IoT sensors are becoming increasingly important supplement to traditional monitoring systems, particularly for in-situ based monitoring. Data collected using IoT sensors are often plagued with missing values occurring as a result of sensor faults, network failures, drifts and other operational issues. Missing data can have substantial impact on in-field sensor calibration methods. The goal of this research is to achieve effective calibration of sensors in the context of such missing data. To this end, two objectives are presented in this paper. 1) Identify and examine effective imputation strategy for missing data in IoT sensors. 2) Determine sensor calibration performance using calibration techniques on data set with imputed values. Specifically, this paper examines the performance of Variational Autoencoder (VAE), Neural Network with Random Weights (NNRW), Multiple Imputation by Chain Equations (MICE), Random Forest-based Imputation (missForest) and K-Nearest Neighbour (KNN) for imputation of missing values on IoT sensors. Furthermore, the performance of sensor calibration via different supervised algorithms trained on the imputed dataset were evaluated. The analysis showed VAE technique to outperform the other methods in imputing the missing values at different proportions of missingness on two real-world datasets. Experimental results also showed improved calibration performance with imputed dataset.

In modern energy aware buildings, lighting control systems are put in place so to maximise the energy-efficiency of the lighting system without effecting the comfort of the occupant. In many cases this involves utilising a set of presence sensors, with actuators, to determine when to turn on/off or dim lighting, when it is deemed necessary. Such systems are installed using standard tuning values statically fixed by the system installer. This can cause inefficiencies and energy wastage as the control system is never optimised to its surrounding environment. In this paper, we investigate a Wireless Sensor Network (WSN) as a viable tool that can help in analysing and evaluating the energy-efficiency of an existing lighting control system in a low-cost and portable solution. We introduce LightWiSe (LIGHTting evaluation through WIreless SEnsors), a wireless tool which aims to evaluate lighting control systems in existing office buildings. LightWiSe determines points in the control system that exhibit energy wastage and to highlight areas that can be optimised to gain a greater efficiency in the system. It will also evaluate the effective energy saving to be obtained by replacing the control system with a more judicious energy saving solution. During a test performed in an office space, with a number of different lighting control systems we could highlight a number of areas to reduce waste and save energy. Our findings show that each system tested can be optimised to achieve greater efficiency. LightWiSe can highlight savings in the region of 50% to 70% that are achievable through optimising the current control system or installing an alternative.

Reducing energy consumption within buildings has been an active area of research in the past decade; more recently, there has been an increased influx of activity, motivated by a variety of issues including legislative, tax-related, as well as an increased awareness of energy-related issues. Energy usage both in commercial and residential buildings represents a significant portion of overall energy consumption; however, much of this may be categorized as waste, that is, energy usage that does not fulfil a definite purpose. In the past decade, the viability of Wireless Sensor Network (WSN) technologies has been demonstrated, leading to increased possibilities for novel services for building energy management. This development has resulted in numerous approaches being proposed for harnessing WSNs for energy management and conservation. This article surveys the state-of-the-art in building energy management systems. A generic architecture is proposed after which a detailed taxonomy of existing documented systems is presented. Gaps in the literature are highlighted and directions for future research identified.

Existing research recognizes the critical role of quality data in the current big-data and Internet of Things (IoT) era. Quality data has a direct impact on model results and hence business decisions. The growth in the number of IoT-connected devices makes it hard to access data quality using traditional assessments methods. This is exacerbated by the need to share data across different IoT domains as it increases the heterogeneity of the data. Data-shared IoT defines a new perspective of IoT applications which benefit from sharing data among different domains of IoT to create new use-case applications. For example, sharing data between smart transport and smart industry can lead to other use-case applications such as intelligent logistics management and warehouse management. The benefits of such applications, however, can only be achieved if the shared data is of acceptable quality. There are three main practices in data quality (DQ) determination approaches that are restricting their effective use in data-shared platforms: (1) most DQ techniques validate test data against a known quantity considered to be a reference; a gold reference. (2) narrow sets of static metrics are used to describe the quality. Each consumer uses these metrics in similar ways. (3) data quality is evaluated in isolated stages throughout the processing pipeline. Data-shared IoT presents unique challenges; (1) each application and use-case in shared IoT has a unique description of data quality and requires a different set of metrics. This leads to an extensive list of DQ dimensions which are difficult to implement in real-world applications. (2) most data in IoT scenarios does not have a gold reference. (3) factors endangering DQ in shared IoT exist throughout the entire big-data model from data collection to data visualization, and data use. This paper aims to describe data-shared IoT and shared data pools while highlighting the importance of sharing quality data across various domains. The article examines how we can use trust as a measure of quality in data-shared IoT. We conclude that researchers can combine such trust-based techniques with blockchain for secure end-to-end data quality assessment.

Recent advances in low-power wireless networking have enabled remote and nonintrusive access to households' electric meter readings, allowing direct real-time feedback on electricity consumption to home owners and energy providers. Fine-grained electricity billing based on appliance power load monitoring has been investigated for more than two decades, but has not yet witnessed wide commercial acceptance. In this paper, we argue that the required human supervision for profiling and calibrating appliance load monitoring systems is a key reason preventing large-scale commercial roll-outs. We propose ANNOT, a system to automate electricity data annotation leveraging cheap wireless sensor nodes. Characteristic sensory stimuli captured by sensor nodes placed next to appliances are translated into appliance operating state and correlated to the electricity data, autonomously generating the annotation of electricity data with appliance activity. The system is able to facilitate the acquisition of appliance signatures, training data and validate the monitoring output. We validate the concept by integrating the automated annotation system to the RECAP appliance load monitoring system.

Continued development of communication technologies has led to widespread Internet-of-Things (IoT) integration into various domains, including health, manufacturing, automotive, and precision agriculture. This has further led to the increased sharing of data among such domains to foster innovation. Most of these IoT deployments, however, are based on heterogeneous, pervasive sensors, which can lead to quality issues in the recorded data. This can lead to sharing of inaccurate or inconsistent data. There is a significant need to assess the quality of the collected data, should it be shared with multiple application domains, as inconsistencies in the data could have financial or health ramifications. This article builds on the recent research on trust metrics and presents a framework to integrate such metrics into the IoT data cycle for real-time data quality assessment. Critically, this article adopts a mechanism to facilitate end-user parameterization of a trust metric tailoring its use in the framework. Trust is a well-established metric that has been used to determine the validity of a piece or source of data in crowd-sourced or other unreliable data collection techniques such as that in IoT. The article further discusses how the trust-based framework eliminates the requirement for a gold standard and provides visibility into data quality assessment throughout the big data model. To qualify the use of trust as a measure of quality, an experiment is conducted using data collected from an IoT deployment of sensors to measure air quality in which low-cost sensors were colocated with a gold standard reference sensor. The calculated trust metric is compared with two well-understood metrics for data quality, root mean square error (RMSE), and mean absolute error (MAE). A strong correlation between the trust metric and the comparison metrics shows that trust may be used as an indicative quality metric for data quality. The metric incorporates the additional benefit of its ability for use in low context scenarios, as opposed to RMSE and MAE, which require a reference for comparison.