Remote communication via avatars in the metaverse has attracted much attention. However, in a PC environment that doesn’t have the motion tracking of a VR headset, it is difficult to determine whether the user's avatar has been continuously participating in a conversation. In this paper, we propose a communication system that represents whether avatars have been continuously participating in a conversation by reflecting users’ nods in avatars and changing the transparency of avatars according to how often they nod in a PC-based metaverse environment. In this study, a prototype was actually developed and evaluated through an evaluation experiment. As a result, it was shown that by changing the avatar's transparency according to how often it nods, it becomes clearer compared to using only the nodding function whether the avatar is listening or not, and whether it has been continuously participating in the conversation or not. The feedback method, reflected in the avatar's transparency change, is expected to be one future way to support communication in the metaverse using avatars.
In recent years, there are increasing demands in manufacturing industry for flexible production systems capable of handling high-mix, low-volume manufacturing, along with a growing population of elderly workers. As a result, reducing workers’ physical burdens and optimizing workspace layouts are urgent challenges. In particular, in repetitive tasks such as assembly and inspection performed at desks, the layout of equipment, worker posture, and movement trajectories between tasks significantly impact the operational efficiency and physical strain. Conventional methods often address these elements independently, resulting in limited effectiveness in achieving integrated spatial optimization. This paper proposes a mathematical optimization model that simultaneously optimizes worker postures, equipment layouts, and task-to-task trajectories for repetitive desk-based operations in manufacturing environments. The proposed method formulates constraints on equipment placement and worker posture by a Mixed-Integer Linear Programming (MILP) model, and integrates it with an A* pathfinding that accounts for obstacles and spatial constraints, enabling the derivation of practical workspace configurations. Furthermore, the model incorporates a dynamic optimization mechanism that reconstructs postures and paths across the entire task cycle during repetitive work, thereby aiming to balance long-term efficiency and safety. Simulation results confirm the effectiveness of the proposed method, demonstrating improvements in trajectory efficiency, reduction of postural strain, and rationalization of inter-equipment distances when compared with conventional layouts.
In this paper, we explore cutting-edge systems for tracking and monitoring cattle, with special attention to camels. We examine technologies like GPS, RFID tags, and wireless sensor networks to understand how each helps farmers manage animal behavior and movement. Our research shows how combining these tools with data analytics and machine learning makes tracking more effective. We discuss the pros and cons of injectable RFID tags, challenges with radio collars, and how IoT systems could revolutionize health monitoring. We also look at how computational systems spot health issues early by analyzing behavior patterns through satellite imagery and sensor data. These technologies give farmers unprecedented insight and control, boosting efficiency and sustainability in camel farming. In our tests, the wireless sensor network predicted theft with 92
NDN-VANET has been developed to enable robust end-to-end communication in VANETs, which are vulnerable networks due to the mobility of vehicles. The NDN functionality enables fast content delivery through network caching and content-centric communication. However, standard NDN does not incorporate any mechanism for priority control based on the significance of the requested data. This paper proposes a new caching method that utilizes spatial locality, based on context awareness of vehicular itineraries. Performance evaluation results confirm that the proposed method effectively improves the cache hit ratio enabling fast delivery of emergency content.
In recent years, for realizing a sustainable society, the deployment of renewable energy such as photovoltaic (PV) power and vehicles that emit no CO_2 , such as electric vehicles (EVs), is highly expected. On the other hand, due to large fluctuations in power generation throughout the day, further utilization of PV induces the duck curve, which degrades the efficiency of traditional thermal power generation. This paper discusses how to utilize PV while mitigating the duck curve by using battery swapping EVs (BSEVs). The proposed system uses the storage capacity of battery swapping stations (BSSs), which hold spare batteries for BSEVs, as stationary batteries for electricity time-shifting. In addition, the proposed system incorporates the mobility of BSEVs to coordinate capacity among BSSs, thereby improving the utilization of PV. To enable this coordination, the system employs two methods to mitigate traffic congestion caused by BSEV operations. This paper clarifies that these two methods effectively alleviate the duck curve though the performance evaluations.
When seeking people needing rescue assistance during disasters, such as earthquakes or fires, reliance on human personnel presents risks of delayed discovery because of constraints on labor and expertise, and the possibility that rescuers might be harmed by secondary disasters. Therefore, interest in using robotic exploration systems as alternatives is growing rapidly. In disaster situations, robots commonly use pre-existing maps to navigate difficult environments, such as indoor spaces, with numerous obstacles. However, indoor structures might change. Alternatively, no prior map might be available. For evaluation in this study, our developed drone system performs victim exploration and map creation simultaneously to investigate and navigate unknown indoor environments efficiently without pre-existing maps. After proposing multiple movement strategies and algorithms for the drone, we conducted simulations to compare the performance achieved by these algorithms.
