Innovative Low Power Multiradio Sensing and Control Device for

Non-Intrusive Occupancy Monitoring

Marco De Donno and Brendan O’Flynn

Tyndall National Institute, University College Cork, Cork, Ireland

Keywords: Smart Sensing, Low Power Consumption, Building Energy Management (BEM), Multiradio Systems,

Multisensing Systems, Internet of Things.

Abstract: New tools and methodologies to reduce the gap between predicted and actual energy performances at the level

of buildings and blocks of buildings are in continuous development in academic and industry organizations.

The development of Wireless Sensor Networking (WSN) technology plays a core role in this field since their

development enables the monitoring and control of application within the building environment. In this paper

the development of a low power consumption multiradio and multisensing system to monitor building

conditions and enable the interaction of occupants with devices through embedded actuators is described. The

device (named NOD) incorporates a 32-bit ARM-Cortex microcontroller, a variety of sensors to monitor the

ambient conditions – luminance, temperature, humidity, air quality - and multiple radio interfaces -

WiFi/Bluetooth LE/868MHz. The NOD is intended to be used as a desktop device with a dedicated user

interface. A description of the system and its features and functionalities is provided.

1 INTRODUCTION

With the increasing demand for more energy efficient

buildings (Laustsen, 2008), the construction and

energy services industries are faced with the challenge

of ensuring that the energy performance and savings

predicted during energy efficiency measures

definition is actually achieved during operation. There

is, however, significant evidence to suggest that

buildings underperform. A, so called, “performance

gap” which is attributed to a variety of causal factors

related to both predicted and in-use performance,

implying that predictions tend to be unrealistically low

whilst actual energy performance is usually

unnecessarily high (MOEEBIUS, 2016).

It is also important to monitor the air quality in

offices environments and take actions to improve it.

Poor indoor air quality can reduce the performance of

office work by 6-9% and these negative effects on

performance are accompanied by general symptoms

such as headache and concentration (Wyon, 2004).

Wireless Sensor Network (WSN) systems are

becoming an increasingly popular technology that is

used today in a myriad of applications such as

Building Energy Management (BEM), Smart Homes,

Home Area Networking (HAN), smart cities and

environmental monitoring applications.

New architectures are required to offer improved

inter-operability, to improve the reliability of data

communications and to address the spread spectrum

requirements associated with next generation sensor

systems through the development of smart radio

systems. Currently, available platforms exist that have

multiple radios but these tend to operate in a single

Industrial, Scientific and Medical (ISM) band

(typically 2.4GHz) – and not in combination with the

868MHz ISM Band, which is ideal for the built

environment due to its long range of transmission and

low data rate communication properties (O’Flynn,

2016). The use of a wireless sensing system reduces

the total cost of the network installation compared to

traditional wired sensing systems which ranges from

40 to $2000 per linear foot of wiring (Feng Zhao,

Leonidas and Guibas, 2004).

The WSN group in TYNDALL is participating in

“MOEEBIUS – Modelling Optimization of Energy

Efficiency in Buildings for Urban Sustainability”, a

project funded by the European Union’s Horizon 2020

research and innovation programme. MOEEBIUS

introduces a Holistic Energy Performance

Optimization Framework that enhances current

De Donno, M. and O’Flynn, B.

Innovative Low Power Multiradio Sensing and Control Device for Non-Intrusive Occupancy Monitoring.

DOI: 10.5220/0006692601730181

In Proceedings of the 7th International Conference on Smart Cities and Green ICT Systems (SMARTGREENS 2018), pages 173-181

ISBN: 978-989-758-292-9

173

(passive and active building elements) modelling

approaches and delivers innovative simulation tools

which deeply grasp and describe real-life building

operation complexities in accurate simulation

predictions that significantly reduce the

“performance gap” and enhance multi-fold,

continuous optimization of building energy

performance as a means to further mitigate and

reduce the identified “performance gap” in real-time

or through retrofitting of appropriate sensing

technology (Launsten, 2008).

To enable this, the MOEEBIUS NOD has been

developed as a hardware device to monitor building

context conditions and further enable the interaction

of occupants with devices through embedded

actuators.

