1

Ambient IoT: A missing link in 3GPP IoT Devices

Landscape
M. Majid Butt, Nitin R. Mangalvedhe, Nuno K. Pratas, Johannes Harrebek, John Kimionis, Muhammad Tayyab,
Oana-Elena Barbu, Rapeepat Ratasuk, and Benny Vejlgaard

Abstract—Ambient internet of things (IoT) is the network of business case. Energy harvesting technology is key to the
devices which harvest energy from ambient sources for powering success of the IoT market as it can considerably reduce device
their communication. After decades of research on operation operational cost. Energy harvesting devices are ’battery free’
arXiv:2312.06569v1 [cs.NI] 11 Dec 2023

of these devices, Third Generation Partnership Project (3GPP)

has started discussing energy harvesting technology in cellular devices, which harvest energy from natural or ambient sources,
networks to support massive deployment of IoT devices at low e.g., electromagnetic, solar, thermal, pressure etc., and operate
operational cost. This article provides a timely update on 3GPP either with a small battery or without battery and do not require
studies on ambient energy harvesting devices including device battery replenishment. They are also termed as ’zero energy’
types, use cases, key requirements, and related design challenges. devices due to their capability to operate without a dedicated
Supported by link budget analysis for backscattering energy
harvesting devices, which are a key component of this study, we power source [2].
provide insight on system design and show how this technology Energy harvesting communication has been widely dis-
will require a new system design approach as compared to New
Radio (NR) system design in 5G.
cussed in research for more than a decade, e.g., see [3],
[4] and references therein. Recent literature outlines key
Index Terms—Energy harvesting, ambient IoT, backscattering, challenges and complexity to power massive number of IoT
3GPP.
devices [5], [6] using radio frequency (RF) signals, but this is
the first time 3GPP has started discussing energy harvesting
I. I NTRODUCTION and backscattering as a potential IoT technology for future
The internet of things (IoT) market is a major source networks in various study groups [7], [8]. Fundamental aspects
of revenue for future wireless technology. Moving beyond of the technology are well researched, but feasibility of the
enhancement of the mobile broadband devices market that technology in devices under 3GPP networks is still unknown
is supported by fourth-generation (4G) cellular technology, under realistic network operating conditions and business
ultra reliable low-latency communication (URLLC) and ma- model.
chine type communication (MTC) service classes were also The goal of this article is to articulate the main reasons
introduced in fifth-generation (5G) cellular technology, which behind industry interest in energy harvesting as a future
target use cases requiring high reliability and massive de- technology and discuss design considerations and solutions. As
ployment of smaller and cheaper IoT devices, respectively. 3GPP has never before discussed energy harvesting devices,
Sixth-generation (6G) technology will further expand these also called ambient IoT (A-IoT) devices, there are several
market segments. The IoT market is expected to steadily grow challenges to be addressed before energy harvesting technol-
from 2020 to 2025, and further increase is expected beyond ogy is adopted in future networks. This article provides a
that period at a higher rate. Building and commercial, health, comprehensive overview addressing three key questions:
agriculture, infrastructure, and industries are some of the key
sectors targeted by the IoT market. Almost all sectors are • How will energy harvesting IoT technology complement
forecasted to grow in the future, but infrastructure and health existing 3GPP IoT technologies?
segments are expected to grow at higher rates in the future • What are the key use cases and the associated design
compared to others [1]. requirements for the technology?
Due to the large number of IoT devices that are expected • What are the challenges associated both with technology
to be deployed, growth of the IoT market requires future as well as its standardization in 3GPP?
networks to be energy efficient and sustainable to make a
Section II discusses the first question by describing existing
M. Majid Butt, Nitin R. Mangalvedhe and Rapeepat Ratasuk are with
Nokia Standards, Naperville, USA. email:{majid.butt, nitin.mangalvedhe, 3GPP IoT technologies and how energy harvesting technology
rapeepat.ratasuk}@nokia.com. fills the technology gap. Section III addresses the second
Nuno K. Pratas, Johannes Harrebek, Oana-Elena Barbu, and question by reviewing recent status of 3GPP discussions on
Benny Vejlgaard are with Nokia Standards, Aalborg, Denmark.
email: {nuno.kiilerich pratas, johannes.harrebek, oana-elena.barbu, energy harvesting technology and the way forward, while
benny.vejlgaard}@nokia.com. Section IV and Section V attempt to answer the third question
John Kimionis is with Nokia Bell Labs, 600 Mountain Ave, Murray Hill, by focusing on link-level analysis and the main technology
NJ, 07974, USA (e-mail: ioannis.kimionis@nokia-bell-labs.com.)
Muhammad Tayyab is with Nokia Standards, Oulu, Finland. (e-mail: challenges. Finally, we conclude with the main findings of the
muhammad.tayyab@nokia.com. article in Section VI.
 2