AI applications are used in various applications by performing the training task T and inference task R of the machine learning (ML) on servers in data centers. Data processing and decision-making tasks are performed on edge nodes in the edge computing (EC) model of the IoT. Since data is locally processed on a small computer named edge node close to users, the responsiveness is improved, smaller energy is consumed, and private data is protected. In this paper, we newly propose an AEC (AI for Edge Computing) model where the inference task T is performed on edge nodes while the training task R is performed on servers. Parts of the inference task T are distributed on edge nodes, and a group of the edge nodes cooperates to process data from devices. Each edge node obtains output data by performing tasks on input data from devices and other edge nodes. Differently from the EC model, the inference task T changes as the training task R is performed, e.g., the execution time and output data change for the same input data. Each edge node exchanges output data with other edge nodes in addition to delivering the output data to devices. Here, an edge node has to find a succeeding edge node which can process the output data. We propose an algorithm to select an energy-efficient succeeding edge node under the condition where the inference task changes. In the evaluation, we show a succeeding edge node consuming smaller energy in the proposed algorithm.
Wireless Mesh Network (WMNs) offer a cost-effective communication, but finding the optimal mesh router allocation is an NP-hard problem. To deal with this issue, in our previous work, we implemented a hybrid intelligent system based on Particle Swarm Optimization (PSO), Hill Climbing (HC), and a Distributed Genetic Algorithm (DGA). In this paper, we implement in Genetic Algorithm of our simulation system four crossover methods (UNDX, BLX α , SPX, psBLX) and two mutation methods (Boundary Mutation and Uniform Mutation). We carry out a comparison study for these methods considering Subway distribution of mesh clients, Constriction Method (CM), and a middle scale WMN. The simulation results show that the combination of psBLX with Boundary Mutation achieves the best performance.
The facility layout problem remains a central focus in operations research. Researchers have been exploring its complexity and seeking efficient solutions. Identifying optimal solutions for diverse facility layouts frequently presents considerable challenges, resulting in either formidable barriers or necessitating substantial effort to get a globally optimal solution promptly. Recent findings indicate that metaheuristic algorithms, including genetic algorithms and Cuckoo search, are effective for addressing the facility layout problem. Conversely, each possesses distinct advantages and downsides. This research integrates two distinct metaheuristic algorithms to develop a hybrid solution. Our implementation exhibits rapid convergence and identifies superior solutions.
In modern intelligent transport systems, optimizing path selection in stochastic road networks has become crucial for autonomous vehicles and mobility on demand services. The unpredictability of travel times, due to factors like traffic congestion and weather conditions, necessitates a probabilistic approach to optimum path finding routing. This paper proposes an optimal stochastic routing algorithm that balances travel time and probability distribution in road networks where each road segment has uncertain travel times. We model the road network as a directed graph, with each edge representing a road segment characterized by multiple possible travel times and their associated probabilities. The algorithm identifies the shortest time path, a path with the highest probability of reaching the destination, and a trade-off path that balances travel time and the probability of reaching the destination. Simulation results on a Manhattan-like grid network demonstrate that the proposed method effectively reduces the risk of delays while maintaining reasonable travel times, thereby offering a practical solution for real-time path finding in uncertain traffic conditions.
In e-sports, monitors with frequencies higher than 240 Hz are commonly used, and while their use is said to offer advantageous for gameplay, the extent of their contribution remains unclear. Quantifying this effect is essential for creating a fair gaming environment. In this study, a program was developed to simulate monitors with refresh rates ranging from 60 Hz to 480 Hz, and two simple reaction tests were conducted using a 500 Hz monitor. When the color of a rectangle changed from blue to white, responses were faster with the higher refresh rate monitor. Conversely, when the white rectangle moved, responses were slower with the higher refresh rate monitor.
The PR- (Purpose and Read-data-from (RDF)) serializability and the EEPR (Energy-Efficient PR) scheduler are proposed to PR-serialize transactions based on purposes of transactions and RDF relation. The EEPR scheduler aborts some transactions to keep the consistency among objects. A transaction is re-started if the transaction is aborted. The execution time (ET) of each aborted transaction becomes longer since the EEPR scheduler re-schedules a re-started transaction without any prioritization. The EEPR-PAT (EEPR with Prioritization for Aborted Transactions) scheduler is newly proposed in this paper to perform aborted transactions prior to non-aborted transactions by extending the EEPR scheduler. In evaluation, we show the ET of aborted transactions in the EEPR-PAT scheduler becomes shorter compared with the EEPR scheduler.
In recent years, the increasing complexity and volume of information in industry-academia collaboration research have posed substantial challenges for effective knowledge discovery. To address this, the study proposes a semantic retrieval-based natural language processing (NLP) system designed to enhance the retrieval and analytical capabilities of textual data by integrating three main modules (retrieval, filtering, and analysis) and implementing both document-level and sentence-level retrieval strategies. Experimental results demonstrate that sentence-level retrieval outperforms document-level retrieval as well as the traditional lexical matching method BM25. Additionally, through word cloud visualization and analysis of temporal trends, this study highlights the evolution of topics in collaborative research, including research outcomes management and innovation of cross-organizational collaboration mechanisms. Although industry-academia collaboration is used as a case example, the proposed system is broadly applicable to other domains, facilitating the identification of emerging trends and overlooked areas, and providing valuable support for researchers and policymakers across various fields.
