Explore the vast range of titles available.

Federated Learning supports collaborative model training across distributed clients while keeping sensitive data decentralized. Still, non-independent and identically distributed data pose challenges like unstable convergence and client drift. We propose Federated Normalized Loss-based Weighted Aggregation (FedNolowe) (Code is available at https://github.com/dongld-2020/fednolowe ), a new method that weights client contributions using normalized training losses, favoring those with lower losses to improve global model stability. Unlike prior methods tied to dataset sizes or resource-heavy techniques, FedNolowe employs a two-stage L1 normalization, reducing computational complexity by 40% in floating-point operations while matching state-of-the-art performance. A detailed sensitivity analysis shows our two-stage weighting maintains stability in heterogeneous settings by mitigating extreme loss impacts while remaining effective in independent and identically distributed scenarios.

In vehicular ad hoc networks (VANETs), many schemes for a multi-channel media access control (MAC) protocol have been proposed to adapt to dynamically changing vehicle traffic conditions and deliver both safety and non-safety packets. One such scheme is to employ both time-division multiple access (TDMA) and carrier-sense multiple access (CSMA) schemes (called a hybrid TDMA/CSMA scheme) in the control channel (CCH) interval. The scheme can adjust the length of the TDMA period depending on traffic conditions. In this paper, we propose a modified packet transmitted in the TDMA period to reduce transmission overhead under a hybrid TDMA/CSMA multi-channel MAC protocol. Simulation results show that a MAC protocol with a modified packet supports an efficient packet delivery ratio of control packets in the CCH. In addition, we analyze the hybrid TDMA/CSMA multi-channel MAC protocol with the modified packet under saturated throughput conditions on the service channels (SCHs). The analysis results show that the number of neighbors has little effect on the establishment of the number of time slots in TDMA periods and on SCHs under saturated throughput conditions.

Smart homes are an element of developing smart cities. In recent years, countries around the world have spared no effort in promoting smart cities. Smart homes are an interesting technological advancement that can make people’s lives much more convenient. The development of smart homes involves multiple technological aspects, which include big data, mobile networks, cloud computing, Internet of Things, and even artificial intelligence. Digital information is the main component of signal control and flow in a smart home, while information security is another important aspect. In the event of equipment failure, the task of safeguarding the system’s information is of the utmost importance. Since smart homes are automatically controlled, the problem of mobile network security must be taken seriously. To address these issues, this paper focuses on information security, big data, mobile networks, cloud computing, and the Internet of Things. Security efficiency can be enhanced by using a Secure Hash Algorithm 256 (SHA-256), which is an authentication mechanism that, with the help of the user, can authenticate each interaction of a given device with a WebServer by using an encrypted username, password, and token. This framework could be used for an automated burglar alarm system, guest attendance monitoring, and light switches, all of which are easily integrated with any smart city base. In this way, IoT solutions can allow real-time monitoring and connection with central systems for automated burglar alarms. The monitoring framework is developed on the strength of the web application to obtain real-time display, storage, and warning functions for local or remote monitoring control. The monitoring system is stable and reliable when applying SHA-256.

The air quality index (AQI) forecast in big cities is an exciting study area in smart cities and healthcare on the Internet of Things. In recent years, a large number of empirical, academic, and review papers using machine learning (ML) for air quality analysis have been published. However, most of those studies focused on traditional centralized processing on a single machine, and there had been few surveys of federated learning (FL) in this field. This overview aims to fill this gap and provide newcomers with a broader perspective to inform future research on this topic, especially for the multi-model approach. In this survey, we went over the works that previous scholars have conducted in AQI forecast both in traditional ML approaches and FL mechanisms. Our objective is to comprehend previous research on AQI prediction including methods, models, data sources, achievements, challenges, and solutions applied in the past. We also convey a new path of using multi-model FL, which has piqued the computer science community’s interest recently.

With the continuous development of a society in which construction works are ubiquitous, investors often accelerate the progress of construction projects and may neglect to follow the proposed construction processes. As a result, construction works decrease in durability and become more susceptible to cracking, breakages, or collapse, resulting in significant damage to people and property. Meanwhile, labour to monitor the progress of works is often time-consuming, effortful, prone to errors, and vulnerable to risk. This study introduces a system that uses a Wireless Sensor Network (WSN) consisting of a laser-ranging Time-of-Flight (ToF) sensor module for structural crack monitoring in architecture. The proposed system collects data on the variation in amplitude of an existing crack as well as environmental factors that can adversely affect cracks such as temperature, humidity, and vibration. This aids the assessment of safety levels by providing warnings and remedies as quickly as possible, contributing to raising the safety level for workers and people living in the area, and minimizing potential damage. Test results demonstrate the following advantages: the device operates stably in extreme weather conditions; has an expected lifetime of more than ten years; and features high accuracy, long data transmission, and low-cost, low-power operation.

