IOT Based Poultry Farm

IOT Based Poultry Farm — Topics for IoT Students

Ambient IoT: Communications Enabling Precision

Agriculture Ashwin Natraj Arun, Byunghyun Lee, Fabio A. Castiblanco, Dennis R. Buckmaster, Chih-Chun Wang, David J.

Love, James V. Krogmeier, M. Majid Butt and Amitava Ghosh.

Abstract—One of the most intriguing 6G vertical markets is LoRaWAN, facilitate communication for IoT devices. How- precision agriculture, where communications, sensing, control, ever, supporting a high density of IoT devices in rural areas and robotics technologies are used to improve agricultural arXiv:2409.12281v1 [eess.SP] 18 Sep 2024

remains challenging with the current wireless infrastructure outputs and decrease environmental impact. Ambient IoT (A- IoT), which uses a network of devices that harvest ambient energy [2]. Precision agriculture relies on sensor data, GPS-guided to enable communications, is expected to play an important role machinery, and variable rate technology for optimal crop in agricultural use cases due to its low costs, simplicity, and management.

This approach enhances productivity, reduces battery-free (or battery-assisted) operation. In this paper, we resource waste, and supports sustainable farming through data- review the use cases of precision agriculture and discuss the driven decisions. However, achieving this promise requires challenges.

We discuss how A-IoT can be used for precision agriculture and compare it with other ambient energy source sensor deployment at high densities and commonly avail- technologies. We also discuss research directions related to both able communications, such as Wi-Fi, are inadequate for vast A-IoT and precision agriculture. farm areas. Instead, energy-harvesting devices using ambient Index Terms—Ambient IoT, backscatter communication, pre- sources for power and communication are needed to meet cision agriculture, 3GPP, RFID. these requirements.

Devices that harvest energy from ambient sources such as electromagnetic, solar, and thermal to power and communicate and operate within the 3GPP network are I. I NTRODUCTION known as Ambient IoT (A-IoT) devices. A-IoT devices can In recent decades, agriculture has undergone a paradigm operate with or without a battery and do not require a constant shift driven by technological advancements to meet the de- power supply, relying on backscattering to communicate. mands of a growing population.

This evolution has given rise Backscatter communications have been around for decades to precision agriculture, an innovation that leverages cutting- and are used in technologies such as Radio Frequency Identifi- edge technologies to optimize resource utilization, enhance cation (RFID) [3] and Wi-Fi backscatter [4]. However, there is crop yields, and mitigate environmental impacts. Precision little prior work on dense deployments and operations within agriculture aims to integrate data-driven decision-making, au- the existing cellular architecture.

Recently, 3GPP has started tomation, and advanced sensor technologies to create a more discussion on A-IoT devices with potential use cases, relevant efficient and sustainable agricultural system. communication scenarios and topologies of operation for such Concurrently, many economic sectors have been signifi- devices. While there is research looking at the performance cantly changed by the rapid growth of Internet of Things of A-IoT devices in conventional use cases, there is very little (IoT) devices which offer simplification and lower costs in literature on the application of A-IoT devices in the domain a wide range of applications.

This growth is expected to on precision agriculture. We review the feasibility of A-IoT continue, with potential deployments in agriculture, healthcare devices for precision agriculture with a link-budget analysis and manufacturing markets [1]. In agriculture, the advent of and compare A-IoT to RFID devices.

Also, we look at the IoT has enabled improved systems for sensing data, communi- challenges and potential research directions for A-IoT devices. cating, inferring information and making decisions. Currently, In this work, we aim to answer several fundamental questions standards such as Narrowband IoT (NB-IoT) and enhanced relating precision agriculture and A-IoT: machine-type communication (eMTC), which are based on • What are different use cases of A-IoT in precision agri- the 3rd Generation Partnership Project (3GPP), as well as culture? • What is backscattering and how does A-IoT interface with This work is supported in part by the National Science Foundation (NSF) under grants CNS2212565, EEC1941529, CCF2008527, CNS2107363, existing 3GPP architecture?

CNS2235134, CNS2225578, CCF2309887 and Nokia. • What are the challenges and potential research directions Ashwin Natraj Arun, Byunghyun Lee, Fabio A. Castiblanco, Chih-Chun for A-IoT in precision agriculture? Wang, David J.

