Download as pdf or txt
Download as pdf or txt
You are on page 1of 7

Available online at www.sciencedirect.

com
Available online at www.sciencedirect.com
ScienceDirect
ScienceDirect
Available online at www.sciencedirect.com
Procedia Computer Science 00 (2018) 000–000
Procedia Computer Science 00 (2018) 000–000 www.elsevier.com/locate/procedia
ScienceDirect www.elsevier.com/locate/procedia

Procedia Computer Science 134 (2018) 114–120

The 15th International Conference on Mobile Systems and Pervasive Computing


The 15th International Conference on Mobile
(MobiSPC Systems and Pervasive Computing
2018)
(MobiSPC 2018)
Towards a Real-time Occupancy Detection Approach for Smart
Towards a Real-time Occupancy Detection Approach for Smart
Buildings
Buildings
H. Elkhoukhia,b,*, Y. NaitMaleka,c , A. Berouinea,c , M. Bakhouyaaa, D. Elouadghiribb,
H. Elkhoukhi , Y. NaitMalek , A.
a,b,* a,c
M. Berouine
Essaaidicc , M. Bakhouya , D. Elouadghiri ,
a,c

M. Essaaidi
International a
University of Rabat
Faculty of Computinga and Logistics,
International TICLab,
University ofSala
RabatEl Jadida, Morocco
{hamza.elkhoukhi, youssef.nait-malek
Faculty of Computing anass.berouine,
and Logistics, mohamed.bakhouya}@uir.ac.ma
TICLab, Sala El Jadida, Morocco
b
Université My youssef.nait-malek
{hamza.elkhoukhi, Ismail, Faculté desanass.berouine,
sciences, 11201mohamed.bakhouya}@uir.ac.ma
Zitoune, Meknès – Morocco
c
b ENSIAS,
Université My Ismail, Mohamed
Faculté V University,Rabat,
des sciences, Morocco
11201 Zitoune, Meknès – Morocco
c
ENSIAS, Mohamed essaaidi@ieee.org
V University,Rabat, Morocco
essaaidi@ieee.org

Abstract
Abstract
Context-awareness has been considered as a crucial fact for developing context-driven control approaches in which sensing, and
actuation tasks are has
Context-awareness performed accordingastoa the
been considered contextual
crucial fact forchanges.
developingThis could be done
context-driven by including
control approachestheinoccupants’ presence,
which sensing, and
actuationactions
number, tasks are
andperformed
behavioursaccording to the
in up-to-date contextual
context takingchanges. Thisthe
into account could be done
complex by including
interlinked the occupants’
elements, presence,
situations, processes,
number,
and theiractions and Many
dynamics. behaviours
recentinstudies
up-to-date
havecontext
shown taking into account
that occupants’ the complex
information is a interlinked
major leadingelements,
sourcesituations, processes,
of uncertainty when
and their dynamics.
developing Many recent
occupancy-driven studiesapproaches
control have shown forthat occupants’
energy information
efficient buildings.is Comprehensive
a major leading source of uncertainty
and real-time when
fine-grained
developing occupancy-driven
occupancy information control
has to be, approaches
therefore, integratedforinenergy
order toefficient
improve buildings. Comprehensive
the performance and real-time
of these control approaches. fine-grained
The work
occupancyininformation
presented hastowards
this paper is to be, therefore, integrated
the development of in order toplatform
a holistic improvethatthe performance of these
combines recent control
IoT and Bigapproaches. The work
data technologies for
real-time
presented occupancy
in this paperdetection.
is towards We thefocus mainly on
development of occupants’ presence
a holistic platform bycombines
that comparing staticIoT
recent and
anddynamic
Big datamachine learning
technologies for
real-time occupancy
techniques. detection.
Experiments have beenWeconducted
focus mainly on occupants’
and results presence
are presented by the
to assess comparing
usefulnessstatic and
of the dynamic
platform andmachine learning
the effectiveness
techniques.
of real-timeExperiments have been
machine learning conducted
strategies andstreams
for data results processing.
are presented to assess the usefulness of the platform and the effectiveness
of real-time machine learning strategies for data streams processing.
© 2018
© 2018 The
The Authors.
Authors. Published
Published by by Elsevier
Elsevier Ltd.
Ltd.
This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/)
© 2018
This The
is an Authors.
open accessPublished by Elsevier
article under Ltd.
the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/).
Peer-review under responsibility of the scientific committee of the 13th International Conference on Future Networks and
This is an open access article under the CC BY-NC-ND
Communications, FNC-2018 and the 15th International Conference license (http://creativecommons.org/licenses/by-nc-nd/3.0/).
on Mobile Systems and Pervasive Computing, MobiSPC 2018.
Keywords: IoT, Big Data, Context-awareness, smart buildings, machine learning;
Keywords: IoT, Big Data, Context-awareness, smart buildings, machine learning;