Fig. 1. Cellular IoT Technology landscape with major technologies covering different potential use cases.

II. I OT SPACE IN 3GPP power consumption, and small form factor, which can be met
The expansion of IoT to encompass use cases in new by devices that are either battery-less or have the capability
verticals has steadily progressed from the initial applications for limited energy storage. The requirements related to power
in second-generation (2G) cellular technology to the current consumption and device cost/complexity cannot be met by
5G technologies. IoT use cases span a range of requirements in existing cellular technologies. Therefore, new technologies
terms of cost, complexity, delay constraints, and other metrics. under NR in Rel-19 or later and 6G are needed.
Fig. 1 depicts the division of IoT based on requirements into Fig. 1 also shows some key performance indicators (KPIs)
the following classes. for the different cellular IoT technologies [10]. It can be seen
Broadband IoT: This class of IoT comprises one extreme that in terms of peak data rate, broadband IoT is at the high
of IoT applications that require high data rates along with end whereas massive IoT is at the low end. Among the legacy
a low latency. The use cases covered by this class of IoT classes, in terms of reliability and latency, KPIs for critical
include fleet management, industry gateways, video, hot-spots, IoT are the most stringent whereas those for massive IoT are
wearables, and industrial wireless sensors. The requirements the most relaxed. A-IoT is expected to have even more relaxed
can be served by 4G Long Term Evolution (LTE) as well as KPIs.
5G New Radio (NR) technologies (using enhanced Mobile
III. A-I OT U SE C ASES AND R EQUIREMENTS
Broadband, or eMBB, and Reduced Capability or RedCap,
devices). RedCap devices have a lower cost and complexity A. Technology Overview
than NR eMBB devices, making them more attractive for many A-IoT technology includes both energy harvesting devices
IoT applications. with active transmission as well as passive backscattering
Critical IoT: This class of IoT constitutes another extreme of devices. Backscattering devices do not have an active trans-
IoT applications requiring ultra-high reliability and ultra-low mission component and modulate information on the received
latency. Typical use cases for this class include factory au- signal from the exciter where the exciter (often used inter-
tomation, industrial control, robotics, and Augmented Reality changeably with the terms illuminator or activator) can be
(AR) or Virtual Reality (VR). The LTE High-Reliability Low- any node generating RF signals. On the other hand, energy
Latency Communication (HRLLC) feature of LTE, introduced harvesting with active transmission and small storage allows
in Release 15, and the NR URLLC technologies feature of NR, higher range and better quality of service (QoS) as compared
also introduced in Release 15, can serve these applications. to backscattering devices without any storage. Different use
Massive IoT: This class of IoT consists of delay-tolerant cases can be envisioned for both technologies, where low-
applications where the requirements include low-cost devices cost backscattering technology is more suitable to use cases
with low energy consumption, extended network coverage, and such as asset tracking and monitoring, livestock, etc., while
support of a massive number of devices. Example use cases high-end use cases requiring better QoS can be supported by
include fleet management, asset tracking, smart meters, smart energy harvesting technology with active transmission. Both
city, gateways, sensors, voice, and point-of-sales. LTE for Ma- backscattering and active energy-harvesting devices can be
chines (LTE-M) and Narrowband IoT (NB-IoT) technologies battery-less or carry a small battery.
fulfill the requirements for these applications [9]. Fig. 2 shows an overview of broad coverage of A-IoT
Ambient IoT: The last class is the latest addition and technology. Backscattering IoT devices can be used both in
comprises the very low end of IoT use cases where the monostatic and bi-/multi-static configurations, where the term
requirements include ultra-low complexity devices, ultra-low monostatic is used when both the energy exciter and the reader
 3

