13th Int. Particle Acc. Conf.

IPAC2022, Bangkok, Thailand JACoW Publishing

Content from this work may be used under the terms of the CC BY 4.0 licence (© 2022). Any distribution of this work must maintain attribution to the author(s), title of the work, publisher, and DOI
ISBN: 978-3-95450-227-1 ISSN: 2673-5490 doi:10.18429/JACoW-IPAC2022-TUOXGD2

WIRELESS IOT IN PARTICLE ACCELERATORS: A PROOF OF CONCEPT

WITH THE IOT RADIATION MONITOR AT CERN
Salvatore Danzeca, Alessandro Zimmaro, Tony Cass, Alessandro Masi, Rodrigo Sierra, CERN
CH-1211, Geneva 23, Switzerland

Abstract
The Internet of Things (IoT) is an ecosystem of web-
IoT AND RADIATION MONITORING
enabled “smart devices” that integrates sensors and IoT
communication hardware to collect, send, and act on data
The term IoT includes everything connected to the
acquired from the environment around them. In particle
internet. It is a concept that covers devices, networks,
accelerators, IoT is not a new concept: systems and
services, and data. All these layers were defined in the 2014
instruments have long been connected to the network to
Cisco “reference” model [1]. The seven layers defined in
receive and to send data that are stored, analysed, and later
the reference model are not uncommon to particle
used. What has been missing so far, is the IoT concept of
accelerators; indeed, all the equipment, sensors, and
“smart devices”, and particularly of “wireless
devices are connected and well-integrated in what is
connectivity”. Wireless technologies play an important role
usually called a “control system” [2]. One of the big
in the Internet of Things. Indeed, they allow to deploy of
differences between the two relies on the connectivity
operational devices quickly and easily, and they
possibilities. Particle accelerators strongly rely on cabled
significantly reduce infrastructure costs. This paper gives
networks and wired infrastructure, while the IoT concept
an overview of the main advantages of using LoRa, a
gained great momentum mainly thanks to a new generation
particular IoT technology used to deploy wireless radiation
of wireless technologies. If implemented in a particle
monitors within the CERN particle accelerator complex.
accelerator, this allows quick connectivity, easy
INTRODUCTION installation, and virtually no cabling. This has a huge
impact on a) availability: being easy to install when needed
At CERN, measurements of the radiation fields and with small deadtime b) observability: giving the multiple
quantities related to damage to electronics equipment are positions that the devices can reach, and, obviously, c) cost.
carried out by the RadMon system. The RadMon devices
allow to measure radiation levels in the accelerator RadMon
complex and investigate if instantaneous failures of In this work, the radiation monitoring system for CERN
electronic systems are caused by radiation. As with other electronics serves as a practical example of how to embrace
accelerator instruments, these devices have limited IoT technology, highlighting its main requirements and
mobility and are subject to a fixed location, which is capabilities. RadMon is a system capable of measuring the
usually close to critical equipment in the beam lines or main quantities related to the radiation effects on
experimental caverns. Wireless IoT Radiation Monitors electronics, such as the Total Ionizing Dose (TID), the
have been developed as the result of the growing demand Displacement Damage (DD) [3] the High Energy Hadron
to carry out radiation measurements wherever the cabled fluence (HEH), and the Thermal Neutron fluence (ThN).
infrastructure is not available. Here we describe the This device is used to monitor the radiation levels on the
technologies used for the project and the various electronic systems in the accelerators, anticipate electronic
advantages of their deployment in a particle accelerator degradation, benchmark simulations and help in the
environment. Although the focus is on radiation investigation into the cause of failures.
monitoring applications, the work is aimed at developing The RadMon system is fully integrated into the CERN
an IoT platform capable of integrating multiple and diverse infrastructure [3]. It requires two cables, for power and for
sensors and applications. This approach opens up the way communication with a Front-End Computer (FEC) which
for the deployment of different and heterogeneous can manage up to 32 devices on same Fieldbus [4]. In
implementations and applications that would have been operation, it’s very common for users to request
impractical so far. In the first section, the IoT and the measurements in locations where RadMons are not
RadMon concepts are discussed along with their use and installed. It is very difficult, even impossible, to plan
impact in the context of the control of particle accelerators. proactively, because they respond to the need of tracking
The second section describes the requirements and the failures on specific equipment. For these operational cases,
characteristics of a wireless IoT network for particle a wireless, battery-powered, standalone radiation monitor
accelerators, as well as the technology used in this proof of has been introduced. The challenges of these types of
concept. developments are a) wireless communication from a few
Finally, the IoT platform developed for radiation hundred to a few km range with a data rate of ~1kB/h; b)
monitoring is presented along with the challenges and the low power electronics capable of running over batteries for
possibilities that can be exploited in the future. at least one year; c) radiation tolerance up to at least 250
Gy and d) low cost.
TUOXGD2 MC6: Beam Instrumentation, Controls, Feedback and Operational Aspects
772 T18: Radiation Monitoring and Safety
13th Int. Particle Acc. Conf. IPAC2022, Bangkok, Thailand JACoW Publishing

