Sensor Data Streams Research Articles

Streaming service provisioning in IoT‐based healthcare: An integrated edge‐cloud perspective

AbstractBio‐sensor data streaming and analytics is a key component of smart e‐healthcare. However, existing Internet of Things (IoT) ecosystem is unable to materialize the real‐time bio‐sensor data streaming and analytics within resource constrained environment. Moreover, traditional solutions fail to mitigate the edge‐cloud integration within a single sub‐system under IoT periphery which lead to investigate how edge‐cloud hybridization could be realized via similar set of tools. The objective of this article is to implement an integrated dual‐mode edge‐cloud system to serve streaming and analytics in real‐time. This study aims to achieve the aforesaid goal by presenting two different experiments that deals with the real‐time pulse sensor data streaming and analytics while utilizing light‐weight IoT‐supported JavaScript frameworks that includes Node.js, Johnny‐Five, Serialport.js, Plotly client, Flot.js, jQUERYy, Express Server, and Socket.io. Firstly, a standalone IoT‐edge system is developed and later, an integrated IoT‐based edge‐cloud system is developed to compare between the effectiveness of the systems. The implementation results show near correlation between the standalone edge and dual‐mode edge system. However, the dual‐mode edge‐cloud system provides more flexibility and capability to counter the bio‐sensor data streaming and analytics services within the constrained framework.

Knowledge Discovery Using Topological Analysis for Building Sensor Data.

Distributed sensor networks are at the heart of smart buildings, providing greater detail and valuable insights into their energy consumption patterns. The problem is particularly complex for older buildings retrofitted with Building Energy Management Systems (BEMS) where extracting useful knowledge from large sensor data streams without full understanding of the underlying system variables is challenging. This paper presents an application of Q-Analysis, a computationally simple topological approach for summarizing large sensor data sets and revealing useful relationships between different variables. Q-Analysis can be used to extract novel structural features called Q-vectors. The Q-vector magnitude visualizations are shown to be very effective in providing insights on macro behaviors, i.e., building floor behaviors in the present case, which are not evident from the use of unsupervised learning algorithms applied on individual terminal units. It has been shown that the building floors exhibited distinct behaviors that are dependent on the set-point distribution, but independent of the time and season of the year.

Embedded Digital Twins in future energy management systems: paving the way for automated grid control

Abstract Emerging real-time applications in information technology, and operational technology enable new innovative concepts to design and operate cyber-physical systems. A promising approach, which has been discovered recently as key technology by several industries is the Digital Twin (DT) concept. A DT connects the virtual representation of a physical object, system or process by available information and sensor data streams, which allows to gather new information about the system it mirrors by applying analytic functions. Thereby the DT technology can help to fill sensor data gaps, e. g., to support anomaly detection, and to predict future operating conditions and system states. This paper discusses a dynamic power system DT as a cornerstone instance of a new generation of EMS, and a prospective new EMS architecture, to support the increasingly complex operation of electric power systems. Unlike in traditional offline power system models, the parameters are updated dynamically using measurement information from the supervisory control and data acquisition (SCADA) and a wide area monitoring system (WAMS) to tune the model. This allows to derive a highly accurate virtual representation of the mirrored physical objects. A simulation engine, the Digital Dynamic Mirror (DDM) is introduced, in order to be able to reproduce the state of a reference network in real-time. The validation of the approach is carried out by a case study. In a closed loop within EMS applications, the DDM can help to assess contingency mitigation strategies, thus it can support the decision-making process under variable system conditions. The next generation of control centre Energy Management System (EMS) can benefit from this development by augmentation of the dynamic observability, and the rise of operator situation awareness.