Fig. 2. Ambient IoT landscape with both energy harvesting and backscattering devices. Ambient energy sources like sun, TV signals and Wifi signals can
be used for energy harvesting and excitation as well.

functionalities are performed by the same device, while bistatic The RAN plenary later agreed to study A-IoT radio use
deployment specifies the scenario where the energy exciter cases and requirements in [11]. This study has completed in
and the reader are physically two different devices. Fig. 2 3GPP and a summary of main agreements is captured in this
illustrates that 3GPP base stations (BSs, i.e., gNB or dedicated section.
micro cells) and smartphones can be used both as exciters, and 3GPP RAN has agreed to study four deployment topologies
readers, or RF energy source, while an ambient energy source for A-IoT devices:
(e.g., TV, WiFi signals) can also be used for excitation for 1) BS ↔ A-IoT device
backscattering as well as energy source for energy harvesting 2) BS ↔ Intermediate node ↔ A-IoT device
for active transmissions. Besides RF signals, energy harvesting 3) BS ↔ Assisting node UE ↔ A-IoT device ↔ BS
devices can make use of other ambient energy sources like sun 4) UE ↔ A-IoT device
light, pressure, thermal etc, to harvest and store energy and In Topology 1, the A-IoT device directly and bidirectionally
use it to make data transmission. Though 3GPP focus is more communicates with a base station. In Topology 2, the A-
on RF energy harvesting devices in A-IoT studies, devices IoT device communicates bidirectionally with an interme-
powered by other energy sources, e.g., solar which has several diate node between the device and the base station. The
orders of magnitude high power density as compared to RF intermediate node can be a relay, Integrated Access and
source, are not precluded in this study.” Backhaul (IAB) node, user equipment (UE), repeater, etc.,
which can support A-IoT. In Topology 3, the A-IoT de-
vice transmits data/signaling to a base station and receives
B. 3GPP Focus Areas and Use Cases
data/signaling from the assisting node; or the A-IoT device
3GPP has recently studied A-IoT in both Service and receives data/signaling from a base station and transmits
System Aspects working group 1 (SA1) [8] and Radio Ac- data/signaling to the assisting node. In this topology, the
cess Networks (RAN) plenary [11]. The 3GPP work was assisting node can be a relay, IAB, UE, repeater, etc. which can
initiated by the Technical Report in TR 22.840 [8] by SA1 support A-IoT. In Topology 4, the A-IoT device communicates
to capture use cases, traffic scenarios, device constraints of bidirectionally with a UE. 3GPP SA1 has defined a large set
ambient power enabled IoT; and identify new potential service of use cases for ambient power-enabled IoT and 3GPP RAN is
requirements as well as new KPIs. defining representative deployment scenarios for studies, each
Considering the limited size and complexity affordable by covering more use cases and topologies.
practical applications for battery-less devices with no energy Using bi-/multi-static links can enable positioning and will
storage capability or devices with limited energy storage that remove the challenges associated with full duplexing self-
do not need to be replaced or recharged manually, the output interference for a monostatic A-IoT illuminator and reader
power of the energy harvester typically ranges from 1 µW device.
to a few hundreds of µW. Existing cellular devices may not The 3GPP RAN study assumes three A-IoT device types:
work well with energy harvesting due to their peak power • Device A: (Passive) Pure battery-less devices with no
consumption of higher than 10 mW. energy storage capability at all, no independent signal
 4

• Indoor: 3m at 90%
• Outdoor: several 10m at 90%
Based on these simple design targets the ambition is to
standardize radio interface and protocols to enable deployment
of such Ambient IoT devices. 3GPP is expected to study A-
IoT devices in Rel. 19 with the aim to standardize in later
releases. It is largely expected that AIoT technology will be
one of the key bridge topics in 3GPP for 6G.

