Now showing 1 - 10 of 11
  • Publication
    ANNOT : Automated electricity data annotation using wireless sensor networks
    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.
      2876Scopus© Citations 45
  • Publication
    A stable routing framework for tree-based routing structures in wsns
    Achieving industry standards for transmission of data is an ultimate goal for wireless sensor network (WSN) designers. Such standards are difficult to attain due to the challenging communications environment in which WSNs operate where best transmission routes may change rapidly. In such dynamic environments, it is not only advantageous, but commonly necessary to change the used route for better performance. However, with so many changing routes, it is possible to introduce instability, which can hinder the provision of services. Route instability can be particularly detrimental in a tree-based routing structure. In these scenarios, it is beneficial to maintain more stable routes in the tree to preserve high standards. The focus of this paper is the development of a route stability framework, whereby currently used metrics are adapted to promote routes that achieve greater stability in highly dynamic network conditions. The central concept introduces neighborhood heuristics (NHs), a method of combining a sensor's routing metric with those of its neighbors to highlight both the quality of the current route and the quality of the routing options available to the sensor should its current route become unavailable. The additional information afforded by the new combined metric allows sensors to choose good quality routes that can better maintain quality despite the degradation of an upstream link. The NHs framework is implemented with the routing protocol for low power and lossy networks routing protocol. Experiments are conducted both in simulation and on an open public testbed, which compare routing stability using the expected transmissions (ETX) metric and ETX under the NHs stability framework (ETX-NH), showing a marked increase in route stability for ETX-NH over using ETX.
      841Scopus© Citations 30
  • Publication
    Towards Application-Aware Networking: ML-Based End-to-End Application KPI/QoE Metrics Characterization in SDN
    Software Defined Networking (SDN) presents a unique networking paradigm that facilitates the development of network innovations. This paper aims to improve application awareness by incorporating Machine Learning (ML) techniques within an open source SDN architecture. The paper explores how end-to-end application Key Performance Indicator (KPI) metrics can be designed and utilized for the purpose of application awareness in networks. The main goal of this research is to characterize application KPI metrics using a suitable ML approach based on available network data. Resource allocation and network orchestration tasks can be automated based on the findings. A key facet of this research is introducing a novel feedback interface to the SDN's Northbound Interface that receives realtime performance feedback from applications. This paper aim to show how could we exploit the applications feedback to determine useful characteristics of an application's traffic. A mapping application with a defined KPI is used for experimentation. Linear multiple regression is used to derive a characteristic relationship between the application KPI and the network metrics.
      790Scopus© Citations 22
  • Publication
    Establishing Waiting Time Thresholds in Interactive Web Mapping Applications for Network QoE Management
    Customer expectations will continue to drive communication service developers to optimise their use of network resources based on user satisfaction. Thus, network platforms need to be remodelled from Quality of Service (QoS) centric to Quality of Experience (QoE) aware platforms. The perceived QoE for interactive web applications such as Google maps or Openstreetmaps is dominated by waiting time, i.e. the perceived time to render the page and map. Studies have explored waiting time estimation for Web QoE applications (e.g. email, downloads, web pages). Perceived waiting time for web mapping applications have been less comprehensively explored. The relationship between perceived waiting time and network QoS is a key QoE management factor to enable QoE aware networks. In this paper, we review the principle of network QoE management and the perception of waiting times. We present experimental design and methodology that facilitate the identification of waiting time thresholds for web applications, using web maps as a use case. We outline our results along with a statistical analysis and discussion interpreting the results and their applications. Finally, we discuss follow-up experiments and how they could be developed and applied in the network QoE management.
      236Scopus© Citations 6
  • Publication
    COPOLAN : non-invasive occupancy profiling for preliminary assessment of HVAC fixed timing strategies
    Nowadays, control of heating, cooling and ventilation equipment operation is mainly achieved via timers with fixed setback schedules, configured using experience and standard models of space occupancy. Applying generic timing strategies is however rarely optimal. Sensor-based systems offer a solution for dynamic control of equipment operation using real-time space occupancy input, but both deployment time and cost constraints hinder their integration if savings and return on investment are uncertain. This work introduces COPOLAN, a tool that correlates power consumption pat- terns and computers’ VLAN activity. Utilising computers’ VLAN activity auditing is key to obtain the power state of employees’ computer equipment over time, a prime indicator of employees’ presence within a building. At low cost and non-invasively, COPOLAN uncovers misalignment and pro- duces ground for (1) determining opportunities of improv- ing HVAC timing strategies and (2) helping decision making prior to integrating new equipment such as sensor-based systems. COPOLAN has been experimented on within a University department, where misalignment between power consumption and space occupancy patterns have highlighted 10 % energy saving opportunities.
      839Scopus© Citations 12
  • Publication
    Spreading the load in a tree type routing structure
    Many routing protocols have advanced the Wire- less Sensor Network (WSN) paradigm with each new method offering unique ways to maximise Quality of Service (QoS) while minimising energy cost to the network. Tree routing is a well examined method with a proven record in offering a high level of service with trees constructed to fulfil a particular routing objective defined by a chosen metric. The tree structure can be maintained with low overall network overhead but exhibits a weakness with regard to load balancing. Without effective load balancing particular nodes in the network may be subject to excessive network load, leading to uneven energy consumption in the network. This in turn can lead to an unwanted scenario of premature node failure. Neighbourhood metrics is presented in this paper as a means to preserve network objectives while achieving improved load distribution in a tree type routing structure. Neighbourhood metrics offer a framework for expand- ing on currently used metrics to include information on the quality of a nodes neighbourhood in addition to the current forwarding route. Neighbourhood metrics is compared to the current state of the art in the form of the Routing Protocol for Low Power and Lossy Networks (RPL) implementation using the direct Expected Transmissions (ETX) metric. Neighbourhood metrics exhibits improved load distribution in a number of open public testbeds.