Data-driven Performance Fault Detection in Commercial Wind Turbines

The operation and maintenance costs of wind parks make up a major fraction of a park's overall lifetime costs. They also include opportunity costs of lost revenue from avoidable power generation underperformance. We present a machine-learning based decision support method that minimizes these opportunity costs. By analyzing the stream of telemetry sensor data from the turbine operation, estimating highly accurate power reference relations and benchmarking, we can detect performance-related operational faults in a turbine- and site-specific manner. The most accurate power reference model is selected based on combinations of machine learning algorithms and regressor sets. Operating personnel can be alerted if a normal operating state boundary is exceeded. We demonstrate the performance fault detection method in a case study for a commercial grid-connected onshore wind turbine. Diagnosing a detected underperformance event, we find that the observed power generation deficiencies coincide with rotor blade misalignment related to low hydraulic pressure of the turbine's blade actuators.

A distributed sensing data anomaly detection scheme

With the continuous development of Internet of Things, the information society has gradually entered a new era of the Internet of everything. Sensor nodes are important sources of data in the Internet of Things. The abnormal and failure of sensing data in the Internet of Things will affect the connectivity of the network. If the accuracy and reliability of the corresponding perception data can be effectively improved, we can timely and accurately find out the emergency and monitor the working status of the network. Therefore, it is of great significance to detect the abnormal data of data streams in the sensor network nodes and confirm its source. For the low quality of sensor data collected in real time in IoT, this paper proposes an anomaly detection method for sensing data streams based on edge computing. In this algorithm, the sensor data is expressed in the form of time series. On the edge computing based sensor data anomaly detection model, the improved confidence interval is used to detect whether the data is abnormal. The concept of interval difference is proposed as the judgment of the source of the anomaly. The accuracy and effectiveness of the algorithm are verified by experiments. The results show that the detection rate of abnormal data is above 98%, which indicates that the algorithm has certain practicability.

The security of vulnerable senior citizens through dynamically sensed signal acquisition

AbstractTraditional signal recognition methods generally use biosensors for signal acquisition. With senior citizens, sensor signal acquisition will be affected by their movements. These signal fluctuations are large, and if the signal area cannot be fixed, it may result in problems such as data loss. The most important issue caused by data loss is the safety for vulnerable seniors. Therefore, here we study abnormal behavior recognition based on dynamic sensing. In this paper, we look to improve the problems that exist in traditional methods. Using the SW‐520D sensor, activity signals of the elderly are first collected. By comparing the received signal strength sets, dynamic sensor data flow of the abnormal behavior for senior citizens can be determined. A multiple linear regression estimation method is used to solve the problem of data loss in dynamic sensor data flow environments. We obtain system parameter thresholds in both area isolation and segmentation using the stochastic resonance method. From this, a direct notch is constructed that enters the dynamic sensor data stream, and the interference component filtering of abnormal behavior signals is processed. The amplitude‐frequency response feature extraction method is used for high‐precision isolation and segmentation of abnormal behavior signal areas such as falls, improving the accuracy of senior behavior signal recognition, and realizing safety monitoring for the elderly. The improved method was used to identify the signals of abnormal behaviors of young people. The minimum recognition error rate was only 2%, the recognition accuracy rate was as high as 98%, and the calculation time was only 19 ms.

Detecting changes in the second moment structure of high-dimensional sensor-type data in a K-sample setting

The K sample problem for high-dimensional vector time series is studied, especially focusing on sensor data streams, in order to analyze the second moment structure and detect changes across samples and/or across variables cumulated sum (CUSUM) statistics of bilinear forms of the sample covariance matrix. In this model, K independent vector time series are observed over a time span which may correspond to K sensors (locations) yielding d-dimensional data as well as K locations where d sensors emit univariate data. Unequal sample sizes are considered as arising when the sampling rate of the sensors differs. We provide large-sample approximations and two related change point statistics, a sum of squares and a pooled variance statistic. The resulting procedures are investigated by simulations and illustrated by analyzing a real data set.