IV. F EASIBILITY S TUDIES

As 3GPP studies are heavily focused on backscattering A-
Fig. 3. Design targets for A-IoT devices as discussed in 3GPP. IoT devices, we discuss some insights on range coverage
aspects for Devices A and B in this section. Through numerical
evaluation on link budget for both forward and reverse links,
generation/amplification (i.e., capable of only backscat- we show which topologies are more practical, when taking
tering), and completely dependent on the availability of into account the device type, transmit power level, and fixed
an external source of energy. vs mobile deployment. The received power Prx,tag at the A-
• Device B: (Semi-Passive) Devices with limited energy IoT device from a given transmitter is computed by product
storage capability that do not need to be replaced or of two factors A and B such that Prx,tag = AB, where
recharged manually, no independent signal generation but A = PT GT /dγ1 and B = Gtag (λ/4π)2 [12]. Prx,tag is the
backscattering potentially with reflection gain. received power at a tag, PT is output power from a transmitter,
• Device C (Active) Actively transmitting device with lim- GT is transmit antenna gain, Gtag is the tag antenna gain, λ
ited energy storage capabilities based on ambient energy is the carrier wavelength, d1 is the transmitter to tag distance
sources. and γ is the path loss exponent (γ = 2 for free space,
γ = 3 for fading environment). Similarly, at the reader,
C. Design Targets for Cellular AIoT Devices the received backscatter signal power Prx,r is computed by
Prx,r = AB 2 GR M/dγ2 , where GR is the receive antenna gain,
The main design target pillars in focus for 3GPP A-IoT
M is the backscatter tag modulation factor, and d2 is the tag to
are depicted in Fig. 3. The target is low data rate, ultra-low
receiver distance. Note that the above formula is generalized
cost, ultra-low-power devices at small form factor but with
and applies to all backscatter deployment scenarios. For a
improved range compared to Radio Frequency Identification
monostatic case, it is imperative that d1 = d2 and GT may
(RFID) and with positioning enabled at relaxed accuracy.
be equal to GR . For a bistatic case where the transmitter and
The 3GPP RAN study assumes some high-level design
receiver use equal antenna gain (e.g., two access points) the
targets for the A-IoT devices as outlined below.
formula can be simplified by setting GT = GR . For a bistatic
Device complexity:
case where different equipment is used for transmitting and
• For Device A, the complexity target is to be comparable
receiving (e.g. a UE and a BS), all parameters can be set
to ultrahigh frequency (UHF) RFID ISO18000-6C (EPC individually.
C1G2).
Considering a tag performing binary modulation with a bit
• For Device B, the target is such that: Device A complexity
rate of 10 kbps and a target (uncoded) bit error rate (BER) =
< Device B complexity < Device C complexity.
1%, the required backscatter signal to noise ratio (SNR) at the
• For Device C, the complexity target is to be orders-of-
Rx is SNRmin = 4.3 dB. Assuming a 15 kHz signal bandwidth
magnitude lower than NB-IoT.
W , a receiver noise figure N F = 6 dB, a shadowing/fade
Device power consumption: margin of F = 10 dB, the reader reference sensitivity has
• For Device A, the power consumption target during a value of approximately S = −112 dBm (i.e., using S =
transmitting/receiving is ≤ 1 µW. −174 + N F + F + 10 log10 W + SNRmin ).
• For Device B, the target during transmitting/receiving is For battery-less devices that harvest RF energy (Device A),
such that: two conditions must be met for successful communication:
– Device A power consumption << Device B power 1) Received excitation power at the tag Ptag must exceed
consumption < Device C power consumption; or a certain power-up threshold Pthr in order for the tag
– Device A power consumption ≤ Device B power to ”wake up”. In the following calculations, a threshold
consumption < Device C power consumption. of Pthr = −19 dBm is assumed (common value for
• For Device C, the device power consumption is ≤ 1 mW. passive RFID devices). This typically governs the ”max
Device data rates: operating range”.
Maximum supported data rate not less than 5 kbps and 2) Received backscattered power at the receiver must ex-
minimum supported data rate not less than 0.1 kbps. ceed the reader sensitivity requirement S (S = -112 dBm
Positioning accuracy (SA1 study based): in the following calculations).
 5

Fig. 4. UE illumination scenarios with different receive options for device type A and B.

