Founded in 1946, it has since 1973 been a constituent college of Jawaharlal Nehru Technological University, as set by The Jawaharlal Nehru Technological University Act, 1972. In 2008 it had received autonomous status by the Jawaharlal Nehru Technological Universities Act, 2008.
Software-defined networking (SDN) is considered a next-generation networking model. Several networking components are managed through a centralized controller that enables efficiency and flexibility in configuring network devices, implementing policy decisions, and managing the underlying network infrastructure through a programmable unit. Despite its default security protocols, SDN is considered to be contradictory towards DDoS attacks. It is observed from state-of-art studies that intrusion in SDN is possible at various layers of its core architecture. Addressing this problem, this article presents a novel ensemble mechanism inspired by quantum cryptography to secure various layers of SDN. This paper presents a two-fold mechanism to secure communications at the SDN architecture's data plane and control plane. It was firstly addressing the secured communication at the data plane, a novel quantum protocol devised. Further, a machine learning-inspired ensemble classifier is devised to detect DDoS attack-prone traffic at the control plane. Simulation studies presented in this article evidenced that the proposed mechanism outperforms the state of art mechanisms in terms of Accuracy and rate of prediction.
Modern network environments need sophisticated techniques for intrusion detection in order to reliably identify and neutralize threats in the face of increasing volumes and complexity of data. The novel intrusion detection framework proposed in this research combines robust machine learning classification, deep learning-based feature extraction, and advanced normalization techniques. The network traffic data is first normalized to reduce the impact of extreme values and ensure consistent data properties. After the pre-processing, the class imbalance issue is solved by using the Enhanced Synthetic Minority Oversampling Technique (ESMOTE) to balance the classes and reduce the overfitting issue. The balanced data is then processed through a hybrid model, the Semi-skipping Layered Gated Recurrent Autoencoder combined with Efficient Network (SLGRAE-ENet) for deep feature extraction. SLGRAE-ENet incorporates a semi-skipping layer within the Gated Recurrent Autoencoder (GRAE) for improving accuracy. The optimal feature set is selected by using Iterative Minimum Redundancy Maximum Relevance (ImRMR) to minimize the training time and reduce the feature dimensionality issue. An ensemble of machine learning classifiers, such as XGBoost, AdaBoost, and Random Forest, is used for the classification stage to enhance detection performance. Experimental results show that the proposed approach attains high accuracies of 99.7 %, 99.94 %, 99.96 %, 99.76 %, 99.75 % and 98.99 % on the Bot-IoT, CIC-DDoS2019, CSE-CIC-IDS2018, NSL-KDD, CIC-IoT2023, and CIC-IoMT 2024 datasets, respectively. Thus, the proposed integrated approach enhances the accuracy, precision, recall, F-measure, and overall efficiency of intrusion detection systems while minimizing the FAR and MSE.
In this research work, a novel deep learning framework is suggested to recognize highly connected acoustic signals from a noisy mixture for enhancing the reliability and clarity of underwater communication. At first, essential Underwater Acoustic Signals (UAS) are collected from the standard resources. Then, the gathered signals are given to the noise reduction phase for enhancing the signal clarity. Here, noises presented in the UAS are reduced using the developed Adaptive Residual Autoencoder with Spatial-Temporal Attention (AResASTA) model. Further, the parameters in the proposed AResAe-STA are tuned using the Renovated Random Attribute-based Golf Optimization Algorithm (RRA-GOA). It can effectively handle large volumes of data promptly to generate optimal solutions. This allows the clear detection of UAS, leading to improved signal recognition and classification. At last, the noise reduction outcomes are attained from the developed AResAe-STA mechanism. Further, various performance measures are used to optimally validate the system performance, and it is compared with several existing methods to observe its effectiveness. The proposed method's overall performance is maximized by 7.49% of DPTN, 5.47% of DPRN, 3.21% of BSS, and 1.05% of AResASTA in terms of the PCC measure.
This study explores the geotechnical performance of expansive Black Cotton Soil (BC soil) when stabilized with Red Mud (RM), an industrial by-product, and varying percentages of lime to enhance its suitability as a pavement subgrade material. BC soil is characterised by high swelling and shrinkage behaviour, creating serious challenges in constructing pavements, embankments, and foundations. Simultaneously, RM spills resulting from dyke breaches raise environmental concerns, making its safe and beneficial utilisation a priority. Stabilizing RM with lime provides an environmentally sustainable approach for geotechnical applications. Laboratory investigations were performed with six lime dosages (3
Diabetic retinopathy (DR) represents a primary cause of vision impairment, highlighting the importance of early and precise detection to reduce its advancement. This study presents DiaRetULS-Net, an innovative Ensembled model developed for the automated detection and classification of diabetic retinopathy severity utilizing retinal fundus images. The proposed methodology utilizes advanced preprocessing techniques, such as Contrast Limited Adaptive Histogram Equalization (CLAHE) for image enhancement, alongside robust feature extraction methods including Discrete Wavelet Transform (DWT) and Local Binary Patterns (LBP) to effectively capture essential frequency and texture-based features. The DiaRetULS-Net architecture combines U-Net for accurate segmentation of retinal abnormalities, the Liquid Time Constant Neural Network (LTCN) for the extraction of dynamic spatial and temporal features, and a Multi-Class Support Vector Machine (SVM) for precise classification of diabetic retinopathy severity levels. The model was assessed using the Messidor-2 dataset and a 5-fold crossvalidation approach, resulting in notable performance metrics: 98.83% accuracy, 98.87% specificity, and 99.21% sensitivity. Comprehensive analyses, such as the Receiver Operating Characteristic (ROC) curve, confusion matrix, and error histogram, substantiate the model's reliability and efficiency.