* Corresponding author. Tel.: +212-605-537-314.


E-mail address:author.
* Corresponding hamza.elkhoukhi@gmail.com
Tel.: +212-605-537-314.
E-mail address: hamza.elkhoukhi@gmail.com
1877-0509 © 2018 The Authors. Published by Elsevier Ltd.
This is an ©
1877-0509 open
2018access article Published
The Authors. under theby
CC BY-NC-ND
Elsevier Ltd. license (http://creativecommons.org/licenses/by-nc-nd/3.0/).
This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/).
1877-0509 © 2018 The Authors. Published by Elsevier Ltd.
This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/)
Peer-review under responsibility of the scientific committee of the 13th International Conference on Future Networks and
Communications, FNC-2018 and the 15th International Conference on Mobile Systems and Pervasive Computing, MobiSPC 2018.
10.1016/j.procs.2018.07.151
H. Elkhoukhi et al. / Procedia Computer Science 134 (2018) 114–120 115
2 Elkhoukhi et al. / Procedia Computer Science 00 (2018) 000–000

1. Introduction

Recent studies have shown that occupant information (e.g., number, presence, behavior, activities) is a major input
for control approaches in energy efficient buildings [1,2,3]. In fact, comprehensive fine-grained occupancy
information could be integrated to improve the performance of occupancy-driven control of HVAC, lighting, and
ventilation systems. Occupancy information can be classified into two main categories as follows. Spatial and
temporal properties provide information about occupants’ physical information, while the behavioral properties
provide information about occupants’ activities [4]. For instance, carbon dioxide devices sensors are commonly used
in buildings for demand-driven control of ventilation systems because they can infer an estimate of space occupancy.
Furthermore, the carbon dioxide (CO2) concentration is often available in monitored indoor environment in order to
assess the indoor air quality. Mainly, the measurement of the amount of CO2 in a space was considered as a main
indicator for occupancy prediction [5].
Many research works have stated that accurate occupancy detection can be achieved by fusing a multi-sensor data,
such as CO2, temperature, humidity, sound, motion sensors [1, 6, 7]. In fact, accurate detection of the actual occupancy
could help in developing context-driven control approaches in which sensing, and actuation tasks are performed
according to the contextual changes. Furthermore, with recent advances in wireless sensors networks, many industries
and researchers have confirmed the potential of IoT as an enabler to the development of intelligent and context-aware
services and applications [8,9,10]. These services could dynamically react to the environment changes and users’
preferences. For instance, in the context of smart buildings, occupancy information could be used for controlling
window opening and shading, HVAC [11,12], and Lighting [13] with the aim is to decrease energy consumption while
maintaining visual, air quality and thermal comfort of buildings’ occupants. However, handling dynamic and frequent
context changes is a difficult task without real-time event/data acquisition and processing platforms [14].
In this work, we investigate the integration of machine learning algorithms for occupancy detection, mainly the
occupants’ presence, which we have used for controlling ventilation systems. We have integrated SAMOA platform
(Scalable Advanced Massive Online Analysis), which was developed mainly for mining big data streams [15]. It’s
worth noting that, to the best of our knowledge, the work presented in [15] is considered as a pioneering work in the
domain of distributed and real-time machine learning on streaming data. The focus of our work is to shed more light
on the usefulness of integrating recent IoT, Big Data technologies, and real-time machine learning algorithms for
occupant’s presence detection.
The remainder of this paper is structured as follows. Section 2 presents recent work related to occupancy detection
algorithms. In Section 3, the architecture of the holistic platform is introduced. Section 4 shows the efficiency of
dynamic detection of occupants’ presence when using real-time machine learning compared to static approaches.
Conclusions and some perspectives are given in Section 5.