Three different bistatic use cases with UE excitation are Device B can be used, and the uplink range from the device to
depicted in Fig. 4 with some example communication ranges. a base station can reach 300 meters, which can accommodate
The path loss exponent γ is assumed to be 3 for outdoor cases similar inter-site distances (ISD).
and 2.5 for indoor cases. For passive A-IoT devices (Device Considering the above link budgets, the topology of illumi-
A), the modulation factor is M =0.25 and for semi-passive nator and receiver to be deployed will depend on application
A-IoT devices (Device B) M =1. Note that all devices are and the environment. However, it is clear that for large scale
assumed to have an omni-directional antenna (Gtag = 2 dBi), outdoor or enterprise deployments, UE illumination will be
which represents a baseline scenario. In the first case from favorable, since the power source can be brought close to the
the left, excitation of devices is performed in an area by a AIoT device. Then uplink ranges of several hundred meters
UE and reception is performed by a fixed BS. The middle can be achieved, and the AIoT device backscattering can be
case involves indoor small-cell deployment with excitation of received by fixed infrastructure gNBs.
devices is performed by a UE and reception by a small-cell
access point. The case to the right involves indoor small-cell
deployment with device excitation and reading by two different V. A-I OT: TECHNOLOGY CHALLENGES AND
UEs. SOLUTIONS

It is interesting to note that for bistatic cases, the two links’ A. A-IoT Device Access Techniques
(exciter (Tx)-to-device and device-to-receiver (Rx)) maximum Energy harvesting IoT devices have the following key
range depends on each other. For a short Tx-device distance, a difference as compared to NR UE and associated challenges
significantly higher device-Rx distance can be achieved. This must be resolved before deploying A-IoT technology.
has also been shown experimentally in [13]. The opposite
1) New Terminal States: A major consideration for the
is also true for semi-passive devices (Device B), but not
access techniques in energy harvesting communication
for passive A-IoT devices that need to be energized by the
is the uncertainty in the availability of energy. Unlike
exciter. Fig. 5(a,b) shows this contrast for the UE excitation-
conventional communication scenarios, availability of
gNB reception case, where a yellow shaded region corre-
ambient energy is not stable, and it varies spatially as
sponds to successful communication operation. The Figure
well. When designing access techniques, this uncertainty
5(a) corresponds to an A-IoT device (Device A), for which
has to be considered [14]. For energy harvesting devices
the Tx-device distance is very limited, since the device power-
with active transmission (Device C), duty cycle design,
up threshold must be exceeded. Despite that fact, very long
scheduling decisions and wakeup occasions need to be
device-Rx ranges can be achieved (several hundreds of meters)
designed based on availability of intermittent energy.
for a sufficiently sensitive receiver. Fig. 5(b) corresponds to an
As devices are not always connected with the network,
energy-assisted backscattering device (Device B). Since there
NR functionality based on radio resource control (RRC)
is no RF power-up constraint, the total operating region is
states (active/idle) may not be valid and we may need
larger compared to Device A. Keeping a short Tx-device range
new device state definitions.
allows for exponentially larger device-Rx range and vice versa.
2) Device Registration: NR UEs are registered with the
Similar trends are observed for the case of UE exciter and
network and a context is maintained in the network. As
indoor gNB reception (Fig. 5(c) and Fig. 5(d)).
A-IoT devices are low-cost devices with small form fac-
Fig. 4 provides example numbers for the feasibility of each tor, it will be difficult to register them with the network
deployment (RF harvesting devices (Device A) and energy- through subscriber identity module (SIM). However, it is
assisted devices (Device B)). In general, Device B can afford important to establish a simplified form of A-IoT device
longer Tx-device ranges than Device A, due to the relaxed identification in a 3GPP network and 3GPP SA2 studies
power-up requirements, and therefore can be used for larger will look into this aspect. It is not yet clear how A-
area coverage. For UE excitation cases, where the UE can be IoT device identification will be managed, i.e., through
brought to the proximity of the devices, both Device A and Subscription Permanent Identifier (SUPI) by the network
 6

