions, and tests whether different execution environments affect simulation results. The researchers create a MANET scenario using the Ad hoc On-Demand Distance Vector AODV routing protocol to conduct identical tests in both environments. The assessment uses Throughput, End-to-End Delay, Packet Delivery Ratio PDR, and Packet Loss Ratio PLR as performance indicators to evaluate node densities of 20, 30, and 50. The researchers conduct multiple experiments and analyse them using the mean and standard deviation Mean SD to obtain statistically reliable results. The testing results show that all performance measurements remain the same across both native systems and virtualised environments, proving that NS-2.35 delivers platform-independent performance that remains constant under the same system conditions. The test runs produce stable simulation results because two tests show extremely low standard deviation values. The simulation results remain accurate because execution efficiency and system performance differences between the two systems result in only minor discrepancies in the test results. The study concludes that researchers can use both native Linux and virtualised environments for NS-2.35 simulations, as virtualisation preserves simulation fidelity. This work validates the cross-platform performance of NS-2.35, thereby building trust in simulation research findings.

  • Open Access

    Assessing the Impact of Node Density on MANET Routing: A Comparative Study of NTSM, AODV, and DSDV Using NS2 Simulation

    by Vinay Sahu, Rani Sahu

    MANETs provide a dynamic, decentralised communication architecture that enables nodes to form spontaneous networks. They are, therefore, best suited for situations where fixed infrastructure does not exist. However, performance in MANETs depends heavily on the routing protocol used and the node density. This paper describes the performance evaluation of three necessary routing protocols, NTSM, AODV, and DSDV, with respect to different node density conditions. We use extensive simulations on the NS2 simulation framework to analyses the performance of these protocols at several node densities. The critical performance metrics under study include packet delivery ratio, end-to-end delay, network throughput, and routing overhead. The systematic comparative study aims to identify the strengths and weaknesses of the protocols and their adaptability to dynamic variations within the network. This study explains in detail how NTSM, AODV, and DSDV respond to changes in node density in simulated MANET environments. They could serve as valuable guidelines for selecting the best routing protocol for different deployment types. Furthermore, these results highlight that, when designing and optimising MANETs, network dynamics and node density variations should also be considered, particularly through simulation tools such as NS2 for performance evaluation.

  • Open Access

    Integrating LSTM and CNN for Stock Market Prediction: A Dynamic Machine Learning Approach

    by Sujata Joshi, Bangaru Lakshmi Mahanthi, Pavithra G, Kiran Sree Pokkuluri, Swapnil S. Ninawe, Rani Sahu

    This paper investigates the application of machine learning models, including Convolutional Neural Networks (CNN), Long Short-Term Memory (LSTM), and a hybrid LSTM-CNN model, for predicting stock prices of companies listed on the National Stock Exchange (NSE). The proposed hybrid model leverages CNN’s capability for spatial feature extraction and LSTM’s proficiency in modelling temporal dependencies, effectively addressing the complex and volatile nature of stock price movements. Using a comprehensive dataset of historical stock prices and trading volumes from various sectors, the hybrid model achieved superior performance with a 15% improvement in RMSE compared to standalone CNN and LSTM models. Results demonstrate its robustness, particularly in volatile market conditions, showcasing its potential for accurate and reliable predictions. This study contributes a novel hybrid approach that integrates spatial and temporal learning to enhance stock market prediction, offering valuable insights for investors and financial analysts while providing a scalable framework for broader financial applications.


  • Open Access

    Revolutionizing MANET Route Discovery with INTSM: An Innovative Load Balancing Approach

    by Rani Sahu, Neetu Sahu, Vinay Sahu

    Communication challenges in ad hoc networks arise due to the mobility of nodes, causing frequent changes in connections and locations. Maintaining network equilibrium to prevent node overload and underutilization is crucial. However, imposing static behaviors on nodes to improve performance can lead to delays, especially in core nodes. Addressing these issues, this research proposes the Intermediate Node Traffic Sharing Model (INTSM) for ad hoc networks. INTSM prioritizes congestion control and load balancing during route discovery, aiming to optimize network resource utilization and traffic distribution, thereby reducing packet delays. The model employs dynamic traffic sharing algorithms that consider real-time network conditions, enabling nodes to adjust their behaviour adaptively. This approach minimizes congestion by distributing traffic loads more evenly across the network, preventing bottlenecks at central nodes. Additionally, INTSM incorporates predictive analysis to foresee potential congestion points and reroute traffic proactively, enhancing overall network stability and performance. Extensive simulations demonstrate that INTSM significantly reduces average packet delay and improves throughput compared to traditional routing protocols. The results highlight the model's efficacy in diverse scenarios, including high mobility and varying traffic loads, proving its robustness and scalability. The primary objective of this study is to enhance navigation and equilibrium mechanisms to improve the performance of ad hoc networks, contributing to more reliable and efficient wireless communication systems. The findings of this research have significant implications for the design of future ad hoc networks, particularly in applications requiring high reliability and quick adaptation to changing network conditions, such as disaster recovery, military operations, and mobile sensor networks. By addressing the critical challenges of congestion control and load balancing, INTSM offers a promising solution to enhance the resilience and efficiency of ad hoc networks.