      478Scopus© Citations 9
  • Publication
    Beyond First Impressions: Estimating Quality of Experience for Interactive Web Applications
    The number of web applications for both personal and business use will continue to increase. The popularity of web applications has grown, increasing the need to estimate Quality of Experience for web applications (Web QoE). Web QoE helps providers to understand how their end-users perceive quality and point towards areas to improve. Waiting time has been proven to have a significant influence on user satisfaction. Most studies in the field of Web QoE have focused on modelling Web QoE for the user’s first interaction with the application, e.g., the waiting time for the first page load to complete. This does not include a user’s subsequent interactions with the application. Users keep interacting with the application beyond the first page load resulting in an experience that consists of a series of waiting times. In this study, we have chosen web maps as a use case to investigate how to measure waiting time for a user’s interactions across a web browsing session, and to measure the correlation between waiting time and user-reported perceived quality. We provide a short survey of existing Web QoE estimation metrics and models. We then propose two new measures: interactive Load Time (iLT) and Total Completed interactive Load (TCiL) to establish the waiting time associated with a web application user’s interactions. A subjective study confirms a logarithmic relationship for interactive web application sessions between iLT and perceived quality. We compare the correlation between QoE for iLT and the state of the art, non-interactive equivalent, Page Load Time (PLT)/Waiting Time. We demonstrate how the iLT/QoE fitting curve deviates from PLT/QoE. The number of clicks in completing tasks and TCiL are explored to explain the connections between user’s interactions behaviour and the perceived quality.
      329Scopus© Citations 18
  • Publication
    How Crisp is the Crease? A Subjective Study on Web Browsing Perception of Above-The-Fold
    Quality of Experience (QoE) for various types of websites has gained significant attention in recent years. In order to design and evaluate websites, a metric that can estimate a user’s experienced quality robustly for diverse content is necessary. SpeedIndex (SI) has been widely adopted to estimate perceived web page loading progress. It measures the speed of rendering pixels for the webpage that is visible in the browser window. This is termed Above-The-Fold (ATF). The influence of animated content on the perception of ATF has been less comprehensively explored. In this paper, we present an experimental design and methodology to measure ATF perceptionfor websites with and without animated elements for various pagecontent categories. We found that pages with animated elements caused people to have more varied perceptions of ATF under different network conditions. Animated content also impacts the page load estimation accuracy of SI for websites. We discuss how the difference in the perception of ATF will impact the QoE management of web applications. We explain the necessity of revisiting the visual assessment of ATF to include the animated contents and improve the robustness of metrics like SI.
      306Scopus© Citations 6
  • Publication
    A Framework to Implement IoT Network Performance Modelling Techniques for Network Solution Selection
    No single network solution for Internet of Things (IoT) networks can provide the required level of Quality of Service (QoS) for all applications in all environments. This leads to an increasing number of solutions created to fit particular scenarios. Given the increasing number and complexity of solutions available, it becomes difficult for an application developer to choose the solution which is best suited for an application. This article introduces a framework which autonomously chooses the best solution for the application given the current deployed environment. The framework utilises a performance model to predict the expected performance of a particular solution in a given environment. The framework can then choose an apt solution for the application from a set of available solutions. This article presents the framework with a set of models built using data collected from simulation. The modelling technique can determine with up to 85% accuracy the solution which performs the best for a particular performance metric given a set of solutions. The article highlights the fractured and disjointed practice currently in place for examining and comparing communication solutions and aims to open a discussion on harmonising testing procedures so that different solutions can be directly compared and offers a framework to achieve this within IoT networks.
      383Scopus© Citations 10
  • Publication
    Improving Data Quality of Low-cost IoT Sensors in Environmental Monitoring Networks Using Data Fusion and Machine Learning Approach
    Environmental monitoring has become an active research area due to the current rise in the global climate change crises. Current environmental monitoring solutions, however, are characterized by high cost of acquisition and complexity of installation; often requiring extensive resources, infrastructure and expertise. It is infeasible to achieve with these solutions, high density in-situ networks such as are required to build refined scale models to facilitate robust monitoring, thus, leaving large gaps within the collected dataset. Low-Cost Sensors (LCS) can offer high-resolution spatiotemporal measurements which could be used to supplement existing dataset from current environmental monitoring solutions. LCS however, require frequent calibration in order to provide accurate and reliable data as they are often affected by environmental conditions when deployed on the field. Calibrating LCS can help to improve their data quality and ensure they are collecting accurate data. Achieving effective calibration, however, requires identifying factors that affect sensor’s data quality for a given measurement. This study evaluates the performance of three Feature Selection (FS) algorithms including Forward Feature Selection (FFS), Backward Elimination (BE) and Exhaustive Feature Selection (EFS) in identifying factors that affect data quality of low-cost IoT sensors in environmental monitoring networks. Applying the concept of data fusion, sensors data were merged with environmental factors and integrated into a single calibration equation to calibrate cairclipO3/NO2 and cairclipNO2 sensors using Linear Regression (LR) and Artificial Neural Networks (ANN). The study showed the effectiveness of calibration in improving low-cost IoT sensor data quality and also demonstrated the convenience of feature selection and the ability of data fusion to provide more consistent, accurate and reliable information for calibration models. The analysis showed that the cairclipO3/NO2 sensor provided measurements that have good correlation with reference measurements whereas the cairclipNO2 sensor showed no reasonable correlation with the reference data. Calibrating the cairclipO3/NO2 yielded good improvement in its measurement outputs when compared to reference measurements (R2=0.83). However, calibrating the cairclipNO2 sensor data yielded no significant improvement in its data quality.
      334Scopus© Citations 68