2. Related work

IoT technologies have been recently developed in order to connect a variety of building sub-systems along with
other environmental & contextual sensors, such as CO2, temperature, humidity and motion sensors. This will allow
monitoring and collecting useful building’s information that could be used for occupancy detection purposes [16, 17].
In fact, instantaneous indoor occupancy information in buildings becomes a major factor to improve occupants’
comfort and to greatly save and reduce energy consumption by developing optimal control approaches of HVAC and
lighting systems [18].
Recent studies show that combining these technologies with a various existing machine learning approaches (e.g.,
classification algorithms) can significantly improve occupancy detection accuracy by making data-driven prediction
through establishing an occupancy model from a sample inputs [19,20,21] (i.e. CO2, Temperature, Humidity, Light).
For example, authors in [7] used electricity consumption data sets gathered from smart electricity meters in five
households to train classification models based on support vector machines (SVM), K-nearest neighbor (KNN),
thresholding (THR) and hidden markov model (HMM) in order to detect the presence of occupants. The reported
accuracy was above 80% compared to the truth occupancy, which was obtained using a tablet computer installed in
the main entrance in order to record the true values of occupants’ presence.
116 H. Elkhoukhi et al. / Procedia Computer Science 134 (2018) 114–120
Elkhoukhi et al. / Procedia Computer Science 00 (2018) 000–000 3

In [22], authors applied different statistical classification models, mainly LDA (Linear Discriminant Analysis),
CART (Classification And Regression Trees) and RF (Random Forest) on the dataset that contains light, temperature,
humidity, CO2 values. They have shown that including information related to the time of the day and week status
(weekend, weekdays) increased the accuracy of occupancy detection by 32% with high accuracies (around 97%) by
using only two predictors. However, the ground truth was obtained by using a digital camera for supervised
classification model training. Authors in [23] applied seven different Artificial Neural Network algorithms on the
same dataset used in [22] to train the classifiers model. The result showed that the Limited Memory Quasi-Newton
algorithm has the highest accuracy rate with 99.061%. In the work presented in [24], authors proposed an approach
for accurate occupancy estimation based on a wireless sensor network, which combines environmental sensors with
uncertain contextual information. The study also used SVM and K-nearest neighbor to train the classifiers models.
In [25], a model of occupancy detection was evaluated in a single occupancy room using twelve ambient sensors
variables to train six machine learning classification algorithms: SVM, KNN, Artificial Neural network (ANN), Naõve
Bayesian (NB), Tree Augmented Naõve Bayes Network (TAN), and Decision Tree (DT). The authors found that CO2,
door status and light variables have important contributions to the final modeling results. The model accuracy ranged
from 92.2% to 98.2% according to the used classifier algorithm. A predictive model from electricity and water
consumption data was evaluated using Monte Carlo simulations [26]. Authors found that the Random Forest and
Decision Tree classifiers model under their boosting version had the best classification performance with an F-measure
of 83.37% and 82.79%, respectively. The truth occupancy was found using a door counter sensor. In [27], authors
proposed an approach based on data from thermal energy storage of electric water heaters to train classifiers model.
However, despite the importance of these algorithms in detecting occupancy from both static and stream data,
dynamic and real-time detection approaches are required to be applied in near real-time, for example in HVAC and
ventilation control. The integration of IoT and Big data technologies into a holistic platform together with machine
learning algorithms could enable new potentials in smart buildings for real-time occupancy detection. The work
presented in this paper investigates this research direction by integrating recent IoT and Big Data technologies for
occupancy detection, mainly the occupants’ presence. The architecture of the platform prototype is introduced in next
section.

