What are you looking for?

Tag Archives: ara

Measurement Study of Dynamics and Liquid Data Transport in OneWeb LEO Satellite Networks

Author Taimoor Islam Posted on

Authors: Evan Gossling, Owen Perrin, Daji Qiao, Hongwei Zhang

This work conducts a measurement study of the Eutelsat/OneWeb Low Earth Orbit (LEO) Satellite Network (LSN) using ARA’s backhaul OneWeb user terminal (UT). OneWeb has been an under-studied network, so we first conduct a measurement study of the OneWeb LSN from a user’s perspective, and then examine the potential benefits of liquid data transport to increase the reliability of LSNs, as some LSN deployments may observe up to 2% packet loss1.

Our topology figure showcases ARA’s backhaul UT deployed on the rooftop of Iowa State University’s Wilson Residence Hall. Through ARA’s online portal1, we can run experiments from a Dell PowerEdge server directly attached to the OneWeb indoor unit (IDU). Additionally, we utilize a virtual machine (VM) from the Google Cloud Platform near Ashburn, Virginia.

The UT connects via the Ku band to the OneWeb satellite constellation. The connected satellite then connects back to OneWeb ground stations via the Ka band in a bentpipe fashion. These ground stations transmit to and from OneWeb points of presence (PoPs), which are larger data centers where traffic within the OneWeb network is peered with the wider Internet. The service-level agreement (SLA) for ARA’s OneWeb connection is 100 Mbps downlink and 20 Mbps uplink.

We experiment with liquid data transport over our existing OneWeb LSN and deploy both endpoints of the transport in our architecture between the ARA server on ISU campus in Ames, Iowa, and the Google Cloud VM at Ashburn, Virginia.

We examine the statistical performance of liquid data transport in comparison to the baseline transport of TCP with no added coding scheme (aside from any coding which OneWeb may internally utilize). Our results figure compares the performance of liquid data with that of un-encoded TCP with BBR and CUBIC congestion control algorithms. We examine a time series of transferring data at a target rate of 10 Mbps over the LSN, which is subject to an artificial loss rate of 1%.

This additive 1% loss is used to demonstrate the effectiveness of liquid data in instances where the LSN drops packets, as is possible with LSNs such as Starlink, and to a certain extent, OneWeb. Overcoming this packet loss is important due to the network requirements of real-time applications, of which such applications will incur performance degradation when retransmissions must be made, which is exasperated when using an LSN.

Our figure shows that the performance of CUBIC increases significantly when liquid data transport is utilized compared to without it. To better understand this, we can further analyze the cwnd shown in the figure. As we can see, the cwnd value never increases to a suitable quantity for CUBIC (with no coding) when consistent losses are experienced. Essentially, these loss-based congestion control mechanisms prevent any substantial growth of the cwnd. On the other hand, when a tunnel is established between the TCP sender and receiver by the liquid data transport, packet losses are overcome inside the tunnel using the redundant repair data generated by the erasure code, thus shielding the TCP endpoints. As a result, the cwnd remains around a constant high value that corresponds to the target data rate of 10 Mbps.

BBR’s performance is comparatively similar in both cases as BBR is a delay-based protocol and its model of the link is not drastically affected by these packet losses; not to the degree that CUBIC’s loss-based model is. However, as the loss rate increases, the amount of retransmissions BBR (with no coding) must make increases correspondingly2, whereas liquid data transport is able to overcome the packet loss without incurring additional retransmissions. For the traces shown in Fig. 9, BBR with no coding incurs 1767 retransmissions throughout the trace to transmit the application data, while BBR with liquid data transport only experiences 81 retransmissions.

Using liquid data transport, packet losses are shielded from the TCP endpoints where the congestion control algorithm is executed. It also eliminates or significantly reduces the amount of packet retransmissions. Both factors allow for natural growth and steady-state moderation of the cwnd size, which, in turn, results in increased and stable throughput. Overall, we find that OneWeb generally fulfills its service-level agreements, and liquid data transport may be used to further increase the reliability of TCP flows over LSNs.