Section 1 of this paper introduces the subject

matter, the application space and the objective of this

work within the project. Section 2 reports the state of

the art in current wireless sensing system technologies

for indoor air quality monitoring and multiradio

platforms. Section 3 describes the “NOD” developed

within the project. Section 4 describes the

configuration and the operational mode of the device.

Section 5 investigates power consumption

characteristics of the platform. Section 6 concludes the

work and outlines some directions for future work and

research in this area.

2 PREVIOUS WORKS

A variety of wireless sensors systems are available and

are continuously under development using different

communication standards, proprietary and non-

proprietary, such as ZigBee, IEEE 802.15x and

Bluetooth. The use of one protocol instead of the other

carries some advantages but also some disadvantages.

Indoor range of transmission above the GHz

frequency is quite limited especially for indoor

applications with dense obstacles that will cause

interference (Harwood, 2009). The Wi-Fi technology

overcomes those issues by using higher transmission

power (up to 100 times higher than ZigBee/802.15.4),

which is of course not suitable for battery powered

systems in low power WSN systems (O’Flynn, 2016).

A number of the currently available systems for

indoor air quality are reported here as well as a

selection of multiradio platform currently in use in

such systems.

The “M-POD” (Jiang, 2011) is the portable IAQ

sensing device wireless embedded sensing,

computation, and communication device based on the

Arduino BT (Figure 1.a). It is capable of sensing the

concentrations of a number of air pollutants and either

storing these data or transmitting them to nearby

smartphones via its Bluetooth interface. The M-pod

has a humidity sensor, light sensor, two temperature

sensors – one upstream to measure ambient air

temperature and the other downstream to measure the

temperature near the sensors, a CO2 sensor, and low-

cost metal oxide gas sensors. The battery lifespan is

approximately 5.5 hours if an M-pod is continuously

on and greater than 24 hours when in low-power

mode.

The “AirSense” (Fang, 2011) is a IAQ sensing

platform developed to use with the Arduino Uno

Ethernet board. The platform has three onboard

sensors including temperature, humidity and VOCs

sensors (Figure 1.b) and is enclosed in a 3D printed

case. The sampled sensor data are transmitted to the

cloud server via the onboard Ethernet port. Besides the

onboard sensors, AirSense also incorporates a

standalone consumer-grade Particulate Matter (PM)

sensor DC1700 from Dylos (AQMD, 2015) to

measure the concentration of indoor PM 2.5.

The “Authentic” board (O’Flynn, 2016) (Figure

1.c) is a Multiradio Sensing Systems for Home Area

Networking and Building Management based on a 32-

bit microcontroller, incorporating temperature and

light sensors and three radio modules

(ZigBee/6LoWPAN, Bluetooth LE, 868MHz). The

configurability of the system can increase the range

between single sensor points and can enable the

implementation of adaptive networking architectures

of different configurations (O’Flynn, 2016). The

system needs an external gateway in order to send the

data to internet.

The BtNode V3 (BTNODE, 2007) (Figure 1.d) is

a platform with two radios onboard. It incorporates a

Chipcon CC1000 low power radio (433-915 MHz)

and also has an additional ZV4002 Bluetooth radio

(2.4 GHz).

The Wasp Mote (LIBELIUM, 2017) is a platform

tha has separate 868 and 900MHz radio modular plug-

in boards, but only a single radio module can be

operated at a time and true multi-radio operation is not

feasible.

The NOD device described in this publication can

be also used as repeater increasing the range of the

network. Moreover it offers the possibility to interact

with the user through buttons and display or an

Android Application for maintenance or data

visualization.

SMARTGREENS 2018 - 7th International Conference on Smart Cities and Green ICT Systems

174

Figure 1: Multiradio systems: a) M-POD b) AirSense c)

Authentic board d) BtNode V3 e)Wasp Mote.

3 SYSTEM IMPLEMENTATION

The Nod is a desktop device which senses human

presence, ambient temperature and humidity, air

quality and light conditions in the user’s vicinity and

communicates these parameters wirelessly to a

gateway. The Nod device also provides an intuitive

and easy to use interface with feedback that allows a

user to input and send commands to the Building

management layer through a gateway to control the

ambient environment in order to maximize user

comfort. The latency (delay) is in the order of

milliseconds and is not apparent to the user.