3. The platform prototype architecture

This section presents the architecture of the platform prototype we have deployed for real-time detection of
occupants’ presence. As shown in Fig.1, it was designed as generic as possible to be applied in different smart
environments that require real-time monitoring and processing, such as in intelligent transportation [28, 29] and
health-care [14,30]. As illustrated in this figure, the platform is composed of three main layers, sensors and actuators
layer, processing layer, and services layer. The first layer is composed of different sensors and actuators that, when
deployed, could configured and remotely controlled through the Net. In fact, sensors are used to gather indoor data
such as CO2, humidity and temperature, which are submitted to the processing layer. This layer is composed of a pre-
processing unit to make sure that data to be submitted and stored is clean, and a real-time processing unit, in which,
IoT and Big Data tools are deployed to process data streams (e.g., Storm, Kaa). It includes so far, a machine learning
algorithm for occupants’ presence detection based on streaming data.
In this work, we placed a great emphasis on Vertical hoeffding tree algorithm for occupancy detection, which is
considered as an input for the services’ layer, such as HVAC system, lighting, and ventilation. We show how this
algorithm could be used with deployed IoT and Big data technologies for real-time occupancy detection. Regarding
data processing, as illustrated in Fig. 2, we have used the prequential evaluation, which uses each data instance first
to test and then train the model. In fact, it consists of three processing units, the source processor, the classifier and
the evaluator processor. The source processor reads data streams using apache Kafka (a publish/subscribe messaging
system) and sends them to the classifier. This later sends the results to the evaluator processor, which is designed to
evaluate the performance of the classifier by supporting the basic classification performance evaluators. It mainly
measures the accuracy of the classifier model since the beginning of the evaluation.
H. Elkhoukhi et al. / Procedia Computer Science 134 (2018) 114–120 117
4 Elkhoukhi et al. / Procedia Computer Science 00 (2018) 000–000

Fig. 1. Architecture of the platform prototype for real-time machine learning using IoT and Big Data technologies

Fig. 2. Prequential evaluation task topology

4. Experimental results

The main purpose of this study is to analyze and explore the occupancy information in buildings by integrating
recent IoT and Big Data technologies. We have used an occupancy detection dataset that includes indoor sensors data,
mainly light, temperature, humidity, and CO2 [22]. A platform prototype was developed and used to predict the status
of occupants’ presence (i.e., 0 for not occupied, 1 for occupied). We have compared both offline and online machine
learning techniques for occupants’ presence detection. In fact, offline machine learning or batch learning is a technique
that generates at once the best predictor by learning on the entire training data, while online machine learning updates
the best predictor at each time future data streams arrives. For offline machine learning, we have used LDA, which
showed better accuracy in the training and the test sets evaluation compared to RF and CART. Regarding online
machine learning, we have integrated the Vertical Hoeffding Tree, which showed its potential for analyzing data
streams [15].
Experiments have been first conducted using offline machine learning in order to figure out its potential in detecting
the occupants’ presence. As shown in Fig. 3, this algorithm has the ability to estimate the occupancy with high
accuracies. However, the LDA is used for batch analysis. In fact, it needs samples of cleaned sensors data together
with a model, which is trained either manually or automatically using additional sensors, in order to be used for
context’s detection and prediction. More precisely, offline machine learning is almost used by the community to study
the accuracy of machine learning algorithms (e.g., SVM, NN), and are not directly suitable for real-time processing
of data streams.
Online occupancy detection is, however, a potential technique that could be used to extract new insights from
sensors data streams. For instance, its main aim is to select the best suitable action (e.g., HVAC, ventilation control)
according to the contexts’ changes. Fig. 4 shows the usefulness of online machine learning for presence detection. We
choose to read data from dataset as data streams with a fixed frequency (e.g. 1, 10, 50 samples). These streams are
118 H. Elkhoukhi et al. / Procedia Computer Science 134 (2018) 114–120
Elkhoukhi et al. / Procedia Computer Science 00 (2018) 000–000 5