Weakly Supervised Multi-Task Representation Learning for Human Activity Analysis Using Wearables

Sensor data streams from wearable devices and smart environments are widely studied in areas like human activity recognition (HAR), person identification, or health monitoring. However, most of the previous works in activity and sensor stream analysis have been focusing on one aspect of the data, e.g. only recognizing the type of the activity or only identifying the person who performed the activity. We instead propose an approach that uses a weakly supervised multi-output siamese network that learns to map the data into multiple representation spaces, where each representation space focuses on one aspect of the data. The representation vectors of the data samples are positioned in the space such that the data with the same semantic meaning in that aspect are closely located to each other. Therefore, as demonstrated with a set of experiments, the trained model can provide metrics for clustering data based on multiple aspects, allowing it to address multiple tasks simultaneously and even to outperform single task supervised methods in many situations. In addition, further experiments are presented that in more detail analyze the effect of the architecture and of using multiple tasks within this framework, that investigate the scalability of the model to include additional tasks, and that demonstrate the ability of the framework to combine data for which only partial relationship information with respect to the target tasks is available.

Adversarial Multi-view Networks for Activity Recognition

Human activity recognition (HAR) plays an irreplaceable role in various applications and has been a prosperous research topic for years. Recent studies show significant progress in feature extraction (i.e., data representation) using deep learning techniques. However, they face significant challenges in capturing multi-modal spatial-temporal patterns from the sensory data, and they commonly overlook the variants between subjects. We propose a Discriminative Adversarial MUlti-view Network (DAMUN) to address the above issues in sensor-based HAR. We first design a multi-view feature extractor to obtain representations of sensory data streams from temporal, spatial, and spatio-temporal views using convolutional networks. Then, we fuse the multi-view representations into a robust joint representation through a trainable Hadamard fusion module, and finally employ a Siamese adversarial network architecture to decrease the variants between the representations of different subjects. We have conducted extensive experiments under an iterative left-one-subject-out setting on three real-world datasets and demonstrated both the effectiveness and robustness of our approach.

Low Power Unsupervised Anomaly Detection by Nonparametric Modeling of Sensor Statistics

This article presents anomaly detection by examining sensor stream statistics (AEGIS), a novel mixed-signal framework for real-time AEGIS. AEGIS utilizes kernel density estimation (KDE)-based nonparametric density estimation to generate a real-time statistical model of the sensor data stream. The likelihood estimate of the sensor data point can be obtained based on the generated statistical model to detect outliers. We present CMOS Gilbert Gaussian cell-based design to realize Gaussian kernels for KDE. For outlier detection, the decision boundary is defined in terms of kernel standard deviation ( $\sigma _{\mathrm{ Kernel}}$ ) and likelihood threshold ( $P_{\mathrm{ Thres}}$ ). We adopt a sliding window to update the detection model in real time. We use time-series data set provided from Yahoo to benchmark the performance of AEGIS. A ${f}$ 1-score higher than 0.87 is achieved by optimizing parameters such as length of the sliding window and decision thresholds which are programmable in AEGIS. Discussed architecture is designed using 45-nm technology node and our approach on average consumes ~75- $\mu \text{W}$ power at a sampling rate of 2 MHz while using ten recent inlier samples for density estimation.

An Integrated Approach of Belief Rule Base and Deep Learning to Predict Air Pollution.

Sensor data are gaining increasing global attention due to the advent of Internet of Things (IoT). Reasoning is applied on such sensor data in order to compute prediction. Generating a health warning that is based on prediction of atmospheric pollution, planning timely evacuation of people from vulnerable areas with respect to prediction of natural disasters, etc., are the use cases of sensor data stream where prediction is vital to protect people and assets. Thus, prediction accuracy is of paramount importance to take preventive steps and avert any untoward situation. Uncertainties of sensor data is a severe factor which hampers prediction accuracy. Belief Rule Based Expert System (BRBES), a knowledge-driven approach, is a widely employed prediction algorithm to deal with such uncertainties based on knowledge base and inference engine. In connection with handling uncertainties, it offers higher accuracy than other such knowledge-driven techniques, e.g., fuzzy logic and Bayesian probability theory. Contrarily, Deep Learning is a data-driven technique, which constitutes a part of Artificial Intelligence (AI). By applying analytics on huge amount of data, Deep Learning learns the hidden representation of data. Thus, Deep Learning can infer prediction by reasoning over available data, such as historical data and sensor data streams. Combined application of BRBES and Deep Learning can compute prediction with improved accuracy by addressing sensor data uncertainties while utilizing its discovered data pattern. Hence, this paper proposes a novel predictive model that is based on the integrated approach of BRBES and Deep Learning. The uniqueness of this model lies in the development of a mathematical model to combine Deep Learning with BRBES and capture the nonlinear dependencies among the relevant variables. We optimized BRBES further by applying parameter and structure optimization on it. Air pollution prediction has been taken as use case of our proposed combined approach. This model has been evaluated against two different datasets. One dataset contains synthetic images with a corresponding label of PM2.5 concentrations. The other one contains real images, PM2.5 concentrations, and numerical weather data of Shanghai, China. We also distinguished a hazy image between polluted air and fog through our proposed model. Our approach has outperformed only BRBES and only Deep Learning in terms of prediction accuracy.