Content from this work may be used under the terms of the CC BY 4.0 licence (© 2022). Any distribution of this work must maintain attribution to the author(s), title of the work, publisher, and DOI
ISBN: 978-3-95450-227-1 ISSN: 2673-5490 doi:10.18429/JACoW-IPAC2022-TUOXGD2

WIRELESS TECHNOLOGIES proper RF modulation methods and with additional layers

for data acknowledgments.
In this paper, a Wireless Sensor Network (WSN) is
described. It aims to monitor different devices and sensors
and send data to a central system for storage and analysis.
Tunnels Wireless Networking Solutions
An effective wireless sensor network must be power
efficient, scalable, responsive, reliable, and mobile. In
underground tunnel environments, such as particle
accelerators, other constraints should be considered such as
RF interference and ionizing radiation. Figure 1: LoRaWAN architecture implemented in the LHC
Wireless technology emission in CERN’s underground accelerator.
facilities and experimental areas are coordinated with the The radiating cable used at CERN is suited for
host state radio frequency regulator (ANFR in France and communication at frequencies below 1GHz, and the use of
OFCOM in Switzerland) to prevent any interference or the ISM band as the carrier frequency for LoraWAN
disruption with external RF services. At the same time, the perfectly fits this requirement.
selected frequency ranges must avoid any RF system Nowadays, LoRaWAN is present in the entire LHC
present in the area for physics or operational purposes tunnel. The gateways are installed in the shielded caverns
(accelerators require RF cavities to perform their duties as connected to the radiating cables in the tunnel. This allows
an example). The second constraint (ionizing radiation) is having full coverage of the LHC. In Fig. 1, is possible to
more restrictive. The wireless technology emitter must be see the LoRaWAN architecture implemented in the LHC
either radiation tolerant or installed in a shielded area. accelerators.
When coupled with tunnel lengths of more than 1 km
without shielded zones, the topology choice for wireless WIRELESS MONITORING PLATFORM
solutions is significantly reduced.
To offer wireless connectivity in the accelerator and An IoT Radiation-Tolerant Platform
experimental areas, CERN deployed around 60 km of The IoT Wireless monitoring platform [8], is a LoRaWan
radiating cables connected to more than 50 injection points enabled, battery-powered, radiation-tolerant, and modular
located in non-radiation areas [5-6]. However, the system capable of facing the requirements described in the
frequency bands used must be below 1 GHz, since the previous paragraphs. All the embedded components are
losses in the cable increase with the frequency. The Components Off the Shelf (COTS), which ensure
radiating cable is already used for TETRA and compliance with low-cost and more stringent low power
TETRAPOL in the 400 MHz band for CERN’s fire brigade requirements. This choice required intense radiation
and public safety agencies, UMTS in the 900 MHz band, qualification and conformity testing following the CERN
LTE in the 800 MHz band, and 5G in 700 MHz band. The guidelines for radiation tolerant system testing [8-10]. All
ISM band in Europe (868 MHz) is not used and was a good the components embed low-power features and modes,
candidate to explore IoT wireless network solutions in the which allowed the system to respect lifetime requirements.
tunnels. An overview of the system is depicted in Fig. 2. System
Low Power Wireless Solutions operations are handled by a microcontroller (MCU).
This work is focused on Low Power Wide Area Network
solutions (LPWAN), because they are extensively used in
industrial IoT projects and are suited for long range and
low data rate applications, such as that required for
radiation monitoring of the electronics in the CERN
accelerator complex. Three main LPWAN solutions were
investigated: LoRaWAN, SigFox, and NB-IoT [7]. The
three networks are similar in terms of scope and features.
LoRaWAN was finally chosen because it is provider-
independent which allows private networks to be created
with open-source tools. In a LoRaWAN network, data Figure 2: Overview of the Wireless IoT platform.
transmitted by a node are typically received by multiple This component is preferred over the more widely used
gateways. Each gateway forwards the received packet from Field Programmable Gate Array (FPGA) in radiation
the end node to the network server. This architecture allows environments because of its power-saving mode that
redundancy and high coverage with the drawback of safeguards the battery duration. A hardware solution
possible collisions and bottlenecks in communication. (watchdog circuit) was implemented to mitigate Single
These last problems are mitigated in LoRaWAN with Event Function Interruption (SEFI) [11]. A radio
transceiver was used to implement the physical and MAC

