LMLakshmana Phaneendra Maguluri

Lakshmana Phaneendra Maguluri

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

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Lakshmana Phaneendra Maguluri has a total of 59 co-authors. Here are the top co-authors by number of shared publications:

  • A. Arulmozhi

    Department of Physics, St. Joseph’s College of Engineering, OMR, Chennai 600119, Tamil Nadu, India

    1 shared publication
  • A. Kavitha

    Department of Electronics and Communication Engineering, K. Ramakrishnan College of Technology, Trichy 621112, Tamil Nadu, India

    1 shared publication
  • Ajeet Kumar Srivastava

    Department of Electronics and Communication Engineering, School of Engineering and Technology, Chhatrapati Shahu Ji Maharaj University, Kanpur 208024, Uttar Pradesh, India

    1 shared publication
  • Battula Nancharaiah

    Professor & HOD Department of ECE Usha Rama College of Engineering and Technology 521109 Telaprolu India

    1 shared publication
  • Bhaskar Kathirvelu

    Department of Automobile Engineering, Rajalakshmi Engineering College, Rajalakshmi Nagar, Thandalam, Chennai 602105. Tamilnadu, India

    1 shared publication
  • D. Baburao

    Department of Computer Science and Engineering (Data Science), Vidya Jyothi Institute of Technology (VJIT), Aziz Nagar Gate, C. B. Post, Hyderabad 500075, Telangana, India

    1 shared publication

Latest Publications

  • Hybrid DeepXAI Fusion Network for Interpretable Big Data Analytics and Secure Cloud Decision-Making Systems

    by B. Nancharaiah, Ramakrishna Kolikipogu, J. Sulthan Alikhan, Lakshmana Phaneendra Maguluri, Elangovan Muniyandy, D. Baburao

    Big data analytics has become a cornerstone of modern intelligent systems, yet its complexity and opacity often limit interpretability, trust, and security in decision-making. To address these challenges, we present the Hybrid DeepXAI Fusion Network, a novel deep learning framework designed to fuse temporal, relational, and contextual features while embedding explainable artificial intelligence (XAI) modules for transparent inference. The system leverages 1D-CNNs for temporal feature extraction, Transformer encoders for long-term contextual modeling, and Graph Neural Networks (GNNs) for capturing inter-feature dependencies, integrated through an attention-based multimodal fusion mechanism. Additionally, Class Activation Mapping (CAM) and explanation loss are incorporated to ensure interpretability consistency, enhancing trustworthiness in sensitive applications. The model is deployed in a federated cloud architecture, ensuring data privacy through homomorphic encryption and blockchain auditing. Experimental evaluation across healthcare, finance, and energy datasets demonstrates superior performance, achieving 98.42% accuracy, 97.9% precision, 98.1% recall, and 98.0% F1-score, surpassing state-of-the-art baselines including CNN-LSTM, Transformer-only, and XGBoost. The robustness of the proposed system was validated under adversarial attacks, where it retained over 91.5% accuracy even against PGD perturbations, while scalability tests confirmed its ability to handle streaming loads of 80,000 messages per second with sub-200 ms latency. These results confirm that the Hybrid DeepXAI Fusion Network offers an efficient, interpretable, and secure solution for next-generation cloudbased decision-support systems.

  • Cognitwin: A Neuro-Digital Synthesis for Predictive Decision Intelligence in Smart Ecosystems

    by Frederick Ruby Helen, Kattupalli Sudhakar, Rajendra Kodamanchili, Mamta Bansal, Naveen Kumar Sanku, Lakshmana Phaneendra Maguluri