then transmitted to the platform for processing and classification. As shown in Fig.4, the algorithm detects accurately
the occupants’ presence. It’s worth noting that we have conducted similar experiments with several data streaming
frequencies, but we did not include them since they showed similar behavior. Furthermore, we have evaluated the
accuracy of the online machine learning technique. As shown in Fig. 5, a sudden decrease of the classification correct
for the first 10 instances was detected, and improved quickly to reach 95%, but during the first 50 samples.

Fig. 3. Offline detection of occupants’ presence using LDA

Fig. 4. Online detection of occupants’ presence using VHT.

Fig. 5. The classification corrects for online machine learning

These preliminary results show the efficiency of online occupancy detection when integrated with IoT and Big Data
technologies into a holistic platform. In fact, these experiments were first conducted to study and analyze the
effectiveness of the platform for data processing using real-time machine learning for occupancy presence detection.
Further experiments have been conducted in a real sitting scenario using the equipment that we have deployed in
our EEBLab. Experiments were taken place on 17th May starting from 9:25am to 4:00pm. The preliminary results,
depicted in Fig. 6, show that the platform is able to detect occupants’ presence with high accuracy (i.e., around 80%
H. Elkhoukhi et al. / Procedia Computer Science 134 (2018) 114–120 119
6 Elkhoukhi et al. / Procedia Computer Science 00 (2018) 000–000

in average). However, while these preliminary results are promising, more experiments are ongoing and new results
will be reported to show the effectiveness of the platform for occupancy detection (e.g., number), which is required
to deploy context-aware applications, such as for smart HVAC, lighting, and ventilation system control.

Fig. 6. The true and estimated values of occupants’ presence

5. Conclusions and perspectives

In this paper, we have presented a first prototype platform towards real-time machine learning integration with
recent IoT and Big data technologies. We compared both offline and online prediction for occupants’ presence
detection in smart buildings and preliminary results showed the potential and accuracy of real-time machine learning.
Ongoing work focuses on refining the platform prototype by using it in our EEBLab for occupants’ presence detection
including other parameters, such as the occupants’ number and their activities. More experiments will be also
conducted to show the usefulness of online machine learning in real sitting using the developed context-driven
approaches for ventilation, lighting, shading and HVAC control [31, 32].

Acknowledgements

This work is supported by CASANET (Context-Aware Sensor-Actuator Networks for Energy-Efficient Buildings)
project (2016-2019), which is funded by «le Ministere de l’Enseignement Superieur, de la Recherche Scientifique et
de la Formation des Cadres (MESRSFC)» and «le Centre National pour la Recherche Scientifique et Technique
(CNRST)», and partially supported by MIGRID project (grant 5-398,2017-2019), which is funded by USAID under
the PEER program.