Love and James V. Krogmeier are with the Elmore Family School of Electrical and Computer Engineering, Purdue University, West Lafayette, IN 47907, USA (e-mails: {ashwin97, lee4093, fcastibl, chihw, II. P RECISION AGRICULTURE WITH A MBIENT I OT djlove, jvk}@purdue.edu).

Dennis R. Buckmaster is with the School of Agricultural and Biological Precision agriculture uses sensors, GPS, drones, and other Engineering, Purdue University, West Lafayette, IN 47907, USA (e-mail: technologies to optimize crop management to enhance effi- dbuckmas@purdue.edu) M. Majid Butt and Amitava Ghosh are with Nokia Standards, USA (e-mails: ciency, increase yields, and minimize environmental impact. {majid.butt, amitava.ghosh}@nokia.com).

Fig. 1 illustrates a landscape of use cases where precision agriculture benefits significantly from A-IoT devices.

Figure 1. Landscape of A-IoT devices used in Precision Agriculture.

A. Sensing in Precision Agriculture need for dense sensor deployment, it is essential to use devices A-IoT devices can be used in precision agriculture to that are small, low-cost, power-efficient, and biodegradable. monitor environmental conditions in farms and greenhouses. A-IoT devices are ideal for this use case due to these charac- These devices can measure parameters such as air temperature, teristics.

Communication from A-IoT devices to the network humidity, CO2 concentration, light, soil temperature, and pH. can be facilitated using a mobile relay node on farm machinery By monitoring these parameters, predictions can be made such as tractors, sprayers and harvesters. As these machines about soil health, crop yield, pest management etc. A-IoT move across the field, the A-IoT devices become active and devices are ideal for this purpose because they do not require send data to the node mounted on the machine. an external power source and are compact.

This allows farmers In corn fields, there is a notable distinction between surface to deploy a large number of sensors in fields to monitor both soil moisture and the moisture content deeper underground. crops and soil efficiently. Sensors can be placed in open fields, Corn plants develop deep roots to access water, emphasizing greenhouses with A-IoT devices connected to the network the need for soil moisture monitoring at multiple depths. through various discussed topologies.

Unlike conventional sensors that require periodic removal for data retrieval, underground A-IoT devices offer the advantage of data collection without physical intervention, which is B. Actuation and Controllers in Precision Agriculture particularly beneficial for crops like corn. In agricultural production, various factors such as soil, climate, and water significantly affect crop growth and D.

Smart Livestock Management yield. Agricultural equipment, like pesticide sprayers, fertilizer Smart livestock farming employs innovative production spreaders and irrigation systems, can be actuated or controlled systems to enhance sustainability and reduce food waste periodically based on sensor data to ensure optimal yield, and crises. Body temperature is a vital health indicator for and effective crop management.

Providing a continuous power livestock and is crucial for farmers to identify and act on supply for these controllers in outdoor farms is a challenge. potential diseases early. Changes in body temperature signify A-IoT enables seamless communication for these control and illness, making it a precision health indicator. Importantly, actuation devices by harvesting ambient energy and triggering acquiring body temperature data is not latency-intensive, and the controllers.

This ensures reliable data transmission and the increasing herd size necessitates the use of cost-effective remote control capabilities, even in areas with limited infras- ear tags over high-power IoT devices. A-IoT devices are ideal tructure. Since the controllers operate periodically, the A-IoT for this application due to their low cost, small form factor precision agricultural controllers can be activated periodically and battery-free operation.

Cattle and pigs can be equipped by the farm management system. This integration of A-IoT with small ear tags that monitor their body temperature and with the cellular architecture enhances the responsiveness and transmit data to the farm. efficiency of agricultural operations. E.

Food Supply Chain C. Underground Soil Sensing in Precision Agriculture Innovative solutions are essential to tackle food waste and Existing sensors used in precision agriculture are mostly ensure food safety. A controlled environment is crucial for above ground or have antennas above ground.

To accurately fresh foods, such as vegetables and meat, to maintain their measure nitrogen levels, soil moisture, and other important safety and shelf life. An efficient method is needed to monitor metrics, sensors need to be placed underground. Given the every stage of the food supply chain to reduce food waste.

