A Scalable Room Occupancy Prediction with Transferable Time Series Decomposition of CO2 Sensor Data

@article{Ang2018ASR,
  title={A Scalable Room Occupancy Prediction with Transferable Time Series Decomposition of CO2 Sensor Data},
  author={Irvan Bastian Arief Ang and Margaret Hamilton and Flora D. Salim},
  journal={ACM Transactions on Sensor Networks (TOSN)},
  year={2018},
  volume={14},
  pages={1 - 28}
}
Human occupancy counting is crucial for both space utilisation and building energy optimisation. [...] Key Method DA-HOC++ is able to predict the number of occupants with minimal training data: as little as 1 day’s data. DA-HOC++ accurately predicts indoor human occupancy for five different rooms across different countries using a model trained from a small room and adapted to other rooms. We evaluate DA-HOC++ with two baseline methods: a support vector regression technique and an SD-HOC model. The results…Expand
ODToolkit: A Toolkit for Building Occupancy Detection
TLDR
This paper extends the design and implementation of an open-source toolkit for occupancy detection by implementing novel domain-adaptive occupancy detection algorithms and comparing them with the benchmark supervised learning algorithms on multiple data sets. Expand
A framework to identify key occupancy indicators for optimizing building operation using WiFi connection count data
TLDR
A framework for extracting occupancy indicators from WiFi traffic data utilizing several machine learning algorithms and statistical analysis methods to predict patterns of building occupancy as well as to identify peak occupancy time and earliest/latest arrival and departure times. Expand
Estimating Number and Dwell Time of Visitors from CO2 Concentration using Partial Modeling with Variable Time Window
TLDR
This paper proposes partial modeling with a variable time window that can make a partial estimation model that automatically corresponds to differences in the change speed between two variables: visitors and CO2 concentration. Expand
Building occupancy forecasting: A systematical and critical review
  • Yuan Jin, Da Yan, A. Chong, Bing Dong, Jingjing An
  • Computer Science
  • 2021
TLDR
This article reviews the literature regarding future building occupancy predictions (forecasting) and focuses on the research purpose, physical routine, and complete methodology of occupancy forecasting. Expand
A review of building occupancy measurement systems
TLDR
Many occupancy measurement systems based on different sensors, especially images/videos based methods, and future trends are presented, including the fifth generation mobile network (5 G), cloud computing platform and artificial intelligence especially deep learning technologies. Expand
Day-ahead prediction of plug-in loads using a long short-term memory neural network
TLDR
The results showed that the developed long-short-term memory (LSTM) model is applicable to the tested buildings without the need for occupant-wise or building-wise calibration, hence, addressing an important gap in the existing literature. Expand
Inferring Transportation Mode and Human Activity from Mobile Sensing in Daily Life
TLDR
This paper focuses on simultaneous inference of transportation modes and human activities in daily life via modelling and inference from multivariate time series data, which are streamed from off-the-shelf mobile sensors in real-world dynamic environments. Expand
Typical weekly occupancy profiles in non-residential buildings based on mobile positioning data
  • Xuyuan Kang, Da Yan, Jingjing An, Yuan Jin, Hongsan Sun
  • Computer Science
  • 2021
TLDR
This research utilizes mobile positioning data from social media platforms to extract typical weekly occupancy profiles of non-residential buildings by cluster analysis and suggests significant difference on energy load profiles and load distributions with the real occupancy profiles. Expand
Genetic Programming with Transfer Learning for Urban Traffic Modelling and Prediction
TLDR
This work forms this challenge as a symbolic regression problem and solves it using Genetic Programming, which enhances with a lag operator and transfer learning, and utilises knowledge collected from other road segments in order to predict vehicle flow through a junction where traffic data are not available. Expand
Transfer Learning in Smart Environments
TLDR
The potential of communication and transfer learning between smart environments in order to empower a decentralized and peer-to-peer ecosystem for seamless and automatic transfer of services and machine learning models is investigated. Expand
...
1
2
3
...

References

SHOWING 1-10 OF 44 REFERENCES
DA-HOC: semi-supervised domain adaptation for room occupancy prediction using CO2 sensor data
TLDR
A semi-supervised domain adaptation method for carbon dioxide - Human Occupancy Counter (DA-HOC), a robust way to estimate the number of people within in one room by using data from a carbon dioxide sensor. Expand
RUP: Large Room Utilisation Prediction with carbon dioxide sensor
TLDR
This work presents large Room Utilisation Prediction with carbon dioxide sensor (RUP), a novel way to estimate the number of people within a closed space from a single carbon dioxide Sensor, which de-noises and pre-processes the carbon dioxide and indoor human occupancy data. Expand
PerCCS: person-count from carbon dioxide using sparse non-negative matrix factorization
TLDR
The PerCCS algorithm is presented that explores the possibility of estimating person count from CO2 sensors already integrated in everyday room air-conditioning infrastructure by using task-driven Sparse Non-negative Matrix Factorization and outperformed the comparative methods in predicting occupancy count. Expand
A multi-sensor based occupancy estimation model for supporting demand driven HVAC operations
TLDR
The results indicate the ability of the proposed system to monitor the occupancy information of multi-occupancy spaces in real time, supporting demand driven HVAC operations. Expand
Human occupancy recognition with multivariate ambient sensors
TLDR
This paper aims to identify which ambient sensor is the most dominant in recognising human presence, and explains the methodology for integrating large amounts of sensor data, and discusses the experiments and findings in relation to the binary occupancy of a single person office. Expand
Real-time Building Occupancy Sensing for Supporting Demand Driven HVAC Operations
Accurate knowledge of localised and real-time occupancy numbers can have compelling control applications for Heating, Ventilation and Air-conditioning (HVAC) systems. However, a precise and reliableExpand
Accurate occupancy detection of an office room from light, temperature, humidity and CO2 measurements using statistical learning models
Abstract The accuracy of the prediction of occupancy in an office room using data from light, temperature, humidity and CO2 sensors has been evaluated with different statistical classification modelsExpand
SD-HOC: Seasonal Decomposition Algorithm for Mining Lagged Time Series
TLDR
Seasonal Decomposition for Human Occupancy Counting (SD-HOC), a customised feature transformation decomposition, novel way to estimate the number of people within a closed space using only a single carbon dioxide sensor is presented. Expand
Occupancy monitoring using environmental & context sensors and a hierarchical analysis framework
TLDR
This work presents an approach for accurate occupancy estimation using a wireless sensor network that only collects non-sensitive data and a novel, hierarchical analysis method, and shows how the system is used for analysing historical data and identify effective room misuse and thus a potential for energy saving. Expand
Performance comparison of occupancy count estimation and prediction with common versus dedicated sensors for building model predictive control
Model predictive control is a promising approach to optimize the operation of building systems and provide demand-response functionalities without compromising indoor comfort. The performance ofExpand
...
1
2
3
4
5
...