MC6: Beam Instrumentation, Controls, Feedback and Operational Aspects TUOXGD2

T18: Radiation Monitoring and Safety 773
 13th Int. Particle Acc. Conf. IPAC2022, Bangkok, Thailand JACoW Publishing

Content from this work may be used under the terms of the CC BY 4.0 licence (© 2022). Any distribution of this work must maintain attribution to the author(s), title of the work, publisher, and DOI
ISBN: 978-3-95450-227-1 ISSN: 2673-5490 doi:10.18429/JACoW-IPAC2022-TUOXGD2

layer of the LoRaWAN protocol. In case of wireless 2/year), the system should work for over 10 years and reset
communication unavailability, a 16 MB SPI non-volatile less than once per year.
Flash memory is embedded in the design. The IoT
Radiation device was developed for radiation to electronics Operational Feedbacks
measurements using a dedicated sensor board derived from A first test on the long-range capability of the IoT
the RadMon. Its modularity using a detachable sensor Wireless monitoring platform was successfully performed
board allows possible future users to implement their own inside the LHC[14]. The test was carried out in Point 1
custom sensor. According to the same concept, the battery where 2 LoRaWAN gateways were installed in the UPS
board can also be swapped with a radiation tolerant power RE18 alcove. Every 200 minutes five packets transmitted
supply inherited from the RadMon system. by the tested device were collected, and the corresponding
Received Signal Strength Indicator (RSSI) was analyzed.
Low Power Design Choices In Fig. 3, the RSSI is depicted for each transmission
Low power constraints and the need for flexibility drove point. The image shows that the device was still able to
most of the design choices. The operating principle also transmit the information over 2 km. This test proved that
considers these two characteristics. The system is based on LoRa technology is a valid network that fits the long-range
a duty cycle consisting of cycles of sleep, wake-up, constraint of such applications in an accelerator.
measurements, storage, transmission, and sleep. Increasing
the interval between two consecutive awakenings, also
called the “measurement period”, allows the battery life to
be improved. As reported in Table 1, extension of the
measurement period allows to meet the requirement of
more than one year of lifetime. The flexibility to allow the
measurement period to be changed on-the-fly when finer
measurements are requested, is a feature currently being
developed.
Table 1: System Lifetime as a Function of the
Measurement Period
Figure 3: This picture shows the RSSI degradation when
Measurement Period Lifetime 𝜏 [Month] moving further away from the gateway position
5 Minutes 3.31 (RE18)[14].
1 Hour 22.9
24 Hours 47.36 In 2021, several systems were deployed to characterize
the radiation levels of specific areas in CERN’s Super
Radiation Tolerance Proton Synchrotron (SPS) [15]. For these measurements,
The radiation qualification of the IoT Wireless the IoT Radiation platform was used in standalone mode,
monitoring platform can be divided into 2 main phases as because the wireless network was not available. The
described in [8]: component level and system level standalone mode tested the IoT Radiation platform
qualification. During the first screening phase, all the performance for storing the measurements in the non-
components were qualified at the Paul Scherrer Institute volatile flash memory. These tests not only provided useful
(PSI) in Villigen, Switzerland, using a high-energy proton measurements for the characterization of these locations
beam of 200 MeV [12] and then selected according to the but also validated the proper functioning of the system.
radiation tolerance requirements of the system. In a second
phase, the platform was tested at system level with a final
CONCLUSION
validation at the CERN High energy AcceleRator Mixed As part of this work, the opportunities for utilizing a
field (CHARM) facility [13]. Those tests allowed the wireless IoT technology have been demonstrated by using
system to be assessed in terms of sensitivity and radiation a radiation monitoring device as a test application. The
tolerance, as well as being used to validate single event system requirements, the wireless challenges, and
mitigation schemes. The platform withstood a TID of 275 implementations have been described. LoRaWAN was
Gy [8]. With this information, the expected lifetime and found to be the best choice for implementing a wireless
failure rate can be estimated by considering specific sensor network in the accelerator, fitting all the
locations for the future High-Luminosity (HL) LHC [9]. In requirements. The Wireless IoT platform for radiation
very harsh tunnel areas, such as some of the magnet cells monitoring has been developed to be both radiation
of the dispersion suppressor region, where a TID of 100 tolerant and compliant with this IoT technology. The
Gy/year and 3·1012 [Link]-2/year is foreseen, the proper tuning of the device parameters allows it to last on
system should work for nearly 3 years performing self- batteries for up to three years. The platform is modular with
recovered resets around 200 times per year. Conversely, in the capability of being adapted for other applications.
shielded alcove areas where the foreseen radiation levels Finally, the wireless capabilities of the device have been
are lower (TID of 25 Gy/year and 1.4·1010 [Link]- tested successfully inside the LHC with a coverage of
around 2 km. Currently, the LoRa connectivity is available