Using Rough Sets to Improve Activity Recognition Based on Sensor Data.

Activity recognition plays a central role in many sensor-based applications, such as smart homes for instance. Given a stream of sensor data, the goal is to determine the activities that triggered the sensor data. This article shows how spatial information can be used to improve the process of recognizing activities in smart homes. The sensors that are used in smart homes are in most cases installed in fixed locations, which means that when a particular sensor is triggered, we know approximately where the activity takes place. However, since different sensors may be involved in different occurrences of the same type of activity, the set of sensors associated with a particular activity is not precisely defined. In this article, we use rough sets rather than standard sets to denote the sensors involved in an activity to model, which enables us to deal with this imprecision. Using publicly available data sets, we will demonstrate that rough sets can adequately capture useful information to assist with the activity recognition process. We will also show that rough sets lend themselves to creating Explainable Artificial Intelligence (XAI).

Fusing wearable and remote sensing data streams by fast incremental learning with swarm decision table for human activity recognition

Human activity recognition (HAR) by machine learning finds wide applications ranging from posture monitoring for healthcare and rehabilitation to suspicious or dangerous actions detection for security surveillance. Infrared cameras such as Microsoft Kinect and wearable sensors have been the two most adopted devices for collecting data for measuring the bodily movements. These two types of sensors generally are categorized as contactless sensing and contact sensing respectively. Due to hardware limitation, each of the two sensor types has their inherent limitations. One most common problem associating with contactless sensing like Kinect is the distance and indirect angle between the camera and the subject. For wearable sensor, it is limited in recognizing complex human activities. In this paper, a novel data fusion framework is proposed for combining data which are collected from both sensors with the aim of enhancing the HAR accuracy. Kinect is able to capture details of bodily movements from complex activities, but the accuracy is dependent heavily on the angle of view; wearable sensor is relatively primitive in gathering spatial data but reliable for detecting basic movements. Fusing the data from the two sensor types enables complimenting each other by their unique strengths. In particular, a new scheme using incremental learning with decision table coupled with swarm-based feature selection is proposed in our framework for achieving fast and accurate HAR by fusing data of two sensors. Our experiment results show that HAR accuracy could be improved from 23.51% to 68.35% in a case of almost 90 degrees slanted view of Kinect sensing while a wearing sensor is used at the same time. The swarm feature selection in general is shown to enhance the HAR performance compared to standard feature selection method. The experiment results reported here contribute to the possibilities of using hybridized sensors from the machine learning perspective.

SmartWheels: Detecting urban features for wheelchair users’ navigation

People with mobility impairments have heterogeneous needs and abilities while moving in an urban environment and hence they require personalized navigation instructions. Providing these instructions requires the knowledge of urban features like curb ramps, steps or other obstacles along the way. Since these urban features are not available from maps and change in time, crowdsourcing this information from end-users is a scalable and promising solution. However, it is inconvenient for wheelchair users to input data while on the move. Hence, an automatic crowdsourcing mechanism is needed.In this contribution we present SmartWheels, a solution to detect urban features by analyzing inertial sensors data produced by wheelchair movements. Activity recognition techniques are used to process the sensors data stream. SmartWheels is evaluated on data collected from 17 real wheelchair users navigating in a controlled environment (10 users) and in-the-wild (7 users). Experimental results show that SmartWheels is a viable solution to detect urban features, in particular by applying specific strategies based on the confidence assigned to predictions by the classifier.