Fig. 5. Operating regions (Yellow indicates sufficient received SNR) (a) for UE-device-gNB scenario device A operation (left) is governed by power-up
threshold (b) for UE-device-gNB scenario device B operation (right) is receiver-sensitivity limited (c) for UE-device-indoor gNB scenario device A operation
(left) is governed by power-up threshold (d) for UE-device-indoor gNB scenario device B operation (right) is receiver-sensitivity limited.

or through application-defined IDs managed by a 3GPP backscattering link budget analysis but illuminator-to-
network. backscattering device analysis as well.
3) Mobility Tracking: Typical use cases targeted by AIoT 5) Distributed Energy Sources: As shown in Section IV,
technology, e.g., inventory tracking, may not require due to a large link budget requirement, gNB is not
continuous mobility management as in NR. Mobility the best exciter for backscattering devices. In 3GPP
management requires maintaining context of UE in networks, all the NR UE access mechanisms are handled
the network and preparing handover when UEs move by gNB. For backscattering devices, we need to rely on
between cells. As some of the target use cases do not distributed low-complexity 3GPP network illuminators,
require this functionality, this overhead can be avoided readers and/or smartphones, which require more coordi-
by finding solutions for ’on demand’ mobility manage- nation between network devices to service A-IoT devices
ment. and demand for new network protocol design.
4) Link Budget Analysis: Backscattering devices are low- 6) Simplified Protocols: Due to small form factor and
cost devices and typically first harvest some energy low-cost requirement, A-IoT devices cannot support
to activate their circuit and then backscatter informa- full stack access protocol. Thus, simplified protocol
tion. Thus, 3GPP system design not only involves design is a key requirement for such devices, particularly
 7