TUOXGD2 MC6: Beam Instrumentation, Controls, Feedback and Operational Aspects

774 T18: Radiation Monitoring and Safety
13th Int. Particle Acc. Conf. IPAC2022, Bangkok, Thailand JACoW Publishing

Content from this work may be used under the terms of the CC BY 4.0 licence (© 2022). Any distribution of this work must maintain attribution to the author(s), title of the work, publisher, and DOI
ISBN: 978-3-95450-227-1 ISSN: 2673-5490 doi:10.18429/JACoW-IPAC2022-TUOXGD2

in the entire LHC tunnel allowing the deployment and the

usage of this wireless platform. [9] R. García Alía et al., “LHC and HL-LHC: Present and Fu-
ture Radiation Environment in the High-Luminosity Colli-
REFERENCES sion Points and RHA Implications”, in IEEE Transactions
[1] Cisco, “The internet of things reference model.” White Pa- on Nuclear Science, vol. 65, no. 1, pp. 448-456, Jan. 2018.
per, 2014. doi:10.1109/TNS.2017.2776107
[Link] [10] S. Uznanski et al., “Qualification of electronics components
ence_Model_White_Paper_June_4_2014.pdf for a radiation environment: When standards do not exist -
[2] S. Deghaye and E. Fortescue-Beck, “Introduction to the BE- High Energy Physics”, CERN, Geneva, Switzerland, Tech.
CO Control System,” CERN Document Server, Dec. 18, Rep, 2017. [Link]
2020. [Link] [11] R. G. Alía et al., “Single event effects in high-energy accel-
[3] R. Ferraro et al., "Study of the Impact of the LHC Radiation erators”, Semicond. Sci. Technol., vol. 32, no. 3, p. 034003,
Environments on the Synergistic Displacement Damage and Feb. 2017. [Link]
Ionizing Dose Effect on Electronic Components," in IEEE cle/10.1088/1361- 6641/aa5695/meta
Transactions on Nuclear Science, vol. 66, no. 7, pp. 1548- [12] W. Hajdas, F. Burri, C. Eggel, R. Harboe-Sorensen and R.
1556, July 2019, doi: 10.1109/TNS.2019.2902441 de Marino, “Radiation effects testing facilities in PSI during
[4] G. Spiezia et al., “A New RadMon Version for the LHC and implementation of the Proscan project”, IEEE Radiation Ef-
its Injection Lines,” IEEE Transactions on Nuclear Science, fects Data Workshop, pp. 160-164, 2002.
vol. 61, no. 6, pp. 3424–3431, Dec. 2014. doi:10.1109/REDW.2002.1045547
doi:10.1109/TNS.2014.2365046 [13] J. Mekki, M. Brugger, R. G. Alia, A. Thornton, N. C. D. S.
[5] S. Agosta, R. Sierra, and F. Chapron, “High-Speed Mobile Mota and S. Danzeca, “CHARM: A Mixed Field Facility at
Communications in Hostile Environments”, J. Phys. Conf. CERN for Radiation Tests in Ground, Atmospheric, Space
Ser., vol. 664, no. 5, p. 052001, Dec. 2015, and Accelerator Representative Environments”, in IEEE
doi:10.1088/1742-6596/664/5/052001 Transactions on Nuclear Science, vol. 63, no. 4, pp. 2106-
2114, Aug. 2016. doi:10.1109/TNS.2016.2528289
[6] R. Sierra and H. Odziemczyk, “EDP Sciences : Readying
CERN for connected device era”, EPJ Web Conf., vol. 245, [14] M. Cejp, “Wireless radiation monitoring system for particle
p. 07015, 2020, doi:10.1051/epjconf/202024507015 accelerators”, Ms. C. thesis, Department of Measurement,
Czech Technical University in Prague, September 2020.
[7] K. Mekki, E. Bajic, F. Chaxel, and F. Meyer, “A compara-
tive study of LPWAN technologies for large-scale IoT de- [15] Y. Q. Aguiar et al., “Implications and mitigation of radiation
ployment,” ICT Express, vol. 5, no. 1, pp. 1–7, Mar. 2019, effects on the CERN SPS operation during the machine
doi:10.1016/[Link].2017.12.005 commissioning of run 3”, presented at the 13th Int. Particle
Accelerator Conf. (IPAC’22), Muang Thong Thani, Thai-
[8] A. Zimmaro et al., “Testing and Validation Methodology for land, June 2022, this conference
a Radiation Monitoring Systems for Electronics in Particle
Accelerators”, in IEEE Transactions on Nuclear Science,
doi:10.1109/TNS.2022.3158527

MC6: Beam Instrumentation, Controls, Feedback and Operational Aspects TUOXGD2

T18: Radiation Monitoring and Safety 775