References

[1] AN. Tuan, A. Marco, “Energy intelligent buildings based on user activity: a survey,” Energy Build 2013;56:244–57.
[2] A. Akbar, M. Nati, F. Carrez, and K. Moessner, “Contextual Occupancy Detection for Smart Office by Pattern Recognition of Electricity
Consumption Data,” http://iot-cosmos.eu/node/1566, 2014.
[3] H.-C. Shih, “A robust occupancy detection and tracking algorithm for the automatic monitoring and commissioning of a building,” Energy and
Buildings 77 (0) (2014) 270–280.
[4] F. Lachhab, M. Bakhouya, R. Ouladsine, M. Essaaidi, “Monitoring and Controlling Buildings Indoor Air Quality Using WSN-based
technologies,” in the 4th International Conference on Control, Decision and Information Technologies, April 5-7, 2017, Barcelona.
[5] D. Calì, P. Matthes, K. Huchtemann, R. Streblow, D. Müller, “CO2 based occupancy detection algorithm: experimental analysis and validation
for office and residential buildings,” Build. Environ. 86 (2015) 39–49.
[6] Z. Yang, N. Li, B. Becerik-Gerber, M. Orosz, “A multi-sensor based occupancy estimation model for supporting demand driven HVAC
operations,” in: Proceedings of the 2012 Symposium on Simulation for Architecture and Urban Design, Society for Computer Simulation
International, San Diego, CA, USA, 542 2012, pp. 49-56.
[7] W. Kleiminger, C. Beckel, T. Staake, and S. Santini, “Occupancy detection from electricity consumption data,” BuildSys. ACM, 2013.
[8] K. Akkaya, I. Guvenc, R. Aygun, N. Pala, A. Kadri, “IoT-based Occupancy Monitoring Techniques for Energy-Efficient Smart Buildings,”
IEEE Wireless Commun. and Networking Conf. Wksps., 2015, pp. 58–63
120 H. Elkhoukhi et al. / Procedia Computer Science 134 (2018) 114–120
Elkhoukhi et al. / Procedia Computer Science 00 (2018) 000–000 7