A specification process was undertaken with

consortium partners from industry and academia to

identify the core requirements associated with a

wireless system for deployment in offices. Technical

features which were assessed and considered

included: functionality requirements as regards

actuation and control, quality of service, latency,

number and types of sensors/meters and interfaces,

programming methods (wireless/non wireless), power

supplies/energy harvesting compatibility, radio

frequency band, standards/non standards

communications and data transmission range.

Figure 2: MOEEBIUS NOD Platform.

The platform described in the following sections

of this paper is a novel low power consumption

multiradio and multisensing system based on a 32-bit

ARM-Cortex-M4 microcontroller, incorporating

multiple radio interfaces - Bluetooth

LE/868MHz/WiFi - to provide increased connectivity

in deployment, and potentially reduce the interference

impact on the network as the system can hop from

ISM band to ISM band automatically. It also includes

sensors for detecting human presence, ambient

Innovative Low Power Multiradio Sensing and Control Device for Non-Intrusive Occupancy Monitoring

175

temperature, humidity, air quality and light conditions

in the user’s vicinity (~5m) in a 360° field of view.

The network of NOD devices adopts a

standardized wireless mesh topology, architecture and

information flow (to overcome deployment site

obstacles, such as walls in indoor environments, and

maximize communication reliability). The selection of

the network topology takes into account the building

types and installations and the maximum distance

from the gateway.

The NOD can receive user inputs via an intuitive

and easy to use interface. The users will be able to set

control settings on the different device types (HVAC

and lighting) and further receive information about the

current status of each device through a custom

designed user interface.

Figure 3: Block diagram of the NOD platform.

The final embedded system was designed around

10x10cm size (shown in Figure 3) and deployed in

offices for preliminary tests and characterization. The

main components of the NOD are the following.

Microcontroller: The heart of the system is the

STM32F303ZET6, a 32-bit ARM Cortex M4 Core,

72MHz Maximum, 512KB flash, 64KB RAM, USB

2.0 (ST, 2016).

Sensors:

- Proximity and Light sensor (VCNL4020)

(Vishay, 2014) interfaced to the

microcontroller via I2C and which can detect

proximity within 20cm. It is also able to

sensing light level in the range 0.25lux –

16klux. It has also an interrupt function that

will pull down the interrupt pin when the

sensor detects the presence within the set

threshold (20cm). This functionality will be

used to activate the screen of the NOD when

the device is in sleep mode.

- Temperature and Humidity sensor (Si7020)

(Silicon Labs, 2016), the humidity and

temperature sensors are factory-calibrated and

the calibration data is stored in the on-chip

non-volatile memory. This ensures that the

sensors are fully interchangeable, with no

recalibration or software changes required.

The sensor is interfaced to the microcontroller

via I2C.

- CO2 and VOC sensor (CCS811) (AMS,

2015), an ultra-low power digital gas sensor

solution which integrates a metal oxide

(MOX) gas sensor for monitoring indoor air

quality (IAQ) including a wide range of

Volatile Organic Compounds (VOCs) with a

microcontroller unit (MCU), an Analog-to-

Digital converter (ADC), and an I2C

interface. The CCS811 supports multiple

measurement modes that have been optimized

for low-power consumption during an active

sensor measurement and idle mode extending

battery life.

- Occupancy sensor (EKMB1301111K)

(Panasonic, 2016), a PIR motion sensor that

guarantees an optimal detection capability and

high reliability. The sensor is able to detect motion

in 5m range with 94° of horizontal detection area

and 82° of vertical. In order to have a 360° of

coverage the solution adopted for the NOD is to

use 4 sensors (one on each edge of the NOD

device) to cover the entire range (Figure 4). When

presence is detected the output of the sensor will

be pulled high and it will cause an interrupt for the

microcontroller.

Figure 4: Solution adopted for presence detection.