In this paper, we investigate applying ensemble methods to spatial data stream classification within the context of emergency services, specifically severe weather events. Emergencies demand swift, accurate decisions to mitigate impacts and protect lives. Ensemble methods improve accuracy predictions by combining outputs from multiple neural networks, each trained on diverse aspects of the data, including geographic coordinates, weather data, spatial data, and logistical factors. These models collectively contribute to more precise decision-making, particularly in assessing evacuation priorities. We collected and generated relevant data for affected regions and evacuation centres pertinent to severe weather events. To apply class labels, we utilize various clustering techniques. We find that the system architecture achieves high classification accuracy of spatial stream data, potentially leading to more effective emergency responses.
The concept of displaying the browsing history of web pages in the form of a tree was first introduced in the MosaicG browser in 1995. However, this feature is no longer available in contemporary browsers. This paper describes how to implement a plugin extension for the Chrome browser that displays the browsing history in a tree. The browsing history is recorded as a chronological list of the user’s browsing actions detected in the background. On request by the user, the extension generates a tree representing the browsing history from the user’s perspective by scanning the chronological list of actions in the following steps. First, the root node of the tree stores the URL of the first page visited. The current node, representing the page on which the user is currently focused, is initially set to the root node. When the user navigates to another page in the same tab, the current node moves accordingly. If the destination URL matches that of the parent or any existing child node, the current node moves to the respective parent or child node as if the move were a backward or forward move, respectively. Otherwise, a new node is created as a child of the current node, stores the URL of the destination page, and becomes the new current node. When active tabs are switched, the current node of the new active tab is marked and the current node of the inactive tab is retained persistently until the user returns to the tab again.
We investigate the accuracy of continuous nearest neighbour queries along several trajectory types. Many Location Based Services (LBS) for continuous k nearest neighbour queries utilize a safe region approach when processing the query locally on a user’s device. Some approaches, however, are approximate as the safe region may be missing points of interest (POIs) which makes the query result invalid. This paper further investigates a cluster-based safe region approach that is proposed in the literature. Past studies of this approach did not focus on the accuracy of random trajectories. Therefore, we evaluate the cluster-based strategy using different random path scenarios to determine how close to 100
This paper presents a study of the implementation of One-Time Passwords (OTP) as the primary authentication mechanism for home wired networks. The study also introduces reauthentication factors based on some predetermined thresholds. For a large number of Internet users, convenience is the number one priority. Unfortunately, security is neglected because of this. This study is divided into 3 main sections; 1. The implementation of the OTP server as the primary authentication method. 2. Introduction and testing of reauthentication mechanisms. 3. The usability test and finally the implementation of an email notification system that would inform the network owner of a reauthentication event. The goal of these projects is to ensure that users can easily and conveniently access their home networks while maintaining good security.
Many malware detection methods have been developed for Android. One way is to detect Android malware by representing Android applications (hereinafter, “apps”) as a method call graph. However, there are cases where common libraries are included in the analysis target, and if these libraries are also included in the graph, the detection accuracy is affected and also graph size is unnecessarily large because they represent behaviors that do not occur in the original application. One way to deal with common libraries is to add them to a white list and exclude them from analysis, but since there are many common libraries, it is difficult to exclude all of them in practice. In this study, we define two types of component graphs, devise an algorithm for generating component graphs, and propose an Android malware detection method using component graphs in order to improve the detection rate of Android malware. To confirm whether component graphs are effective for malware detection, we used graph neural networks as a classification algorithm and conducted experiments to evaluate the accuracy of Android malware detection using component graphs and method call graphs. The results showed that our proposed method using component graphs, which are much smaller graphs than method call graphs, succeeded in detecting malware with 96.18
During a disaster, quick and accurate assessment of the situation is required. However, information at disaster sites is fragmented, and collecting and organizing relevant data is time-consuming. Therefore, we propose a system that automatically summarizes, classifies, and visualizes damage situations based on disaster images posted by citizens. In this system, citizens upload disaster images with location information using a web form and the captions of the damage situation are generated from the images. The system uses a large-scale language model to automatically classify images under different damage categories from the generated captions and visualize them on a disaster map. This enables real-time structuring of images posted by citizens and intuitive understanding of disaster sites.
This study integrates the second-generation Robot Operating System (ROS 2) with embedded systems and proposes a secure one-to-one communication mechanism. The design incorporates access control, identity authentication, and message encryption using an out-of-band approach. By default, all UDP packets are blocked, disabling ROS 2’s underlying communication. To initiate communication, a node must first establish a TCP socket as an out-of-band channel to perform identity authentication and negotiate a session key. Upon successful authentication, the system whitelists the IP addresses to allow UDP traffic, enabling ROS 2 message exchange. Unauthorized nodes remain blocked, effectively enforcing access control. A hierarchical certificate authority and certificate chain verification ensure authenticity, while the session key guarantees the confidentiality and integrity of the communication.