The increasing demand for organically sourced food items also of approximately a decade. The distinctive features of A-IoT requires a reliable way to guarantee production and processing devices further enhance their appeal for various applications. methods in the food supply chain. Equipping the food supply These characteristics sharply contrast with existing low power chain with A-IoT devices in each transport item can address wide area network (LPWAN) technologies like LoRaWAN, these needs.

These devices update relevant information at NB-IoT, and eMTC, as seen in Table I. In essence, A-IoT each stage, from the seed stage to harvesting and packaging leverages backscattering ambient energy sources, targeting a facilities. The end user can verify the produce by accessing different class of devices compared to existing IoT devices. the data from the A-IoT device.

Similarly, suppliers can use these devices to track real-time demand and stock accordingly, Table I reducing food waste. This principle applies at each stage of C OMPARISON OF I OT NETWORK ARCHITECTURES the food supply chain, decreasing overstocking and waste. Network Max Coverage Data Rates Architecture Power III.

BACKSCATTER C OMMUNICATIONS LoRaWAN 25mW 10-15km (rural) 0.3-5.5 kbps NB-IoT 200mW 5-15km (rural) 250 kbps A-IoT devices use backscatter communication, a wireless Active A-IoT 10mW 500 m 5 kbps technique where devices reflect and modulate an incoming Battery-free A-IoT 10µW 500 m 5 kbps radio frequency (RF) signal to transmit data [5]. This method is particularly useful for devices with limited or zero internal B. A-IoT with 3GPP power, as it does not require the generation of new RF signals. 3GPP has recently started discussions about A-IoT devices There are two types of backscatter communication: monostatic and has included them in technical reports (TRs) 38.848 [6] and bistatic as seen in Fig. 1. and 22.840 [7].

The specifications of A-IoT in Rel-19 study In a monostatic system, the energy exciter and signal item are summarized in [8]. receiver are the same device, whereas in a bistatic system, they The three types of A-IoT devices are battery-less (BL), are separate devices. At the heart of backscatter communica- battery-assisted (BA), and battery and signal-assisted (BSA) tion lies impedance mismatching. Varying the load impedance devices.

BL devices have no energy source and no independent of an A-IoT device changes its reflection coefficient, affecting signal generation/amplification, relying solely on backscatter the amplitude and phase of the reflected wave. This allows communication. BA devices have an energy source for am- the A-IoT device to perform various modulation techniques, plifying the backscattered signal but no independent signal such as amplitude-shift keying (ASK), amplitude modulation generation.

BSA devices have an energy source and inde- (AM), phase-shift keying (PSK), phase modulation (PM), and pendent signal generation, using active RF components for their combinations via load modulation [5]. The modulation transmission. The major specifications of these devices are order is proportional to the number of load impedance states listed in Table II. in the A-IoT device.

A-IoT devices are designed to support indoor environments Backscatter communication is widely used in technologies with coverage ranging from 10 to 50 meters and outdoor such as RFID, Wi-Fi backscatter, and Bluetooth Low Energy environments from 50 to 500 meters. The data rate for uplink (BLE) backscatter. In RFID, a dedicated RF transceiver emits and downlink transmissions ranges from 0.1 kbps to 5 kbps.

RF energy towards RFID tags, which reflect the RF signal Each device can handle message sizes up to 1000 bits for back to the transceiver, modulated with data. This approach reception and transmission. A-IoT devices support up to 150 is efficient for inventory management, asset tracking, and devices per 100 square meters indoors and up to 20 devices access control.

While RFID systems are useful, they require per 100 square meters outdoors. dedicated transceivers to enable backscattering for the RFID 3GPP has defined the following network topologies for A- tags. Further, the reader has to be in close proximity to RFID IoT devices to connect to the network via base stations (BSs) tags for backscattering, which is a key limitation of RFID and user equipments (UEs). In these topologies, the links technology.

In contrast, A-IoT leverages existing ambient RF may be unidirectional or bidirectional. The following main (cellular or TV) signals for operation, distinguishing it from topologies are considered in recent 3GPP discussions: traditional RFID systems. This adaptability positions A-IoT 1) BS ↔ A-IoT device: Direct, bidirectional communica- as an exciting research area that aligns well with established tion. communication infrastructures. 2) BS ↔ intermediate node ↔ A-IoT device: Bidirectional communication between A-IoT device and the interme- IV.