Radio Communication: in operation the NOD

will be part of a standardized wireless mesh network

and, in order to send data to the middleware, the NOD

needs to have at least one wireless communication

module on board. As the NOD has more than one

SMARTGREENS 2018 - 7th International Conference on Smart Cities and Green ICT Systems

176

wireless module on board, when compared to single-

end radio devices, it has the potential to provide

increased connectivity in deployment, and can

potentially reduce the interference impact on the

network, as the system can hop from ISM band to ISM

band in an autonomous and opportunistic manner to

minimize retransmission and reduce power

consumption.

- Bluetooth Low Energy (Cypress

CYW20737S) (Cypress, 2017), enables the

user to communicate with the NOD using a

smartphone or tablet App. The module

includes an embedded BLE antenna, 24 MHz

clock, and 512 Kb EEPROM. The Bluetooth

stack and several application profiles are built

into the module.

- Wi-Fi (ATWINC1500-MR210PB) (Atmel,

2015), for interfacing the device to the

middleware sending sensing data. It is a low-

power consumption 802.11 b/g/n IoT

(Internet of Things) module, which is

specifically optimized for low power IoT

applications with a fully integrated power

amplifier, LNA, switch, power management,

and PCB antenna. The embedded processor

performs many of the MAC functions,

including but not limited to: association,

authentication, power management, security

key management, and MSDU aggregation/de-

aggregation. In addition, the processor

provides flexibility for various modes of

operation, such as STA and AP modes.

- Sub-1GHz radio chip at 868MHz (SPIRIT1)

(St, 2016), enables the NOD as a multiradio

platform. It is a very low power consumption

device and the over the air data rate is

programmable from 1 to 500 kbps. It can be

used in systems with channel spacing of

12.5/25 kHz, complying with the EN 300 220

standard. The SPIRIT1 provides an optional

automatic acknowledgement, retransmission,

and timeout protocol engine in order to reduce

overall system costs by handling all the high-

speed link layer operations. It also supports an

embedded CSMA/CA engine and An AES

128-bit encryption co-processor for secure

data transfer.

Battery Management: the battery used is a

rechargeable Lithium-ion Polymer battery with a

capacity of 3000mAh, which is recharged through the

USB port or through the use of energy harvesting

systems compatible with the built environment

connected to the battery charger chip (MP2617)

(MPS, 2015). Onboard is also present a fuel gauge

chip used to track the battery relative state of charge

(SOC) continuously over widely varying charge and

discharge conditions.

External Interface:

- Micro USB, used for multiple purposes. It is

the power input for the board and for battery

charging and it is also available for data

transfer as it is connected on the on the

interface of the microcontroller (USB 2.0).

- JTAG, to download the firmware into the

microcontroller

- UART, enable the NOD to communicate

with external peripherals and it is also used for

debugging purpose)

Figure 5: Moeebius NOD Android App.

User Interface: Electronic Paper Display (EPD),

1.44’’, resolution of 128x96 (E1144CS021)

(Pervasive Displays, 2016) and 5 capacitive touch

buttons enable the human computer interface. The

display gives a feedback to the user of the real time

status of the NOD (battery level, connection status,

sensors data). Touch buttons are used as inputs for

setting different device types (HVAC and lighting).

Another solution for the user interface can be adopted

using a mobile app that will be able to connect to the

NOD (via Bluetooth LE). It shows the NOD’s status

and is able to receive inputs from the user to set

temperature and light level.

4 NOD CONFIGURATION AND

OPERATING MODE

In this section is described how the NOD can be

configured and the how it operates in the MOEEBIUS

Network.

Innovative Low Power Multiradio Sensing and Control Device for Non-Intrusive Occupancy Monitoring

177

4.1 Configuration

In order to act as sensing device as part of a wireless

network the NOD needs to be configured when it is

powered on for the first time. This can be done through

the Wi-Fi provisioning, a process of connecting a new

Wi-Fi device (station) to a Wi-Fi network. The

provisioning process involves loading the station with

the network name (often referred to as SSID) and its

security credentials.

Wi-Fi AP Provision mode primarily demonstrates

how to configure the credentials (such as, SSID and

Passphrase) in ATWINC1500 remotely. The

configured credentials are used to connect with a

desired access point.