[9] F. Lachhab, M .Bakhouya, R .Ouladsine and M. Essaaidi. “Towards a context-aware platform for complex and stream event processing”.
In High Performance Computing & Simulation (HPCS), 2016 International Conference on (pp. 961-966), July 2016.
[10] F. Lachhab, M. Bakhouya, R. Ouladsine, M. Essaaidi, “Performance Evaluation of Linked Stream Data Processing Engines for Situational
Awareness Applications”, in Concurrency and Computation: Practice and Experience Journal, 30 (12), 2018, DOI: 10.1002/cpe.4380.
[11] S.W. Wang, J. Burnett, H. Chong, “Experimental validation of CO2- based occupancy detection for demand-controlled ventilation, Indoor
Built Environ,” 8 (6) (1999) 377–391.
[12] F. Oldewurtel, D. Sturzenegger, M. Morari, “Importance of occupancy information for building climate control,” Applied Energy, 101 (2013)
521-532
[13] C. de Bakker, M. Aries, H. Kort, A. Rosemann, “Occupancy-based lighting control in open-plan office spaces: A state-of-the-art review,
“Build Environ, 112 (2017) 308-321.
[14] Y. Nait Malek, A. Kharbouch, H. El Khoukhi, M. Bakhouya, V. Deflorio, D. Elouadghiri, S. Latre, C. Blondia., “On the use of IoT and Big
Data Technologies for Real-time Monitoring and Data Processing,” proc. of the 7th International Conference on Current and Future Trends of
Information and Communication Technologies in Healthcare (ICTH2017), Volume 113, pp. 429-434, Lund, Sweden.
[15] G. D. F. Morales and A. Bifet, “SAMOA: Scalable advanced massive online analysis,” J. Mach. Learn. Res., vol. 16, pp. 149–153, Jan. 2015.
[16] J. Pan, R. Jain, S. Paul, T. Vu, A. Saifullah, M. Sha, “An Internet of Things Framework for Smart Energy in Buildings: Designs, Prototype,
and Experiments”, IEEE Internet of Things Journal, Vol. 2, Issue 6, pp. 527-537, Dec. 2015.
[17] C. Marche, M. Nitti, V. Pilloni, “Energy efficiency in smart building: a comfort aware approach based on Social Internet of Things”, Global
Internet of Things Summit (GIoTS), 6-9 June 2017, Geneva, Switzerland.
[18] T. Labeodan, W. Zeiler, G. Boxem, Y. Zhao, “Occupancy measurement in commercial office buildings for demand-driven control applications:
A survey and detection system evaluation,” Energy and Buildings, 93 (2015) 303-314.
[19] A. Ebadat, G. Bottegal, D. Varagnolo, B. Wahlberg, K.H. Johansson, “Estimation of building occupancy levels through environmental signals
deconvolution,” in: Proceedings of the 5th ACM Workshop on Embedded Systems For Energy-Efficient Buildings, ACM, Rome, Italy, 2013,pp.
1-8.
[20] B. Dong, B. Andrews, K.P. Lam, M. Höynck, R. Zhang, Y.-S. Chiou, D. Benitez, “An information technology enabled sustainability test-bed
(ITEST) for occupancy detection through an environmental sensing network,” Energy and Buildings, 42 (7) (2010) 1038-1046.
[21] M. Jin, N. Bekiaris-Liberis, K. Weekly, CJ. Spanos, AM. Bayen, “Occupancy detection via environmental sensing,” IEEE Trans. Autom. Sci.
Eng, 2016, no. 99, pp. 1-13.s
[22] L.M. Candanedo, V. Feldheim, “Accurate occupancy detection of an office room from light, temperature, humidity and CO2 measurements
using statistical learning models,” Energy Build. 112 (2016) 28–39.
[23] K. Tutuncu, O. Cataltas, M. Koklu, “OCCUPANCY DETECTION THROUGH LIGHT, TEMPERATURE, HUMIDITY AND CO2
SENSORS USING ANN,” Proceedings of ISER 45th International Conference, Rabat, Morocco, 8th -9th December 2016, ISBN: 978-93-
86291-60-8
[24] A. Khan, J. Nicholson, S. mellor, T.Ploetz, “Occupancy monitoring using environmental & context sensors and a hierarchical analysis
framework. Embedded Systems for Energy-Efficient Buildings,” BuildSys. ACM (2014). p. 90–9. doi:10.1145/2674061.2674080.
[25] Z. Yang, N. Li, B. Becerik-Gerber, M. Orosz, “A systematic approach to occupancy modeling in ambient sensor-rich buildings,” Simulation,
90 (8) (2014) 960-977.
[26] T. Vafeiadis, et al., “Machine Learning Based Occupancy Detection Via The Use of Smart Meters,” ISCSIC. International symposium, pp. 6-
12, 2017.
[27] D. Chen, S. Kalra, D. Irwin, P. Shenoy, and J. Albrecht, “Preventing occupancy detection from smart meters,” IEEE Trans. Smart Grid, vol.
6, no. 5, pp. 2426–2434, Sep. 2015.
[28] A. Lahlou1, F. Ossart, E. Boudard, F. Roy, M. Bakhouya, “A Dynamic Programming Approach for Thermal Comfort Control in Electric
Vehicles, ” in the 2018 Vehicle Power and Propulsion Conference (VPPC), Chicago, Illinois, USA, 2018.
[29] I. Amezzane, Y. Fakhri, M. El Aroussi, M. Bakhouya, “Towards an Efficient Implementation of Human Activity Recognition for Mobile
Devices, ” in EAI Endorsed Trans. Context-aware Syst. & Appl. 4(13), 2018.
[30] A.Kharbouch, H. El Khoukhi, Y. Nait Malek, M. Bakhouya, V. De Florioa,, D. El Ouadghiri, S. Latre, C. Blondia , “Towards an IoT and Big
Data Analytics Platform for the Definition of Diabetes Telecare Services“, The 2nd International Conference Onsmart Applications And Data
Analysis For Smart Cities (SADASC’18), Casablanca, Morocco.
[31] F. Lachhab, M. Bakhouya, R. Ouladsine, M. Essaaidi, “Towards an Intelligent Approach for Ventilation Systems Control using IoT and Big
Data Technologies, ” in The 9th International Conference on Ambient Systems, Networks and Technologies (ANT 2018), Porto, Portugal,
Procedia Computer Science, Elsevier, Volume 130, pp. 926-931.
[32] F. Lachhab, Y. Nait Malek, M. Bakhouya, R. Ouladsine, M. Essaaidi, “A Context-Driven Approach using IoT and Big Data Technologies for
Controlling HVAC Systems, ” in the 5th International Conference on Control, Decision and Information Technologies, CoDIT’2018,
Thessaloniki, Greece, pp. 1-6, 2018.

You might also like