一种基于深度学习的异常数据清洗算法

在物联网(IoT)中采用合适的异常数据清洗算法能极大地提升数据质量.许多研究人员采用统计学方法或分类聚类等方法对时-空相关数据进行清洗.但这些方法需要额外的先验知识,会给汇聚节点带来额外的计算开销.该文根据低秩-稀疏矩阵分解模型,提出一种基于深度神经网络的快速异常数据清洗算法,来解决物联网中时-空相关数据的清洗问题.结...     

论智慧轨道交通及其系统架构

提倡用“智慧轨道交通”来描述未来的“高效、便捷、安全、可视、可预测、环保和智慧”的现代轨道交通行业和宏观系统的发展远景视图,并提出了一个“智慧轨道交通系统架构”。该架构以“智慧轨道交通智能化地面基础设施”（SRT-IGI）和智慧列车为基础实现更透彻的感知;以“轨道交通全联网”(SRT-IoT)作为信息交换与共享平台实现更广泛的互联互通;以“与人类融合智慧管理、决策与指挥”（HCA-IMDC）应用系统为顶层结构。三部分组成结构及分布其中的“智能化”功能共同实现整个交通行业的“智慧化”。同时分别对三部分的技术内涵作了进一步的探讨。鉴于笔者尚未发现从信息技术（IT）角度全面探讨“智慧轨道交通”的文献,所以希望本框架的提出能够为从信息技术（IT）角度全面研究智慧轨道交通提供参考,进一步推动智慧轨道交通在技术上的跨越式发展,使我国轨道交通以“更透彻的感知、更广泛的互联互通、更深入的智能化处理能力”的雄姿,为国家经济建设发展和人性化地为人民服务。     

基于系统论的产品低碳设计质性研究

目的 建筑与工业产品的低碳设计策略有明显差异,其中由属性差异衍生出不同的设计形式。将低碳设计与对象属性的联系作为研究目标,提出在产品设计中,对象属性决定低碳设计形式的假设。方法 将建筑“高技化”与产品“物联网化”进行比较研究,探明其中的联系。基于系统理论从整体性、相关性、动态性、目的性、层次性五个方面进行分析,得出两类人造物在属性上的差异。基于系统设计中的功能、输入输出、程序与层次、环境、媒介及人六个方面,最终形成六个对比因子。以此判断低碳设计与对象属性间的联系,进而解释其设计表象上的差异。结果 对象属性决定低碳设计的形式。结论 对象属性对“功能”“能量转换”“环境”“媒介”四个设计因子产生影响,而对“人”和“程序”两个设计因子不产生影响,设计“人”和“程序”两个因子的方法带有普适性。     

Toward intelligent resource management in dynamic Fog Computing-based Internet of Things environment with Deep Reinforcement Learning: A survey

Fog computing has already started to gain a lot of momentum in the industry for its ability to turn scattered computing resources into a large-scale, virtualized, and elastic computing environment. Resource management (RM) is one of the key challenges in fog computing which is also related to the success of fog computing. Deep learning has been applied to the fog computing field for some time, and it is widely used in large-scale network RM. Reinforcement learning (RL) is a type of machine learning algorithms, and it can be used to learn and make decisions based on reward signals that are obtained from interactions with the environment. We examine current research in this area, comparing RL and deep reinforcement learning (DRL) approaches with traditional algorithmic methods such as graph theory, heuristics, and greedy for managing resources in fog computing environments (published between 2013 and 2022) illustrating how RL and DRL algorithms can be more effective than conventional techniques. Various algorithms based on DRL has been shown to be applicable to RM problem and proved that it has a lot of potential in fog computing. A new microservice model based on the DRL framework is proposed to achieve the goal of efficient fog computing RM. The positive impact of this work is that it can successfully provide a resource manager to efficiently schedule resources and maximize the overall performance.     