Remote configuration facilities such as AP

provisioning and HTTP provisioning modes are

available. In HTTP provisioning mode, the HTTP

page is used to configure the credentials. The HTTP

server is running in the WINC firmware and the HTTP

page is also stored in the WINC flash memory

(Microchip, 2017).

The NOD starts as a SoftAP using open

security mode (no security method) and

broadcasts the beacon frames with SSID

MOOEBIUS_PROVISION_AP_IDx (where Idx is

the identifier of the NOD).

A smartphone or tablet can be used to connect to

this AP. A web page will be opened where it is

possible to set the SSID and password for the

MOEEBIUS Network that the NOD will join. Those

parameters are stored in the NOD so every time it is

powered on it will connect to the network

automatically.

Figure 6: Provisioning Web Page.

This procedure needs to be done for each NOD

that will be part of the MOEEBIUS wireless network.

To save time and enable auto configuration this

can be done only on one NOD device (master) and it

will broadcast the information to the other NODs

through the Sub1-GHz module.

Other modalities to configure the device could be

through a smartphone/tablet App that send the

configuration parameters (SSID and password of the

wireless network) using the Bluetooth connection

4.2 Nod Operating Mode

As described in the previous section, when the NOD

is powered up it needs to be configured (it has to be

done only once) and the parameters (SSID and

password) are stored in the flash memory of the Wi-Fi

module. Then the NOD will scan for the Wi-Fi

networks available and it will connect to the

configured one; the NOD is connected to the Internet.

Once the connection is established the NOD will

configure the hardware driver (I2C, UART, SPI,

ADC) in order communicate with the sensors and

communication protocols on board.

By default the microcontroller is programmed to

read sensors value (Temperature, Humidity, Light

level CO2 and VOC) every 5 minutes. Duty cycle can

be configured by the user through the BLE App or

using the touch buttons. Another option is that the

server can send a command to the NODs in order to

change the time between two readings.

When the sensors data are available the NOD will

create a packet with all the readings and it will be sent

to the server in the JSON format using the HTTP

REST API. Along with the sensor data and its

reliability, the packet contains timestamp, country ID,

building ID and point name. An example is the

following:

[{

"timestamp":"2017-02-

16T14:47:00+00:00",

"countryid": "IE",

"buildingid": "TY00",

"pointname": "NOD01.CO2",

"value": "400",

"reliability":"1"

}]

Moreover all the sensors are shown on the display

along with the battery level, connection status and

current date and time.

To save power the NOD will enter in sleep mode

after the packet is sent, then it will wake up after 5

SMARTGREENS 2018 - 7th International Conference on Smart Cities and Green ICT Systems

178

minutes to read data from sensors and send them to the

server.

When the NOD is in sleep mode, it can be woken

up from different sources:

- Occupancy sensors: if presence is detected in

the range of 5 meters the microcontroller will

be woken up by the sensors interrupt and a

packet containing occupancy information

will be sent to the middleware.

- Touch button: if this button is touched by the

user the system wakes up and it will be ready

to receive user’s inputs.

- Proximity interrupt: this is an alternative to

the touch button, so if the user wants to

wakes up the NOD, he has to put his hand

close to the device (distance less than 20cm).

- Sub-1GHz: the NOD is a multiradio device

and when it is in sleep mode and receives a

packet from another device (that can be a

command or configurations instruction) on

the 868MHz ISM band the microcontroller

will wake up and elaborate the information

received.

- Fuel gauge: as described the NOD has a

battery monitoring circuit that will send an

interrupt to the microcontroller when the

battery voltage is lower than a threshold in

order to tell the user to charge the battery.

If the user wants to set desidered temperature and

light level, he simply has to navigate into the menu and

select the temperature or the light level. When the

wanted level is reached (and displayed on the screen)

the user needs to confirm the action. In this case a

command is sent to the middleware in order to set the

user’s preferences regarding environmental

conditions. The middleware will send back a

confirmation that the action has been done.