    The integration of Digital Twins (DTs) with cognitive modeling presents a novel paradigm for enhancing intelligent decision-making in complex, real-time environments. In this paper, Cognitwin, a hybrid neuro-digital model is presented that combines physical systems high-fidelity digital replicas with human-like cognitive processing mechanisms to allow proactive, explainable, and adaptive decision-making. The system architecture integrates physicsbased and data-driven modeling, cognitive architectures (e.g., ACT-R), as well as reinforcement learning and graph neural networks (GNNs) to produce actionable intelligence. Uncertainty-aware perception relies on sensor fusion and DempserShafer theory, whereas the prediction part of learning is based on LSTM and Transformer models. The performance in ten smart areas such as healthcare, mobility and energy has shown better results where the average accuracy is 95.7, F1score of 0.94 and the latency is less than 200 ms in most applications. The importance of the integrated feedback and cognitive modules has shown to be confirmed by an ablation study which has proven that the integrated cognitive and feedback modules are essential in producing strong accuracy. Furthermore, the system has the decision quality rating of 0.94 following the reinforcement learning policy refinement processes. Comparative analysis shows that GNN is better than LSTM in causal reasoning tasks and justifies with lower MSE values and higher causal scores. Cognitwin provides a bidirectional synchronization process that is real time between the physical and the virtual-cognitive layer and creates continuity in learning and in situational awareness.

  • Impact of online community learning programs on TB prevention behaviors

    by Mamilla Aishwarya Rajasekhar, Suresh C. Ranganathan, S. Kuppuraj, V.P. Murugan, Venkatesan Hariram, Lakshmana Phaneendra Maguluri

  • Top Cited Papers

    Intelligent Reflector Metasurface Design for 6G Adaptive Beamforming Using Model‐Level Multi‐Scale Edge Guided Attention Graph Neural Network With Biruni Earth Radius Optimization

    by Jyoti Atul Dhanke, R. Krishnamoorthy, G.S.N. Murthy, Lakshmana Phaneendra Maguluri, Jnaneshwar Pai Maroor, Mangal Singh

    The advancement of sixth‐generation (6G) wireless communication demands intelligent, adaptive solutions to meet the increasing requirements of ultra‐reliable low‐latency communication (URLLC), massive connectivity, and terahertz (THz) data rates. Intelligent Reflecting Metasurfaces (IRMs) have emerged as a revolutionary paradigm that reconfigures the wireless propagation environment by intelligently controlling the behavior of electromagnetic waves. However, traditional beamforming methods face significant challenges in dynamically adapting to complex, time‐varying wireless environments, especially in the presence of high user mobility and signal obstructions. These limitations hinder the efficiency of signal directionality and energy focusing, resulting in sub‐optimal communication performance. To overcome these issues, a novel framework, the Model‐Level Multi‐Scale Edge‐Guided Attention graph neural Network (M‐LMSedGAN), is introduced, enabling efficient learning from complex spatial relationships and heterogeneous signal interactions. Optimization of the reflective phase shifts is achieved using the Biruni Earth Radius Optimization (BERO) algorithm, which enhances the global search capability and convergence efficiency of the system. The proposed method enables real‐time and context‐aware beamforming adjustments, achieving superior signal quality and improved network performance. Experimental evaluations demonstrate a peak beamforming accuracy of 99.9%, indicating robust generalization across dynamic environments and diverse user distributions. This intelligent approach establishes a foundation for deploying energy‐efficient, high‐performance IRM‐assisted 6G communication infrastructures.


  • DDoS Detection and Mitigation in Cloud via Hybrid Architecture With Extractive Feature Set

    by Sumalatha Ippa, Lakshmana Phaneendra Maguluri

    Unlike traditional systems, cloud infrastructures are dynamic and distributed, necessitating specialized methods to handle distributed denial of service (DDoS) attacks that flood services with traffic, causing downtime or performance issues. This research introduces a novel approach for sensing and preventing Cloud DDoS attacks by addressing the unique challenges of dynamic and distributed cloud environments. The novelty of this work lies in three main contributions: (i) Preprocessing, (ii) Feature extraction, (iii) Attack detection, and (iv) Attack mitigation. In the preprocessing phase, the input dataset is balanced using random undersampling and processed with an enhanced decimal scaling (EDS) normalization technique. The feature extraction phase involves retrieving enhanced correntropy‐based features, higher‐order statistical (HOS) features, and raw features. Attack detection is carried out using a novel deep learning (DL)‐based hybrid model named improved LinkNet‐PolyNet (ILPNet), which combines improved LinkNet (ILNet) and PolyNet. This approach offers improved accuracy over the conventional LinkNet model by addressing bottlenecks between the encoder and decoder blocks. Lastly, an enhanced entropy‐based mitigation approach is used to reduce detected attacks. The efficiency of this approach is evaluated through comparisons with existing methods using detailed performance, statistical, and ablation analyses. The ILPNet strategy demonstrated superior performance compared to conventional methods, achieving an accuracy of 0.960, a precision of 0.953, and an F ‐measure of 0.961.


