Transactions on Energy Systems and Engineering Applications

HyTra: Hyperclass Transformer for WiFi Fingerprinting-based Indoor Localization

Muneeb Nasir — 2024-02-13

The emerging demand for a variety of novel Location-based Services (LBS) by consumers and industrial users is driven by the rapid and extensive proliferation of mobile smart devices. Sensors embedded in smart devices or machines provide wireless connectivity and Global Positioning System (GPS) capability, and are co-utilized to acquire location-linked data which are algorithmically transformed into reliable and accurate location estimates. GPS is a mature and reliable technology for outdoor localization but indoor localization in a complex multi-storey building environment remains challenging due to fluctuations in wireless signal strength arising from multipath fading. Location-linked data from wireless access points (WAPs) such as received signal strength (RSS) are acquired as numerical sequences. By conceptualizing a fixed order sequence of WAP measurements as a sentence where the RSS from each WAP are words, we may leverage on recent advances in artificial intelligence for natural language processing (NLP) to enhance localization accuracy and improve robustness against signal fluctuations. We propose the hyper-class Transformer (HyTra), an encoder-only Transformer neural network which learns the relative positions of wireless access points (WAPs) through multiple learnable embeddings. We propose a second network, HyTra-HF, which improves upon HyTra by applying a hierarchical relationship between location classes. We test our proposed networks on public and private datasets varying in sizes. HyTra-HF outperforms existing deep learning solutions by obtaining 96.7\% accuracy for the floor classification task on the UJIIndoorloc dataset. HyTra-HF is amenable to deep model compression and achieves accuracy of 95.95\% with over ten-fold reduction in model size using Sparsity Aware Orthogonal (SAO) initialization and has the best-in-class accuracy for the sparse model.

An Enhanced Energy Efficiency Routing for WSN based on Elephant Herding and Swarm Optimization Approaches

Robin Abraham — 2024-02-27

Energy utilization and inadequacy of sensor nodes are considered major drawbacks in wireless sensor networks (WSNs). This is because the sensor nodes use the battery for recharging energy. To overcome this issue WSN utilized a clustering-routing algorithm. This protocol divides the adjacent sensor nodes into separate clusters to choose a cluster head. Thus, the cluster head gathers information from all clusters and transmits it to the base station. In this article, the proposed method used cluster-based routing protocols to enhance energy efficiency and network lifetime. Moreover, this paper follows three stages to maximize energy efficiency. Initially, the clustering process is performed using dolphin swarm optimization (DSO), where a group of clusters is formed. Then the second stage is composed of cluster head selection among the group of clusters by elephant herding optimization (EHO) strategy. Finally, the collected data are necessary to forward to the base station for transferring the information. A specified path (routing) is selected by chicken swarm optimization (CSO). By using these algorithms, the network nodes support the balance of energy utilization. Experimental analysis proves when evaluated with existing methods the proposed technique has improved energy efficiency with an increase in network lifetime.

A switched-inductor switched-capacitor based ultra-gain boost converter: analysis and design

Neyyala Raju — 2024-02-29

A feature known as high-voltage gain conversion is necessary for a number of applications, including photovoltaic (PV) connected systems, UPS, SMPS, and some inverter applications, specifically for the power processing of low-voltage renewable sources. This article makes a suggestion for an ultra-gain boost converter based on a switched-inductor switched-capacitor (SISC) network. Ultra-voltage gain (> 15) and lower voltage stresses across the switches are the main benefits of the proposed converter. Additionally, compared with other high-gain topologies, the number of components decreases. This paper presents a systematic analysis of the proposed ultra-gain boost DC–DC converter along with a comparison to other topologies that have been previously published in the literature. The simulation model confirmed that the efficiency of the proposed topology is 95.23%.

Study of the properties of a composite material Fe78Si9B13 / GNP in an epoxy matrix

Marcelo Ruben Pagnola — 2024-04-12

This study investigates the properties of a composite material obtained by mixing Fe₇₈Si₉B₁₃ metallic powders (at %) with graphene nanoplates (GNP) in an epoxy matrix. Four composite types were created with GNP weight proportions of 0%, 0.5%, 1.0%, and 1.5%. The composites were embedded in transparent epoxy with weight proportions of 10%, 15%, and 20%, and then filled into 7 x 20 mm cylindrical probes. Twelve samples were prepared, and another 12 samples were subjected to a longitudinal magnetic field of 1 kG. All samples were tested with a Universal Testing Machine (Model WDW 10E) up to a maximum force of 20 kN. The experiment recorded deformation (ΔH) vs. charge force. Most samples showed a maximum compression resistance of 390 MPa, except for a few that did not exceed 100 MPa. The magnetically oriented samples showed a greater elastic limit in the range of 200 to 270 MPa. Optical microscopy was used to observe the ordering of the particles after the application of the magnetic field. Scanning electron microscopy, energy-dispersive X-ray spectroscopy, and X-ray diffraction were used to characterize the structure of the composite components. A vibrating sample magnetometer (VSM) was used to characterize the magnetic behavior of the metallic powders in the composite.