5 POWER CONSUMPTION

ANALYSIS

The current consumption analysis has been carried

out for the normal operation of the NOD (after power

up) that comprises the following behavior:

- Microcontroller, battery charger, fuel gauge

and 4 occupancy sensors always ON

- Read sensors data one at time. After each

sensor reading, it is set to standby mode

- Display sensors data on EPD

- Turn off EPD display

- Turn on Wi-fi and establish connection with

the server

- Transmit data to the server

- Turn off Wi-fi

- Microcontroller goes into sleep mode for 5

minutes

Using the Agilent Technologies N6705B DC

Power Analyzer, the current consumption for each of

the main components of the NOD are reported in

Table 1.

Figure 7 shows the NOD current consumption

profile during the normal operational mode: power on,

initialization, sensors readings and display, Wi-Fi

transmission and sleep mode.

Table 1: NOD components current consumption.

NOD components Current consumption

VCNL4020 (Light and

Proximity sensor)

Standby = 0.02 mA

Sensing = 0.45 mA

CCS811

(CO2/VOC sensor)

Standby = 0.05 mA

Sensing = 29.5 mA

Si7020 (Temperature and

Humidity sensor)

Standby = 0.01 mA

Sensing = 0.2 mA

EKMC1600100

(occupancy sensors)

Sensing = 0.12 mA

STM32F302

(Microcontroller) and all

sensors in stand by mode

ON (no operations) = 8.5 mA

Sleep mode = 4.43 mA

Stop mode = 2.96 mA

Deep sleep mode = 2.69 mA

E144CS021

(EPD display)

Display on = 6.5 mA

ATWINC1510 (Wi-Fi)

Access point mode = 111 mA

Transmitting = 110 mA

Figure 7: NOD current consumption profile (not to scale).

Innovative Low Power Multiradio Sensing and Control Device for Non-Intrusive Occupancy Monitoring

179

The average current consumption measured is

~6.23 mA.

A fully charged 3000 mAh rechargeable battery

supplying the average current will last for

approximately 20 days (1):

No. of days =



.

∗





(1)

This is an estimated value based on the

presumption if the NOD is reading sensors data every

5 minutes configuring the sensors in the lowest power

consumption mode, without any other interaction with

the user (touch buttons and LEDs) or using the

Bluetooth and sub-1Ghz features. Time for reading,

display and transmission data are set at maximum

values, so reducing them the current consumption will

be less.

6 CONCLUSIONS/FUTURE

WORK

This work describes the development and preliminary

characterization of a novel low power consumption

multiradio system for real-time indoor air quality

monitoring incorporating three radio interfaces, WiFi,

Bluetooth LE and 868MHz with all its features and

operating modes that bring it to a sensing/ control

Smart system (integrating a wide range of sensors and

control interface) that is easy to install and configure.

It also provides a solution for network congestion in

environment such as Home Area Network and

Commercial Buildings and Offices thanks to the

embedded multiradio feature. Additional

characterization and optimization of the system are

currently underway and future work will be focused

mostly on the firmware side improving the code

adding the multiradio features using the sub-1GHz

radio implementing protocols for reduced power

consumption.

ACKNOWLEDGEMENTS

This publication has emanated in part from research

supported in part by Science Foundation Ireland (SFI)

and is co-funded under the European Regional

Development Fund, Grant Number 13/RC/2077-

CONNECT. Aspects of this work have been funded

by the European Union Horizon 2020 project

MOEEBIUS under grant agreement 680517.

REFERENCES

International Energy Agency, 2008. Energy Efficiency

Requirements in Building Codes, Energy Efficiency

Policies for New Buildings. [pdf] Paris: OECD/IEA. <

https://www.iea.org/publications/freepublications/publ

ication/Building_Codes.pdf>. [Accessed August 2017].

AMS, 2015. CCS811 Datasheet. [pdf] <https://ams.com/

eng/content/download/951091/.../CCS811_DS000459_

3-00.pdf>. [Accessed June 2017].

AQMD, 2015. Laboratory Evaluation Dylos-DC1700 PM

Sensor. [pdf] AQMD. <http://www.aqmd.gov/docs/

default-source/aq-spec/laboratory-evaluations/dylos---

lab-evaluation.pdf?sfvrsn=2> [Accessed October 2017].