  • Open Access

    Design and implementation of solar-grid based charging station for electric vehicle with fault detection method using R-Pi and IoT processor

    by M. Vaigundamoorthi, S. Karthick, V. S. Chandrika, D. Chithra, K. V. Balaramakrishna, K. Lakshmi Khandan, Lakshmana Phaneendra Maguluri, S. Chandrasekar, M. Janarthanan

    In this research describes the electrical vehicle (EV) charging station using PV panel with fault detection methods. The PV modules will failure for some time, because of some external factors and internal factors. In direct fault condition the monitor and analyze the external factors such as the life span, high intensity and breakage of the PV panels using Raspberry Pi (R-Pi) processor with internet of things (IoT) system. In power demand/day on the PV panel will be evaluated and analyzed through R-Pi processor and IoT. The efficiency and the range values of the PV panels will be monitored and analyzed through IoT. Proposed work explains, how the fault detection techniques have been improved and adopted in using R-Pi processor through IoT platform. The proposed dataset pre-processing system is incorporated with IoT module. The grid fault clearing time will be compared with the actual values through R-Pi processor. The PV panel faults are detected using thermal image processing, that image parameter values analysis through IoT based internal monitoring system.

  • OPTIMIZING ON-STATE CURRENT AND RF FIGURES OF MERIT IN HETERO GATE THROUGH ADVANCED SOURCE DESIGN

    by D. Jessintha, Ponni Bala Muthiah, A. Kavitha, A. Arulmozhi, Ajeet Kumar Srivastava, Lakshmana Phaneendra Maguluri, Ramaswamy Manjith, Vineet Tirth, A. Rajaram

    Using a device simulator developed by technology computer aided design, this study examines how changing the dielectric material close to the source area affects the linearity and analogue/RF figures of merit in hetero gate TFETs (HG-TFETs). This research takes SiO2, Al2O3, and HfO2 into account among other dielectric materials. We take a look at a TM-GS-VTFET structure, which stands for triple-metal-gate stacked III–V vertical tunnel field-effect transistor. One TM-GS-VTFET, Device-A, makes use of a source pocket, while the other, Device-B, introduces a novel source extension technique. The improved ON-state current (ION) with subthreshold swing of source pocket-engaged TFETs are notable improvements over conventional TFETs. Nevertheless, more intricate production processes could be necessary for them. The electric-field distribution is changed, and the tunneling barrier width is reduced close to the source side by the source extension approach. Both the tunneling rate and the device’s functionality are enhanced by this. Silvaco Technology’s Computer-Aided Design (TCAD) program ran the simulation. We looked at a bunch of RF and analog characteristics. According to the results, Device-B had an ION of 234.03 [ ]A/[ ]m, an OFF-state current of [ ] [ ]A/[ ]m, a subthreshold swing of 28.58[ ]mV/decade, and a transconductance of 245.28 [ ]S/[ ]m. Gain bandwidth product (GBP), oscillating frequency (fmax), and RF frequency factor cutoff frequency (fT) were 152.45, 830, and 25.16[ ]GHz, respectively. Compared to Device, Device-B displayed ION that was 11 times higher, GBP that was three and a half times higher, and fmax that was 3.15 times higher.


