19 July 2021 in IEEE Internet of Things Journal
Internet of Things (IoT) is a wireless network of various battery-powered sensing units. Due to the limited battery capacity, the nonaccessible/abandoned nodes demand more energy to be reached or reintegrated to keep the network connected. When a specific tree topology for routing, called destination-oriented directed acyclic graphs (DODAGs), is used, isolated nodes spend maximum energy on the assigned task during data transformation from the sensor field to the DODAG root. The nodes closer to the DODAG root need to rely on faraway nodes and resource-burden-constrained nodes to lead to the quick energy drain. It brings an idea of IoT network nodes in which an extra amount of energy provide to the longer time alive nodes. This article proposes an ARFOR-adaptive ranking fuzzy-based energy-efficient opportunistic routing protocol for sustainable IoT applications. The proposed protocol consists of a parent node (PN) that acts as a head node in a cluster to aggregate the packets to DODAG root; and a volunteer node (VN) acts as a forwarder to transfer the packets to PN with threshold energy limits to increase network lifetime during the transmission cycle. The proposed VN selection is based on fuzzy parameters, such as Canberra distance, residual energy, and threshold. The simulation outcomes depict that the ARFOR fairly justifies the network timeline requirement with maximum percentage area coverage. The percentage gain in terms of network lifetime is comparatively significant for a lower number of VN.
8 May 2021 in International Journal for Numerical Methods in Biomedical Engineering
1 January 2022 in BioMed Research International
Growth of malignant tumors in the breast results in breast cancer. It is a cause of death of many women across the world. As a part of treatment, a woman might have to go through painful surgery and chemotherapy that may further lead to severe side effects. However, it is possible to cure it if it is diagnosed in the initial stage. Recently, many researchers have leveraged machine learning (ML) techniques to classify breast cancer. However, these methods are computationally expensive and prone to the overfitting problem. A simple single‐layer neural network, i.e., functional link artificial neural network (FLANN), is proposed to overcome this problem. Further, the F‐score is used to reduce the issue of overfitting by selecting features having a higher significance level. In this paper, FLANN is proposed to classify breast cancer using Wisconsin Breast Cancer Dataset (WBCD) (with 699 samples) and Wisconsin Diagnostic Breast Cancer (WDBC) (with 569 samples) datasets. Experimental results reveal that the proposed models can diagnose breast cancer with higher performance. The proposed model can be used in the early breast cancer diagnosis with 99.41% accuracy.
22 January 2021 in Journal of Real-Time Image Processing
1 February 2022 in International Journal of Sensors, Wireless Communications and Control
Aims & Objective: The fast depletion of fossil fuels and the growing awareness of environmental protection has become a concerning topic. Because of this fact, the researchers are working for a long time to generate electrical energy sources due to the intermittent nature of unconventional energy sources such as solar, wind geothermal, tidal, and biomass as a sustainable, cost-effective, and environmentally friendly alternative for conventional energy sources. These systems are interconnected and full-fill demands as well as energy storage, which subsequently formed a complex hybrid renewable energy system. Hence, forecasting of energy generation, sizing of equipment is essential for the economic feasibility of a complex hybrid system. Also necessary for the design analysis. Methodology: In this research article, the proposed Functional Link Convolutional Neural Network (FLCNN) is applied to forecast the energy generation from the hybrid solar and wind energy system. Also, the Jaya algorithm has been applied to find the optimal sizing of the solar and wind based hybrid renewable energy system. Results & Discussion: The proposed method is simple in design and implementation, and it also reduces computational complexity and time. The proposed FLCNN technique has been compared with various other Machine Learning (ML) methodology, such as Convolutional Neural Network (CNN), Random Forest (RF), and Xg-Boost. In sizing, Jaya is compared with other heuristic techniques such as Genetic Algorithm (GA), Particle Swarm Optimization (PSO), and Cat Swarm Optimization (CSO). Conclusion: The proposed FLCNN and Jaya optimization techniques successfully applied for tasks like energy forecasting and sizing of the renewable energy system.
1 October 2017 in Journal of Computer Science
Mobile Ad Hoc Network (MANET) is an auto Configuring network. Due to its natural characteristics, a MANET is vulnerable to many security threats. Blackhole attack compromises the performance and the reliability of the network. Since nodes are allowed to move freely within the network, it becomes very important to protect the communication among mobile nodes for the sake of security. In this paper we have investigated various techniques that can detect Blackhole attacks in MANET and we have compared the detection techniques with different matrices such as Average Packet Delivery ratio and Average End-To-End delay.