Towards real-time recognition of activities in smart homes

Many supervised methods have been proposed to infer the particular activities of the inhabitants from a variety of sensors attached in the home. Current activity recognition systems either assume that the sensor stream has been presegmented or use a sliding window for activity segmentation. This makes real-time activity recognition task difficult due to the presence of temporal gaps between successive sensor activations. In this paper, we propose a method based on a set of hidden Markov models that can simultaneously solve the problem of activity segmentation and recognition on streaming sensor data without relying on any sliding window methods. We demonstrate our algorithm on sensor data obtained from two publicly available smart homes datasets.

Towards real-time recognition of activities in smart homes

Many supervised methods have been proposed to infer the particular activities of the inhabitants from a variety of sensors attached in the home. Current activity recognition systems either assume that the sensor stream has been presegmented or use a sliding window for activity segmentation. This makes real-time activity recognition task difficult due to the presence of temporal gaps between successive sensor activations. In this paper, we propose a method based on a set of hidden Markov models that can simultaneously solve the problem of activity segmentation and recognition on streaming sensor data without relying on any sliding window methods. We demonstrate our algorithm on sensor data obtained from two publicly available smart homes datasets.

Managing sensor data streams in a smart home application

Managing sensor data streams in a smart home application

Managing sensor data streams in a smart home application

A challenge in developing an ambient activity recognition system for use in elder care is finding a balance between the sophistication of the system and a cost structure that fits within the budgets of public and private sector healthcare organisations. Much activity recognition research in the context of elder care is based on dense networks of sensors and advanced methods, such as supervised machine learning algorithms. This paper presents the data processing aspects of an activity recognition system based on a simpler, knowledge-based unsupervised approach, designed for a sparse network of sensors. By structuring sensor data management as a streaming system, we provide a simple programming model for the application logic, which facilitates building a fault-tolerant system with the potential for distributed data management within the sensor network. The system, evaluated by a public sector healthcare organisation, constitutes an example of a system that is useful and has a sustainable cost structure.

A Survey on Behavioral Pattern Mining From Sensor Data in Internet of Things

The deployment of large-scale wireless sensor networks (WSNs) for the Internet of Things (IoT) applications is increasing day-by-day, especially with the emergence of smart city services. The sensor data streams generated from these applications are largely dynamic, heterogeneous, and often geographically distributed over large areas. For high-value use in business, industry and services, these data streams must be mined to extract insightful knowledge, such as about monitoring (e.g., discovering certain behaviors over a deployed area) or network diagnostics (e.g., predicting faulty sensor nodes). However, due to the inherent constraints of sensor networks and application requirements, traditional data mining techniques cannot be directly used to mine IoT data streams efficiently and accurately in real-time. In the last decade, a number of works have been reported in the literature proposing behavioral pattern mining algorithms for sensor networks. This paper presents the technical challenges that need to be considered for mining sensor data. It then provides a thorough review of the mining techniques proposed in the recent literature to mine behavioral patterns from sensor data in IoT, and their characteristics and differences are highlighted and compared. We also propose a behavioral pattern mining framework for IoT and discuss possible future research directions in this area.

Framework for Error Detection & its Localization in Sensor Data Stream for reliable big sensor data analytics using Apache Spark Streaming

Internet of Things (IoT) is one of the big sources of Big Sensor data in which collected sensor readings often polluted due to corruption and losses of the sensor readings. In Big Sensor data analytics, fast and efficient detection of error and its localization is a challenging research issue. Most of the existing solutions for the error detection and localization in sensor data use offline techniques for detecting errors. In this paper, we propose a novel framework for online error detection and its localization which uses an online scheme for detecting and localizing errors in sensor data using latest big data processing tools such as Apache Spark streaming. Performance evaluation of the proposed framework is done using two different datasets such as real-time air quality dataset of the Raipur city and Intel sensor dataset in terms of false positive.