  • Open Access

    A comprehensive evaluation of machine learning algorithms for precise energy consumption forecasting in smart homes

    by Lakshmana Phaneendra Maguluri, M. Shankar, R. Aruna, D. Chitra Devi, M. J. Suganya

    Energy is one of the most critical and costly resources, playing a vital role in our daily lives. As technology advances, the demand for energy also increases. This work proposes a model for predicting energy consumption in smart homes, consisting of data preprocessing, performance evaluation, and application. Once the data is processed, it is fed into the prediction module, where various machine-learning algorithms are applied to forecast energy consumption. As smart home environments grow in complexity, selecting the most effective machine learning algorithm becomes increasingly crucial. The persistent challenge lies in manually discerning the best-performing algorithm, given their potential variance in efficacy across diverse use cases or datasets. In the dynamic landscape of energy conservation and cost-effective power generation, precise forecasting of energy consumption is essential, playing a pivotal role in advancing energy sustainability and bolstering economic stability. This introduction explores the intricate terrain of predicting energy utilization within smart homes, a domain that has seen increased interest due to the integration of machine learning algorithms. The primary focus of this exploration is the rigorous evaluation of these algorithms, using key performance metrics such as mean absolute error (MAE), mean squared error (MSE), and R-squared.


  • Open Access

    SVM-Based Classifier For Early Detection Of Alzheimer's Disease

    by Parvatham Niranjan Kumar, Lakshmana Phaneendra Maguluri

    Alzheimer's disease (AD) is a disorder affecting the brain and its functions that results in permanent damage to the brain. This multifaceted disease slowly destroys brain cells, reducing a person's ability to think, remember, and carry out even the most basic duties. Ultimately, this cognitive decline leads to dementia. A significant portion of the global population faces metabolic challenges like Alzheimer's disease and diabetes. Recent research has explored various machine learning approaches aimed at early disease detection. Early AD diagnosis is very important for a speedy recovery and minimizing damage to brain cells. In this proposed work, a machine learning model is developed using Support Vector Machine to detect individuals with dementia (AD) or without dementia (NC). Our model is trained using 2D Magnetic Resonance Imaging (MRI) brain scan images. We computed common performance measures like the F1-score and accuracy to assess the model. The 15-fold cross-validation technique was applied to cross-validate these results. Notably, our results demonstrate that Support Vector Machine (SVM) attained a remarkable accuracy of 99.06% in the detection of AD. Accurate Alzheimer's disease detection through machine learning algorithms can greatly reduce the annual mortality rates related to AD.


  • Top Cited Papers
    Open Access

    An Interpretable Approach with Explainable AI for Heart Stroke Prediction

    by Parvathaneni Naga Srinivasu, Uddagiri Sirisha, Kotte Sandeep, S. Phani Praveen, Lakshmana Phaneendra Maguluri, Thulasi Bikku

    Heart strokes are a significant global health concern, profoundly affecting the wellbeing of the population. Many research endeavors have focused on developing predictive models for heart strokes using ML and DL techniques. Nevertheless, prior studies have often failed to bridge the gap between complex ML models and their interpretability in clinical contexts, leaving healthcare professionals hesitant to embrace them for critical decision-making. This research introduces a meticulously designed, effective, and easily interpretable approach for heart stroke prediction, empowered by explainable AI techniques. Our contributions include a meticulously designed model, incorporating pivotal techniques such as resampling, data leakage prevention, feature selection, and emphasizing the model’s comprehensibility for healthcare practitioners. This multifaceted approach holds the potential to significantly impact the field of healthcare by offering a reliable and understandable tool for heart stroke prediction. In our research, we harnessed the potential of the Stroke Prediction Dataset, a valuable resource containing 11 distinct attributes. Applying these techniques, including model interpretability measures such as permutation importance and explainability methods like LIME, has achieved impressive results. While permutation importance provides insights into feature importance globally, LIME complements this by offering local and instance-specific explanations. Together, they contribute to a comprehensive understanding of the Artificial Neural Network (ANN) model. The combination of these techniques not only aids in understanding the features that drive overall model performance but also helps in interpreting and validating individual predictions. The ANN model has achieved an outstanding accuracy rate of 95%.