A MBIENT I OT diate node (e.g., relay, integrated access and backhaul Ambient IoT devices include both active devices with (IAB) node, UE, repeater). energy harvesting and passive devices with backscattering. These devices can be used in both monostatic and bistatic V. F EASIBILITY S TUDY - U NDERGROUND configurations depending on the requirements of the use cases.

BACKSCATTERING In this section, we explore the use of backscattering tech- A. Comparison with existing IoT architectures nology for underground soil sensing in precision agriculture. A-IoT devices are characterized by ultra-low complexity, This feasibility study examines the potential of backscattering compact size, limited capabilities, and an extended lifespan communication for underground sensing, primarily focusing

Table II

P ROPOSED SPECIFICATIONS FOR A-I OT DEVICES

Device Type Description Power Consumption Complexity

Battery-less (BL) No energy source. Only backscatter ≤ 10µW Comparable to UHF RFID communication. ISO18000-6C (EPC C1G2) Battery-assisted (BA) Energy source for amplifying the Between BL and BSA devices Between BL and BSA devices backscattered signal.

No independent signal generation. Battery and signal-assisted (BSA) Has an energy source. Can indepen- ≤ 10mW Much lower than NB-IoT de- dently generate signal. vices

The next challenge after deployment of these sensors is

reading the data from these sensors. Unlike traditional sensors that can automatically transmit and receive data, backscat- tering devices need excitation to activate these sensors for communication. Currently, RFID devices are excited manually for devices deployed in a large area using RFID readers.

This is an arduous process to implement for a large farm to read the sensor data. Unlike the wireless channel across air, the wireless channel across soil is harsh. Apart from distance- based loss, there are losses from soil moisture, refraction etc.

Thus, reading RFID tags planted underground is much more difficult compared to terrestrial use cases. In our field trials, we automated the reading process using OATSMobile, a communications platform enabling connected Figure 2. RFID tags the size of corn seeds planted in the field with a farms.

OATSMobile, a customized agricultural sprayer (Fig. conventional seed planter. 3), has six RFID antennas mounted on the machine and an RFID reader (Zebra FX9600) in its cabinet. The antennas on A-IoT devices, by leveraging UHF RFID field trials. Given were strategically placed to optimize tag detection, with two the similar device complexities between BL A-IoT and UHF antennas directed towards each row of corn where the RFID RFID devices, we conducted underground field trials using tags were planted.

As OATSMobile moves through the farm, RFID devices due to their commercial availability. Although it positions the antennas between the rows of corn. In a our primary focus is on BL A-IoT devices, the comparable single pass, it covers four rows, reducing obstruction between technology in RFID allows us to draw relevant insights.

We antennas and plants. The experiment was conducted at dif- also perform a link budget analysis to compare the under- ferent corn growth stages to observe seasonal effects. Out of ground read ranges of BL A-IoT and RFID technologies. 288 planted RFID tags, 152 unique tags were successfully read, resulting in a 53.9% reception success rate.

The success A. Field Trials - Backscatter Communication for Underground rate was influenced by factors such as soil moisture, corn Sensing canopy growth, and tag depth. The link-budget analysis in the To assess the feasibility of A-IoT devices for underground following section corroborates the impact of soil moisture on sensing, a high-density deployment of sensors is necessary. the read range of the backscattering devices.

Deploying a large number of sensors in agricultural settings presents several challenges. Currently, sensors are manually buried underground, a method that is labor-intensive and time- B. Underground Backscattering Link Budget Analysis consuming, especially for large farms.

Additionally, the de- We conduct a link budget analysis for both excitation ployment process must minimize disruption to existing crops. (downlink) and backscattered (uplink) connections of A-IoT The high density required for effective monitoring necessitates and RFID devices placed underground to understand our field innovative deployment practices to reduce labor and time. trials. This evaluation allows us to compare the read ranges In our field trials at Purdue University, we deployed RFID of A-IoT devices with RFID devices.