Atmel, 2015. ATWINC1500-MR210PB Datasheet. [pdf]

<http://www.atmel.com/images/atmel-42376-smartcon

nect-winc1500-mr210pa_datasheet.pdf> [Accessed

June 2017].

BTNode, 2007. BTnode rev3 hardware reference. [online]

<http://www.btnode.ethz.ch/Documentation/BTnodeRe

v3HardwareReference > [Accessed October 2017].

Cypress, 2017. CYW20737S Datasheet. [pdf]

<http://www.cypress.com/file/299481/download>.

[Accessed June 2017].

Fang B., Xu Q., Park T., and Zhang M., 2016. Airsense: an

intelligent home-based sensing system for indoor air

quality analytics. In Proc. of ACM UbiComp, 2016.

Feng Zhao, Leonidas, and Guibas J., 2004. Wireless Sensor

Networks: An Information Processing Approach,

Burlington: Morgan Kaufmann.

Harwood M., 2009. CompTIA Network+ N10-004 Exam

Cram, 3rd Edition. Indianapolis: Pearson IT

Certification.

Jiang Y, Li K., Tian L., Piedrahita R., Yun X., Mansata O.,

Lv Q., Dick R. P., Hannigan M., and Shang L.. Maqs: A

personalized mobile sensing system for indoor air

quality. In: UbiComp 2011. 13th International

Conference on Ubiquitous Computing. Beijing, China,

17-21 September 2011. ACM.

LIBELIUM, 2017. Waspmote Datasheet. [pdf]

<http://www.libelium.com/downloads/documentation/

waspmote_datasheet.pdf > [Accessed October 2017].

Microchip, 2017. ATWINC15x0 – Wi-Fi Network Controller

Software Programming Guide. [pdf] < http://ww1.micro

chip.com/downloads/en/DeviceDoc/70005305A.pdf>.

[Accessed June 2017].

MOEEBIUS, 2016. About the project. [online]

<http://www.moeebius.eu/about-the-project>

[Accessed September 2017].

MPS, 2015. MP2617 Datasheet. [pdf] <

http://www.mouser.com/ds/2/277/MP2617_r1.04-

479044.pdf>. [Accessed June 2017].

O'Flynn B, De Donno M, and Magnin W. (2016)

'Multiradio, Multiboot capable sensing systems for

Home Area Networking', Int’ Journal on Advances in

Networks and Services, 9 (3&4), pp.96-106.

Panasonic, 2016. EKMB1301111K Datasheet. [pdf]

<http://versasense.com/files/datasheets/panasonic_EK

MB1201112.pdf>. [Accessed June 2017].

SMARTGREENS 2018 - 7th International Conference on Smart Cities and Green ICT Systems

180

Pervasive Displays, 2016. E1144CS021 Datasheet. [pdf] <

http://www.pervasivedisplays.com/LiteratureRetrieve.a

spx?ID=232432>. [Accessed June 2017].

Silicon Labs, 2016. Si7020 Datasheet. [pdf] <https://

www.silabs.com/documents/public/data-sheets/Si7020-

A20.pdf>. [Accessed June 2017].

ST, 2016. SPIRIT1 Datasheet. [pdf] < http://www.

atmel.com/images/atmel-42376-smartconnect-winc150

0-mr210pa_datasheet.pdf>. [Accessed June 2017].

ST, 2016. STM32F303xE Datasheet. [pdf] <

http://www.st.com/content/ccc/resource/technical/docu

ment/datasheet/2c/6f/d7/64/1f/a3/4f/c9/DM00118585.p

df/files/DM00118585.pdf/jcr:content/translations/en.D

M00118585.pdf> [Accessed June 2017].

VISHAY, 2014. VCNL4020 Datasheet. [pdf] <

https://www.vishay.com/docs/83476/vcnl4020.pdf>.

[Accessed June 2017].

Wyon, D. P., 2004. The effects of indoor air quality on

performance and productivity. Indoor Air, 14(s7),

pp.92-101.

Innovative Low Power Multiradio Sensing and Control Device for Non-Intrusive Occupancy Monitoring

181