  • Open Access

    Exploring the Future of Stock Market Prediction through Machine Learning: An Extensive Review and Outlook

    by Sourabh Jain, Navdeep Kaur Saluja, Anil Pimplapure, Rani Sahu

    A thorough analysis of trends and future directions reveals how machine learning is revolutionizing stock market forecasting. The most recent research on machine learning applications for stock market prediction during the previous 20 years is methodically reviewed in this article. Artificial neural networks, support vector machines, genetic algorithms in conjunction with other methodologies, and hybrid or alternative AI approaches were the categories used to group journal articles. Every category was examined to identify trends, distinct perspectives, constraints, and areas that needed more research. The results provide insightful analysis and suggestions for further study in this developing topic.


  • Open Access

    Challenges and Opportunities of 5G Network: A Review of Research and Development

    by Vinay Sahu, Neetu Sahu, Rani Sahu

    The advent of the fifth generation of wireless technology, commonly referred to as 5G, heralds a transformative era in communication and technology interaction. With its promise of high-speed connectivity, minimal latency, and support for massive machine-type communications, 5G networks are poised to enable a wide array of applications across various domains such as the Internet of Things (IoT), smart cities, virtual and augmented reality, Industry 4.0, and edge computing. However, the deployment of 5G networks presents numerous challenges, including infrastructure requirements, limited availability of high-frequency spectrum, security vulnerabilities, high deployment costs, and interoperability issues with existing networks. This paper meticulously examines the challenges, opportunities, and applications of 5G networks. It delves into the potential of 5G networks to offer rapid connectivity, facilitate novel applications, and revolutionize industries. Moreover, the paper scrutinizes the hurdles associated with deploying and sustaining 5G networks, underscoring the imperative need for further research in this domain to ensure the seamless adoption and integration of 5G technology. As 5G networks continue to evolve, addressing these challenges will be crucial for unlocking the full potential of this transformative technology and realizing its societal and economic benefits on a global scale. Additionally, collaboration between stakeholders including governments, telecommunications companies, technology providers, and regulatory bodies will play a pivotal role in overcoming these challenges and driving the successful deployment and adoption of 5G networks worldwide.


  • Open Access

    "An Energy-Efficient Algorithm for Resource Allocation in H-CRAN((EE H-CRAM)) for 5G Networks"

    by Rani Sahu, Vinay Sahu

    The demand for faster internet has increased rapidly in today's generation due to more people using it. To keep up with this growth, a new network called 5G is being developed to replace the current 4G network. 5G has been designed to meet all the communication needs of users. One promising technology for this new network is Heterogeneous Cloud Radio Access Network (H-CRAN), which is energy-efficient and can handle high data rates. By using this network, better service quality can be achieved by turning small cells into bigger ones. This network is different from the usual networks we have today, because it uses a type of radio system that is much better at reusing radio resources. The structure of this network includes more Remote Radio Heads mounted around the Macro Base Station. This type of network is very helpful in saving energy and achieving high Energy Efficiency compared to regular networks. Its resource allocation techniques can optimize the limited spectrum resources by reducing the interference between users. Traditional cellular networks have not been able to save energy effectively, but this network is seen as a promising solution to the challenges faced by 5G networks, such as high data rates, limited power, and delays. In our paper, we suggest some ways to make H-CRAN more energy-efficient. We propose energy-efficient methods for H-CRAN (EE H-CRAM), incorporating an energy-efficient resource allocation management design that caters to the heterogeneous cloud radio access networks in 5G. We use a special energy-saving program to manage the resources of the network, which is designed for 5G. By using this program, we can improve the speed of the network. Overall, our paper proposes an innovative and energy-efficient approach to H-CRAN design, which can significantly improve the performance of the 5G network.