In our analysis, we tags underground using a conventional seed planter, as shown considered the transmitter and reader above the ground, and in Fig. 2. Using RFID tags comparable in size to corn the tag, which is either underground. seeds, we ensured compatibility with existing seed planters For tags placed underground, the analysis is complex without modifications. This automated deployment method due to soil channel characteristics.

The downlink is an significantly reduces labor and time compared to manual de- aboveground-to-underground (AG2UG) link and the uplink is ployment, offering benefits to farmers without adding overhead an underground-to-aboveground (UG2AG) link. The received for sensor installation. We planted 288 RFID tags in 12 rows tag power is Prx,tag = PT GT Gtag LAG−U G (d1 ) where PT of corn at a depth of 2.5 cm underground and the number of is the transmitter power, GT is the transmit antenna gain, tags were distributed based on the soil moisture levels.

Gtag is the tag antenna gain, LAG−U G (d) is the AG2UG

Figure 3. OATSMobile - communications platform enabling connected farms. OATSMobile is equipped with an RFID reader and six antennas to read RFID tags placed underground alongside corn plants in a field.

path loss and d1 is the distance between the transmitter Figure 4. Activation distance and read distance of A-IoT and RFID devices and the tag [9]. Similarly, the backscattered power received placed underground in both monostatic and bistatic configurations.

In a monostatic configuration, dact = dread while in a bistatic configuration, in the reader is Prx,read = Prx,tag Gtag GR M LU G−AG (d2 ) dact ̸= dread . where GR is the reader antenna gain, M is the backscatter modulation factor, and LU G−AG (d) is the UG2AG path loss. exhibit a higher activation and reading distance than RFID Both LAG−U G (d) and LU G−AG (d) include aboveground path tags, although the activation distance for underground tags is loss, underground path loss, and refractive loss components reduced due to higher soil path loss.

Additionally, the data [10]. The underground path loss depends on factors such as shows a decrease in dact and dread with increasing volumetric volumetric water content and soil clay fraction, which alter water content (VWC), indicating higher path loss in wetter soil permittivity. soil. These findings suggest that A-IoT technology, with its In monostatic cases, d1 = d2 and GT = GR as the lower activation threshold and higher sensitivity, is better exciter and the reader are the same device.

In bistatic cases, suited for robust and reliable subterranean soil sensing appli- d1 ̸= d2 , and GT = GR if the transmitter and reader use cations, enhancing underground communication and precision the same antenna gain, otherwise GT ̸= GR . For successful agriculture techniques. communication, the following conditions must be met: • Received tag power Prx,tag should be exceed the tag ac- tivation threshold Pthr . Only if this condition is met, the VI.

C HALLENGES AND R ESEARCH D IRECTIONS tag wakes up. The distance dact at which Prx,tag = Pthr A. Challenges is called the DL distance or activation distance. • Received backscattered power Prx,read should exceed Ambient IoT faces challenges due to lack of coordination the reader sensitivity S.

The distance dread at which between transmitter and receiver, making it difficult to know Prx,read = S is named the UL distance or read distance. the channel state information (CSI). The ambient RF signal The link budget analysis parameters for RFID and A-IoT can interfere with the receiver, complicating the detection systems are as follows: the transmitted power (PT ) is 30 process. While advanced machine learning techniques can dBm for RFID and 24 dBm for A-IoT.

The tag threshold improve detection and estimation, the simplicity and battery- power (Pthr ) is −10 dBm for RFID and −25 dBm for A-IoT, free nature of A-IoT devices mean that existing wireless algo- with receiver sensitivities (S) of −75 dBm and −100 dBm, rithms cannot be applied directly. Therefore, new techniques respectively. The modulation factors (M ) are 0.33 for OOK are needed to overcome these challenges. in RFID and 0.25 for A-IoT.

Both systems share a path loss 1) Energy Harvesting: Despite the simplicity, backscatter- exponent (γ) of 3, and antenna gains (GT = GR ) of 6 dBi, ing has many challenges in providing reliable communica- with a tag gain (Gtag ) of −1 dB. The transmitter is placed tion for A-IoT devices. Some of the key research problems 0.3 meters aboveground, and the volumetric water content in backscattering comprise energy efficiency, communication (VWC) ranges from 5% to 25%.