Household flow detection using FEAT (flow estimating accelerometer-thermometer) device

The use of IoT devices in water end use disaggregation verification is an emerging field which offers benefits over conventional approaches, in terms of cost, accuracy and scalability. Having reliably disaggregated water appliance consumption data will enable smart water meter data to be used in household water conservation approaches and for understanding water consumption behaviours. The FEAT device provides a low cost, easily applied and scalable solution that is demonstrated to work even for very low flow conditions of 0.03 l/s. The FEAT device is a combination of a battery, Wi-fi board and MPU6050 sensors providing multi-modal accelerometer and thermometer data. The study places 7 of these FEAT devices onto hot and cold water pipes leading to a shower, which is operated 4 times in a high flow situation, 0.13 l/s, and 4 times in a low flow situation, 0.03 l/s. The data is then analysed and compared with a flow logger to determine if the FEAT device can detect when a domestic appliance is using water. There are limiting cases where the level of noise or external interference limits distorts the data, obscuring the distinguishable peaks in the data due to the similarity of the values. By using high and low pass filtering methods it was possible to enhance the peaks but there are still situations where peaks cannot be detected: for example, if a rigid pipe is not able to vibrate easily or if a hot water boiler is not triggered due to the low flow rate. However, the results show it should be possible to overcome these limiting cases, as it is much less likely for both the vibration and temperature data to be adversely affected by noise or external influences simultaneously, therefore decreasing the effect of noise and external influences. In conclusion, this research paper demonstrates that FEAT devices are a low cost, easily applied and scalable solution for detecting flow. By using high and low pass filtering, placing sensors on freely moving pipes and through the use of multi-modal verification, the FEAT device is shown to work on both metal and plastic pipes even in the lowest flow situations of 0.03 l/s. Therefore the FEAT device is a suitable solution for appliance identification in disaggregation verification datasets.     

Integration of Internet of Things and blockchain technology in healthcare domain: A systematic literature review

Healthcare is a vitally important field in the industry and evolving day by day in the aspect of technology, services, computing, and management. Its potential significance can be increased by incorporating Internet of Things (IoT) technology to make it smart in the aspect of automating activities, which is then further reformed in the healthcare domain with the help of blockchain technology. Blockchain technology provides many features to IoT-based healthcare domain applications such as restructuring by securing traditional practices, data management, data sharing, patient remote monitoring, and drug analysis. In this study, a systematic literature review has been carried out in which a total of 52 studies were selected to conduct systematic literature review from databases PubMed, IEEE Access, and Scopus; the study includes IoT technology and blockchain integration in healthcare domain-related application areas. This study also includes taxonomy that mentions the aspects and areas in healthcare domain incorporating the traditional system with the integration of IoT and blockchain to provide transparency, security, privacy, and immutability. This study also includes the incorporation of related sensors, platforms of blockchain, the objective focus of selected studies, and future directions by incorporating IoT and blockchain in healthcare domain. This study will help researchers who want to work with IoT and blockchain technology integration in healthcare domain.     

Optimized Resource Allocation for IoT-D2D Communication Using WSN

The need for a strong system to access radio resources demands a change in operating frequency in wireless networks as a part of Radio Resource Management (RRM). In the fifth-generation (5G) wireless networks, the capacity of the system is expected to be enhanced by Device-to-Device (D2D) communication. The cooperation and Resources Allocation (RA) in the development of Internet of Things (IoT) enabled 5G wireless networks are investigated in this paper. Developing radio RA methods for D2D communication while not affecting any Mobile Users’ (MU) communication is the main challenge of this research. Distinct performance goals such as practising equality in the rates of user data, increasing Network Throughput (NT), and reducing End-to-End Delay (EED) are achieved by RA. The study undertaken on optimising performance for various wireless networks is focused on in this research work. Proposing a polynomial-time Proportional Fair Resource Allocation Method (PFRAM), which considers the MU’s rate requirements, is the prime objective of this paper. Any Resource Allocation Method (RAM) can be used by the proposed method for MU, and the time and differing location channel conditions are the factors to be adapted with. Allotting more than one resource block is allowed by our PFRAM to a D2D pair. The automatic maintenance of battery-less IoT wireless devices’ energy level is done potentially using an Extensible Energy Management System (EEMS). Finally, the device’s Node Transmission Power (NTP) can be managed using an Energy-Saving Algorithm (ESA) designed in this work for Node Uplink Data Transmission (NUDT). The trade-off between the Packet Loss Rate (PLR) and NTP is balanced by the algorithm. The cost of NUDT’s average Energy Consumption (EC) is reduced by locating the optical NTP. In order to free much bandwidth for wireless information, NUDT conserves the harvested energy for minimising Radio Frequency (RF) Energy Transmission (ET). MATLAB simulations are used to assess the proposed EEMS. The IoT device’s NTP is managed using ESA designed for NUDT. The significant minimisation of channel hopping EED and the selection of the premium quality communication channel by the proposed framework are indications of the simulation results. 67.19% is the bandwidth to transmit DPs with the Bandwidth Allocation Algorithm (BAA), which is greater than the cases in its absence.     