(RTT) methods pose an even more challenging task, since

the A-IoT device must, in addition to processing downlink,
also reply with a paired uplink transmission. Given the above,
the only feasible alternative remains uplink positioning, where
the Transmission/Reception Points (TRPs) are tasked with
detecting the device and measuring the required positioning
metrics, conditioned on the availability and proximity of radios
to charge and activate the A-IoT device.
In the second use-case, the A-IoT device may itself be
used to locate other devices (other A-IoT devices or NR
UEs). Specifically, NR localization of a UE requires signaling
from multiple TRPs with known location, to multilaterate
the UE location. When an insufficient number of TRPs are
available and/or in poor line-of-sight (e.g., very often in indoor
scenarios), cheap A-IoT devices may take the role of the TRP,
Fig. 6. AIoT positioning incorporating various network entities. UE to UE and become the so-called A-IoT TRP. However, the location
links (PC5 interface) and UE to gNB links (Uu interface) have been marked. of an A-IoT TRP:
1) is typically unknown to the NR network right after
deployment. That is because such devices are manually
backscattering devices which depend on strong activa-
installed at physical locations associated with areas
tion signals. Access and security mechanisms need a
which have positioning blind spots.
new design approach that provides required authentica-
2) may change over time. The devices may be manually
tion and security to backscattering devices at lower com-
moved around to cover other/new positioning blind spots
plexity and energy cost. Access and security protocol
e.g., where such blind spots are created when an existing
complexity may depend on each device type discussed
TRP gets obstructed due to temporary blockage. For
in 3GPP.
example, a TRP can be blocked in an indoor factory
when large packages are moved around the factory or
B. A-IoT Device Positioning Techniques new equipment is installed.
For several 3GPP A-IoT deployment scenarios and use 3) cannot be computed using NR positioning techniques
cases, support for positioning is a key feature. Fig. 6 shows due to the limited range of such devices. While it has
an example of an A-IoT deployment based on multi-static high cost-reduction potential, it is not straightforward
backscattering with positioning capabilities. The positioning of how to efficiently utilize an A-IoT device as a TRP, since
A-IoT devices can be based on the existing 3GPP positioning the current NR positioning protocols are not equipped
architecture, using a 3GPP node to trigger the event and one to compute and track the position of the A-IoT device.
or more 3GPP nodes to listen to a given reference signal,
e.g., another gNB or Positioning Reference Unit (PRU). The C. A-IoT Authentication and Security
positioning session can be controlled and monitored by the Similar to RFID, security methods for A-IoT systems must
location management function (LMF). A positioning session guarantee at least (a) authentication, i.e., confirming device
can be enabled by a UE in the network or from the network identity and (b) confidentiality, i.e., ensure that the A-IoT
side. Sidelink can be another means to configure UEs in device is not being eavesdropped on. NR security mechanisms
proximity of the A-IoT device as activation exciter and/or as guarantee both requirements, data integrity and availability via
positioning anchors. higher layer protocols and complex cryptography. The com-
A-IoT positioning encompasses two complementary use- plexity of NR security is however both a blessing (for existing
cases: 1) locate the A-IoT device and 2) use the A-IoT device services) and a curse for A-IoT systems. A-IoT systems consist
to locate other devices. of devices that are extremely limited in terms of computational
In the first use case, the A-IoT device is often used for power, that can rarely support basic cryptography. Therefore,
tracking objects in production, shipping, etc., and therefore, it there is a need to develop new security protocols and methods
is necessary for the NR network to locate the A-IoT device. tailored to A-IoT design limitations. The above-mentioned
However, since the A-IoT activation and reading ranges and protocols may instead resort to physical layer security mech-
device processing capabilities are severely limited, the typical anisms. Physical layer security is built on exploiting the
time- and angle-based NR positioning techniques cannot be uniqueness of the propagation channel responses between the
straightforwardly applied. Specifically, A-IoT systems do not A-IoT device and other legitimate radios comprising the A-
lend themselves well to downlink positioning techniques, since IoT system and determining a transmission scheme tailored to
these requires the A-IoT device to be synchronized to the those specific conditions only.
network and be able to detect positioning signals and extract
relevant measurements (differential timing measurements, an- D. Spectrum
gle of departures, carrier phases, etc.), requirements which are The spectrum to be targeted for A-IoT should be common
impossible to fulfill by most A-IoT devices. Round-trip time across as many regions as possible and with the same or
 8