Based on these parameters, range, network responsiveness etc. The battery-free and ultra- we calculated activation and read distances for both A-IoT and low power operation of backscattering makes it a good tech- RFID devices and are illustrated in Fig. 4. nique for A-IoT devices. The ambient RF signals are used Our results demonstrate the superior read range capabilities for backscattering and typically RF harvesting efficiency is as of A-IoT devices compared to RFID for tags placed under- low as 18.2% [5].

Research could be directed on improving ground in monostatic and bistatic configurations. A-IoT tags the energy efficiency of backscattering systems for RF signals.

2) Backscatter Signal Detection/Estimation: Unlike tra- their nature, but space-time codes (e.g. Alamouti) can achieve ditional communication systems, in a backscatter scheme, diversity gain without explicit CSI. A novel space-time code transmitting bits ‘1’ and ‘0’ correspond to whether the de- proposed in [14] outperforms the Alamouti code for multi- vice is backscattering or not.

Since A-IoT devices merely antenna tags. backscatter existing RF signals, it becomes challenging for 2) Multiple Access / Random Access: A-IoT devices have the reader to detect and estimate the data as it cannot estimate ultra-low complexity and a large device density. In order the CSI without pilot symbols [11]. Implementing machine to support the large A-IoT device density concurrently, we learning (ML)-based techniques is also challenging because need state-of-the-art multiple access techniques.

Since A-IoT the acceptable error rates for detection are much lower than devices have high device density, random access techniques those in traditional ML applications [12]. Therefore, further are employed to support high device density. For UHF RFID research is needed in the detection and estimation of ambient devices, schemes such as slotted ALOHA, Q-protocol are used backscattering signals, given the lack of coordination between for random access.

Since BL is of the same complexity as the device and the reader. UHF RFID, A-IoT devices could use similar random access 3) Security: Security is one of the major aspects in any techniques. There are research opportunities in finding better network.

In A-IoT systems, the devices are limited in com- multiple access techniques for all devices types in A-IoT. plexity and therefore cannot employ the complex techniques 3) Advanced Modulation Techniques: In backscattering used in 5G-NR. Also, security in A-IoT systems should be communication, the load impedance of the A-IoT device is focused in physical layer as devices have minimal upper layer varied to modulate the reflected signal to the reader.

Currently, components unlike traditional systems. These devices are very simpler modulation schemes such as ASK, PSK are used in easy to eavesdrop on as they are very similar to RFID systems. backscattering. For larger data rates, we need larger number We need to ensure authentication and confidentiality over of load impedance states in order to support higher order physical layer security.

There is a need to develop new security modulation schemes with low complexity. designs and protocols that work in low-complexity devices 4) Security: A-IoT security should guarantee authentication such as A-IoT devices. and confidentiality for every device in the network. Unlike 4) Access for Stateless Devices: The challenge lies in traditional wireless systems, A-IoT devices have lower com- communicating with devices that have low or no stored energy plexity.

Hence, we should focus on physical layer security and cannot always respond to network paging signals. Current and algorithms beyond 5G-NR [15]. This is achievable via cellular protocols rely on paging mechanisms with Radio lightweight authentication protocols such as hash functions etc.

Resource Control (RRC) states, assuming devices are always 5) Positioning: Positioning A-IoT devices is crucial for var- reachable and can initiate paging. However, A-IoT devices ious precision agriculture applications, enabling the tracking lack defined RRC states, making these assumptions invalid of farm assets, livestock, and machinery. However, passive A- and necessitating new access mechanisms.

IoT devices pose challenges since they cannot process received signals, making downlink-based positioning difficult. Instead, localization should be done by the reader, though large inter- B. Research Directions site distances in rural networks can hinder positioning due to A-IoT presents unique challenges, including lack of co- low signal strength.

Moving nodes like tractors and drones ordination between devices and the reader. This section ex- can read backscattered signals and locate A-IoT devices in plores potential avenues to improve connectivity and energy a bistatic/multi-static manner, with location data reported to efficiency of A-IoT devices. Location Management Function (LMF).