Machine learning-based clustering protocols for Internet of Things networks: An overview

The Internet of Things (IoT) continues to expand the current Internet, opening the door to a wide range of novel applications. The increasing volume of the IoT requires effective strategies to overcome its challenges. Machine Learning (ML) has led to a growing technology that enables computers to solve problems without the need for knowledge of their intricate details. Over the past years, various ML techniques have been used to efficiently manage IoT networks. Clustering is a technique that has proven its performance in the networking domain. Many works in the literature have studied ML-based clustering methods for IoT networks, including their main properties, characteristics, underlying technologies, and open issues. In this paper, we focus on topology-centered ML-based clustering protocols for IoT networks. Specifically, we investigate the potential benefits of adopting the clustering approach to address several IoT challenges. Moreover, we provide a comprehensive taxonomy of ML-based clustering algorithms for IoT networks. Finally, we statistically analyze the incorporation of ML techniques for clustering in various IoT systems and highlight the related open issues.     

Fairness constraint efficiency optimization for multiresource allocation in a cluster system serving internet of things

Massive, diverse, and high-frequency Internet of Things (IoT) applications pose challenges to the operation of cluster systems that serve it. Fair and efficient multidimensional resource allocation is of great significance to the sustainable operation of these systems. However, most of the existing cluster multiresource allocation optimization researches focus too much on the fairness of resource allocation and ignore the efficiency. The unbalanced use of multidimensional system resources reduces the effective utilization of system resources, which seriously affects the service quality of IoT applications. In this paper, we define the multiresource fair and efficient sharing optimization as a fairness-constrained efficiency optimization problem, which is from dynamics, discrete resources, and heterogeneous perspectives according to the characteristics of cluster system in practical. Moreover, we present a dynamic efficiency-aware multiresource fair allocation algorithm, DEF, which can improve the ability of the cluster system to serve diverse IoT applications. In the algorithm, large jobs schedule to the servers that expect the least remaining resources. Simulations performed using Google cluster-usage traces show that DEF can improve system resource utilization and guarantee the fairness of sharing among users.     

Optimized traffic engineering in Software Defined Wireless Network based IoT (SDWN-IoT): State-of-the-art,research opportunities and challenges

Wireless networks have been in focus since the last few decades due to their indispensable role in the future generation networks like the Internet of Things (IoT). However, the associated challenges in wireless network implementation such as distance, line-of-sight, interference, weather, power issues, etc., affect the performance adversely. Software Defined Networking (SDN) is a future generation networking technology and has been proven to alleviate the performance challenges in the existing wireless IoT networks. It helps to evolve the wireless IoT domain in the form of Software Defined Wireless Network based IoT (SDWN-IoT). Traffic Engineering (TE) has been part of traditional network designs since long back, to improve the performance of the communication networks. However, its more optimized forms and their usefulness in SDWN-IoT networks have been under active investigation. This work explores the existing literature related to the major types of SDWN-IoT networks namely, Software Defined Wireless Sensor Network based IoT (SDWSN-IoT) and Software Defined Wireless Mesh Network based IoT (SDWMN-IoT). Additionally, the article also draws some useful inferences, and compares respective contributions and shortcomings. Finally, various research opportunities and challenges have been discussed with respect to the SDWSN-IoT and SDWMN-IoT networks.