similar spectrum access regulations, both to ensure the ease [9] R. Ratasuk, N. Mangalvedhe, Y. Zhang, M. Robert, and J.-P. Koskinen,
of mobility of AIoT devices across different regions as well “Overview of narrowband iot in lte rel-13,” in IEEE Conference on
Standards for Communications and Networking (CSCN), 2016.
as to simplify the standardization of the A-IoT ecosystem. For [10] “5G reduced capability devices,” Nokia, Tech. Rep., May 2022.
example, RFID technology can be deployed across different [Online]. Available: https://onestore.nokia.com/asset/212488? ga=2.
bands, where the actual band depends on the use case and 145096240.545018478.1690233441-1471620813.1690233441
[11] 3rd Generation Partnership Project (3GPP), TR 38.848,
the requirements [15]. However, in order to avoid coexistence “Ambient IoT (internet of things) in RAN,” Tech. Rep.,
issues with legacy RFID technologies and at the same time 2023. [Online]. Available: https://portal.3gpp.org/desktopmodules/
take advantage of the existing cellular deployments, the A- Specifications/SpecificationDetails.aspx?specificationId=4146
[12] J. D. Griffin and G. D. Durgin, “Complete link budgets for backscatter-
IoT system should target licensed frequency bands where radio and RFID systems,” IEEE Antennas and Propagation Magazine,
cellular technologies already operate as well as the unlicensed vol. 51, no. 2, pp. 11–25, 2009.
bands at 5 GHz and 6 GHz. In particular, the unlicensed [13] J. Kimionis, A. Bletsas, and J. N. Sahalos, “Increased range bistatic
scatter radio,” IEEE Transactions on Communications, vol. 62, no. 3,
bands will have both infrastructure-based and sidelink-based pp. 1091–1104, 2014.
access technologies allowing for the coordination of the A- [14] M. M. Butt, N. Mangalvedhe, and R. Ratasuk, “Energy harvesting arrival
IoT communications and aid RF-based charging of the A-IoT aware joint sensing and transmission,” US Patent, 11,540,223,2022,
Dec., 2022.
devices. [15] N. C. Wu, M. A. Nystrom, T. R. Lin, and H. C. Yu, “Challenges to global
RFID adoption,” in Technology Management for the Global Future -
PICMET Conference, vol. 2, 2006, pp. 618–623.
VI. C ONCLUSION
This article provides a comprehensive overview of Ambient
IoT technology in 5G. Ambient IoT is new technology on top
of existing cellular IoT technology and targets market segment
below the existing technologies. Battery replenishment and
cost is major hurdle in deployment of IoT technology and
A-IoT can help fill this gap by targeting reduced complexity
and cost. Before this technology becomes a reality, several
challenges need to be overcome including development of
suitable access, transmission, positioning, and radio resource
management techniques, identifying spectrum and topologies
for various scenarios, and incorporating low complexity secu-
rity protocols. This article summarizes 3GPP discussions on A-
IoT studies and envision an ambient IoT technology that will
make future networks more sustainable and ready for massive
deployment of IoT devices.

R EFERENCES
[1] “Market research on energy harvesting for IoT device,”
Conflow Power group, London, Tech. Rep., 2020. [Online].
Available: https://assets.website-files.com/6225ec4d016842cb92d1e5d3/
629151680aab9776a55aaad2 Conflow-Energy-Harvesting-Report.pdf
[2] S. Naser, L. Bariah, S. Muhaidatand, and E. Basar, “Zero-energy
devices empowered 6G networks: Opportunities, key technologies, and
challenges,” IEEE internet of things magazine, 2023.
[3] D. Ma, G. Lan, M. Hassan, W. Hu, and S. K. Das, “Sensing, computing,
and communications for energy harvesting IoTs: A survey,” IEEE
Communications Surveys & Tutorials, vol. 22, no. 2, pp. 1222–1250,
2020.
[4] Z. Wei, X. Yu, D. W. K. Ng, and R. Schober, “Resource allocation for
simultaneous wireless information and power transfer systems: A tutorial
overview,” Proceedings of the IEEE, vol. 110, no. 1, pp. 127–149, 2022.
[5] O. L. A. López, H. Alves, R. D. Souza, S. Montejo-Sánchez, E. M. G.
Fernández, and M. Latva-Aho, “Massive wireless energy transfer: En-
abling sustainable IoT toward 6G era,” IEEE Internet of Things Journal,
vol. 8, no. 11, pp. 8816–8835, 2021.
[6] K. Tekbyk, D. Altinel, M. Cansiz, and G. K. Kurt, “Wireless Power
Transmission on Martian Surface for Zero-Energy Devices,” IEEE
Transactions on Aerospace Electronic Systems, vol. 58, no. 5, pp. 3870–
3880, Oct. 2022.
[7] 3rd Generation Partnership Project (3GPP), RP-223396, “Ambient IoT
(internet of things),” Tech. Rep., 2022. [Online]. Available: https://www.
3gpp.org/ftp/TSG RAN/TSG RAN/TSGR 98e/Docs/RP-223396.zip
[8] 3rd Generation Partnership Project (3GPP), TR 22.840, “Study
on ambient power-enabled internet of things,” Tech. Rep., 2023.
[Online]. Available: https://www.3gpp.org/ftp/Specs/archive/22 series/
22.840/22840-120.zip