Additionally, A-IoT 1) MIMO Backscatter Communications: Multiple-input devices can serve as anchors to help localize other 3GPP multiple-output (MIMO) has been a key technology, improving nodes, enhancing positioning through triangulation without the capacity and reliability of wireless networks. However, its constructing new Positioning Reference Units (PRUs). This use in backscatter communications has been limited, despite approach is particularly beneficial in rural areas with low its potential to enhance range, capacity, and reliability.

Multi- 3GPP node density. antenna readers can increase the read range through trans- mit and receive beamforming, which requires estimating the VII. C ONCLUSION MIMO backscatter channel. Employing multiple antennas on This work provides an overview of Ambient IoT, a simple, tags is challenging due to their passive nature. low-cost and battery-free technology with significant potential Several open research problems must be addressed to facili- for precision agriculture.

This article explored various use tate MIMO techniques such as beamforming. First, backscatter cases of A-IoT in this field, highlighting its advantages over channel estimation is critical. While acquiring CSI has been RFID, particularly for underground sensing.

Despite promising widely studied in prior MIMO work, backscatter channel prospects, several challenges need to be addressed for A-IoT to estimation remains unexplored due to its unique cascaded become a reality. These include the need for multiple access, structure with forward and backward channels, complicating positioning and efficient multiple access. By exploring these the signal model.

Initial work for the monostatic scenario research directions, we can harness the full potential of A-IoT exists [13], but bistatic/ambient scenarios remain unexplored. in precision agriculture. This article serves as a comprehensive Next, space-time codes can be promising for multi-antenna summary of A-IoT and its promising applications in advancing tag precoding. Passive tags cannot adapt to the channel due to agricultural practices.

R EFERENCES Fabio A. Castiblanco is currently working toward a Ph.D. degree in electrical

and computer engineering with Purdue University, West Lafayette, IN, USA. [1] Alliance for Telecommunications Industry Solutions (ATIS). (2023) His research interests include embedded systems, computer networking, and 6g roadmap for vertical industries. [Online]. Available: https: wireless communications.

Castiblanco received a B.S. degree in electronics //nextgalliance.org/white papers/6g-roadmap-vertical-industries/ engineering from the Universidad Nacional de Colombia (UNAL), Bogota, [2] Y. Zhang, D. J.

Love, J. V. Krogmeier, C.

R. Anderson, R. W.

Heath, Colombia. and D. R. Buckmaster, “Challenges and opportunities of future rural wireless communications,” IEEE Communications Magazine, vol. 59, no. 12, pp. 16–22, 2021. [3] R.

Want, “An introduction to rfid technology,” IEEE Pervasive Comput- ing, vol. 5, no. 1, pp. 25–33, 2006. [4] N. Van Huynh, D. T.

Hoang, X. Lu, D. Niyato, P.

Wang, and D. I. Kim, “Ambient backscatter communications: A contemporary survey,” IEEE Dennis R.

Buckmaster received the B.S. from Purdue University and M.S. Communications Surveys & Tutorials, vol. 20, no. 4, pp. 2889–2922, and Ph.D. degrees in Agricultural Engineering from Michigan State University 2018. in 1984, 1986 and 1989, respectively. He is a Professor of Purdue ABE.

His [5] C. Xu, L. Yang, and P.

Zhang, “Practical backscatter communication current research and teaching interests in agricultural systems management are systems for battery-free internet of things: A tutorial and survey of recent in all aspects of the data pipeline of digital agriculture including sensors, data research,” IEEE Signal Processing Magazine, vol. 35, no. 5, pp. 16–27, flow, analytics, biophysical modeling for decision making, and automation; 2018. his application focus is cropping systems and livestock feeding systems.

He [6] 3rd Generation Partnership Project (3GPP). (2022) Study on ambient is an award winning teacher and mentor. iot (internet of things) in ran. [Online]. Available: https://www.3gpp. org/ftp/Specs/archive/38 series/38.848/38848-i00.zip [7] ——.

(2023) Study on ambient power-enabled internet of things. [Online]. Available: https://www.3gpp.org/ftp/Specs/archive/22 series/ 22.840/22840-j00.zip [8] M. M.

Butt, N. R. Mangalvedhe, N.

K. Pratas, J. Harrebek, J.

Kimionis, M. Tayyab, O.-E. Barbu, R.