Along with the development of the Internet of Things (IoT), waste management has appeared as a serious issue. Waste management is a daily task in urban areas, which requires a large amount of labour resources and affects natural, budgetary, efficiency, and social aspects. Many approaches have been proposed to optimize waste management, such as using the nearest neighbour search, colony optimization, genetic algorithm, and particle swarm optimization methods. However, the results are still too vague and cannot be applied in real systems, such as in universities or cities. Recently, there has been a trend of combining optimal waste management strategies with low-cost IoT architectures. In this paper, we propose a novel method that vigorously and efficiently achieves waste management by predicting the probability of the waste level in trash bins. By using machine learning and graph theory, the system can optimize the collection of waste with the shortest path. This article presents an investigation case implemented at the real campus of Ton Duc Thang University (Vietnam) to evaluate the performance and practicability of the system’s implementation. We examine data transfer on the LoRa module and demonstrate the advantages of the proposed system, which is implemented through a simple circuit designed with low cost, ease of use, and replace ability. Our system saves time by finding the best route in the management of waste collection.

Software-Defined Networking (SDN) has opened a promising and potential approach for future networks, which mostly requires the low-level configuration to implement different controls. With the high advantages of SDN by decomposing the network control plane from the data plane, SDN has become a crucial platform to implement Internet of Things (IoT) services. However, a static SDN controller placement cannot obtain an efficient solution in distributed and dynamic IoT networks. In this paper, we investigate an optimization framework under a well-known theory, namely submodularity optimization, to formulate and address different aspects of the controller placement problem in a distributed network, specifically in an IoT scenario. Concretely, we develop a framework that deals with a series of controller placement problems from basic to complicated use cases. Corresponding to each use case, we provide discussion and a heuristic algorithm based on the submodularity concept. Finally, we present extensive simulations conducted on our framework. The simulation results show that our proposed algorithms can outperform considered baseline methods in terms of execution time, the number of controllers, and network latency.

This paper presents an energy-saving and lifetime-maximizing solution for the LoRa wireless mesh network (WMN). Energy dissipation is a crucial factor affecting the usability of the LoRa WMN. In the worst cases, in systems without electric mains, the life of a sensor node battery may last for only a few hours. Two proposed solutions are characterized as energy-saving due to the use of deep sleep in the ESP8266-12F microcontroller. This allows the optimization of duty cycling, which refers to the ratio between active and inactive periods of sensor nodes power-gating the node, i.e. turning off all circuitries. This solution benefits applications using active power-hungry sensors sampled many times daily. Notably, reducing power consumption during idle time increases the optimal battery life by up to hundreds of times. As a result, the automatic uptime of a LoRa WMN can increase from days to months or even years, depending on usage. Therefore, energy-saving must be optimized if the node is to be installed in locations without a grid or renewable energy source. Experimental results show that the proposed energy-saving solution is more effective than those introduced by previous studies.

As digitization is integrated into daily life, media are increasingly transferred over the Internet. Voice-over-Internet Protocol (VoIP), the most popular media transfer technology, is attracting many researchers and investments. The application of Artificial Intelligence (AI) technology into the Private Branch Exchange (PBX) has played a pivotal role in enhancing the customer experience and is able to unite employees in any company. One technology application used to optimize customer experience in a call centre is the use of an automatic PBX integrated with a Virtual Assistant (VA), which interacts directly with the PBX through voice and in multiple languages without any keystrokes. The Interactive Voice Response (IVR) module forwards the customer's call to an operator or supports automatic processing. This solution can help businesses to handle thousands of calls per day with optimal performance, thus creating a customer care campaign that quickly reaches many users. A PBX integrated with Vietnamese Virtual Assistants (VVA) on an AI technology platform will also help businesses to cut down on operator costs with automated calls. Through comparison with a traditional PBX, this article analyzes, evaluates and optimizes an automatic PBX system with integrated VVA, thereby offering efficient solutions for interest companies.

Heterogeneous populations of cerebellar Purkinje cells (PCs) are arranged into separate longitudinal stripes, which have different topographic afferent and efferent axonal connections presumably involved in different functions, and also show different electrophysiological properties in firing pattern and synaptic plasticity. However, whether the differences in molecular expression that define heterogeneous PC populations affect their electrophysiological properties has not been much clarified. Since the expression pattern of many of such molecules, including glutamate transporter EAAT4, replicates that of aldolase C or zebrin II, we recorded from PCs of different \"zebrin types\" (zebrin-positive = aldolase C-positive = Z+; and Z-) in identified neighboring stripes in vermal lobule VIII, in which Z+ and Z- stripes occupy similar widths, in the Aldoc-Venus mouse cerebellar slice preparation. Regarding basic cellular electrophysiological properties, no significant differences were observed in input resistance or in occurrence probability of types of firing patterns between Z+ and Z- PCs. However, the firing frequency of the tonic firing type was higher in Z- PCs than in Z+ PCs. In the case of parallel fiber (PF)-PC synaptic transmission, no significant differences were observed between Z+ and Z- PCs in interval dependency of paired pulse facilitation or in time course of synaptic current measured without or with the blocker of glutamate receptor desensitization. These results indicate that different expression levels of the molecules that are associated with the zebrin type may affect the intrinsic firing property of PCs but not directly affect the basic electrophysiological properties of PF-PC synaptic transmission significantly in lobule VIII. The results suggest that the zebrin types of PCs in lobule VIII is linked with some intrinsic electrophysiological neuronal characteristics which affect the firing frequency of PCs. However, the results also suggest that the molecular expression differences linked with zebrin types of PCs does not much affect basic electrophysiological properties of PF-PC synaptic transmission in a physiological condition in lobule VIII.