Ratasuk, and B. Vejlgaard, “Ambient iot: Chih-Chun Wang is a Professor of Purdue ECE. His current research A missing link in 3gpp iot devices landscape,” IEEE Internet of Things interests are in the latency minimization of 6G-and-beyond wireless networks, Magazine, vol. 7, no. 2, pp. 85–92, 2024. information and coding theory for ultra-low latency communications.

He [9] J. D. Griffin and G.

D. Durgin, “Complete link budgets for backscatter- received the NSF CAREER Award in 2009. He is an IEEE Fellow. radio and rfid systems,” IEEE Antennas and Propagation Magazine, vol. 51, no. 2, pp. 11–25, 2009. [10] Z.

Sun, I. F. Akyildiz, and G.

P. Hancke, “Dynamic connectivity in wireless underground sensor networks,” IEEE Transactions on Wireless Communications, vol. 10, no. 12, pp. 4334–4344, 2011. [11] G. Wang, F.

Gao, R. Fan, and C. Tellambura, “Ambient backscatter communication systems: Detection and performance analysis,” IEEE David J.

Love is the Nick Trbovich Professor of Electrical and Computer Transactions on Communications, vol. 64, no. 11, pp. 4836–4846, 2016. Engineering at Purdue University. He is a Fellow of the IEEE, American [12] C.

Liu, Z. Wei, D. W.

K. Ng, J. Yuan, and Y.-C.

Liang, “Deep transfer Association for the Advancement of Science (AAAS), and National Academy learning for signal detection in ambient backscatter communications,” of Inventors (NAI). IEEE Transactions on Wireless Communications, vol. 20, no. 3, pp. 1624–1638, 2020. [13] D. Mishra and E.

G. Larsson, “Multi-tag backscattering to mimo reader: Channel estimation and throughput fairness,” IEEE Transactions on Wireless Communications, vol. 18, no. 12, pp. 5584–5599, 2019. [14] H. Luan, X.

Xie, L. Han, C. He, and Z.

J. Wang, “A better than alamouti ostbc for mimo backscatter communications,” IEEE Transactions on James V. Krogmeier received the BSEE degree from the University of Wireless Communications, vol. 21, no. 2, pp. 1117–1131, 2021.

Colorado and the MS and Ph.D. degrees from the University of Illinois. He [15] E. Ruzomberka, D.

J. Love, C. G.

Brinton, A. Gupta, C.-C. Wang, is currently Professor of Electrical and Computer Engineering at Purdue and and H.

V. Poor, “Challenges and opportunities for beyond-5g wireless his research interests are in statistical signal processing. security,” IEEE Security & Privacy, vol. 21, no. 5, pp. 55–66, 2023.

M. Majid Butt is a senior staff research scientist at Nokia, USA. He has

authored more than 80 peer-reviewed articles and contributed to more than 100 filed/granted patents in cellular networks domain. He serves as an associate editor for IEEE Open Journal of the Communication Society and IEEE Open Ashwin Natraj Arun is currently working toward a Ph.D. degree in Electrical Journal of Vehicular Technology. He is a senior member of IEEE and IEEE and Computer Engineering at Purdue University, West Lafayette, IN, USA.

COMSOC distinguished lecturer. His research interests include signal processing for wireless communications and information theory.

Amitabha (Amitava) Ghosh (F’15) is a Nokia Fellow and works at Nokia

Standards and Strategy. He joined Motorola in 1990 after receiving his Ph.D in Electrical Engineering from Southern Methodist University, Dallas. Since joining Motorola he worked on multiple wireless technologies starting from Byunghyun Lee received his B.S. and M.S. degree in electrical engineering IS-95 to 3GPP New Radio (NR).

He has more than 65 issued patents, has from Yonsei University, Seoul, South Korea, in 2017 and 2019, respectively. written multiple books and book chapters and has authored numerous technical He is currently pursuing his Ph.D. in electrical and computer engineering at papers. He is also the chair of the NextG alliance, National Roadmap Working Purdue University, West Lafayette, IN, USA. His research interests include Group. beamforming and estimation/tracking techniques for integrated sensing and communication systems.

Table II

J. Love, C. G.