Check the full article here.

[1] F. Michel, M. Trevisan, D. Giordano, and O. Bonaventure, “A first look at starlink performance,” in Proceedings of the 22nd ACM Internet Measurement Conference, 2022, pp. 130–136.
[2] Y. Cao, A. Jain, K. Sharma, A. Balasubramanian, and A. Gandhi, “When to use and when not to use bbr: An empirical analysis and evaluation study,” in Proceedings of the Internet Measurement Conference, 2019, pp. 130–136.

At-Scale, Real-World srsRAN Experiments in ARA

Author Taimoor Islam Posted on

ARA is pleased to announce new long-range and field-validated open-source 5G and open RAN capabilities using srsRAN across rural and agricultural environments. This large-scale, outdoor deployment demonstrates stable, reliable and high-throughput connectivity using programmable base stations and commercial-off-the-shelf (COTS) UEs. Leveraging high-power and low-noise amplifiers at the programmable base stations, ARA now supports over-the-air coverage up to 1.3km using NI N320 software-defined radios and Quectel RM500Q-GL UEs. Recent outdoor trials demonstrate stable attachment, high throughput, and robust link reliability, even in complex terrains such as valleys and cell edges—showcasing ARA’s ability to expose rich propagation diversity and realistic interference dynamics. Single-cell results shown in Figure 1 confirm strong throughput  scaling with distance and terrain, while multi-cell trials reveal predictable patterns of throughput, latency, and jitter. Figure 3 shows reliability in terms of packet delivery rate in multi-cell interference scenarios. The setup for the results presented in Figure 3 is illustrated in Figure 2. These characteristics make ARA uniquely suited for benchmarking, optimization, and future 6G and Open RAN research in real-world conditions. 

Beyond connectivity, ARA’s srsRAN deployment enables end-to-end measurement, prototyping, and multi-cell experimentation across distributed UE sites and multi-sector gNBs. Researchers can capture per-UE and per-packet metrics—such as throughput, packet delivery reliability, latency, jitter, and interference effects at kilometer-scale ranges. The ARA platform thus enables rigorous exploration of URLLC, HRLLC, multi-UE contention, rural macro-cell behavior and interference modeling under rich outdoor channel conditions leveraging srsRAN software stack. With such capabilities, ARA is well positioned as one of the few living labs to directly support the mission of the Linux Foundation Open Centralized Unit Distributed Unit  (OCUDU) initiative by providing a  large-scale, field-deployed and fully programmable open RAN platform that accelerates open-source ecosystem adoption, reduces reliance on proprietary infrastructure, and enables rigorous next-generation wireless research and prototyping in strategic agricultural and rural settings. Example experiments can be conducted by following the detailed ARA user manual. 

Figure 1: Multi-UE downlink throughput
Figure 2: Multi-Cell deployment scenario 
Figure 3: Packet delivery rate in multi-cell interference scenarios

ARA-to-ICICLE: Remote Data Sourcing for Agricultural AI Processing

Author Sarath Babu Posted on

Modern agricultural AI relies on timely, high-quality field data. Yet collecting and processing this data in real time, especially in rural environments, remains a critical challenge. The ARA-to-ICICLE pipeline addresses this gap by enabling live agricultural data streaming from remote locations directly into an AI processing pipeline, unlocking faster insights and more responsive, data-driven farming.

This collaboration between the ARA project at Iowa State University, ICICLE at The Ohio State University, and the San Diego Supercomputer Center combines expertise in wireless networking, distributed systems, and AI-driven agricultural analytics, demonstrating how cross-disciplinary innovation can transform farming operations.

From a Soybean Field in Iowa to an AI Pipeline in Ohio 

Our demonstration focuses on a practical use case: monitoring a remote soybean field in Ames, Iowa, located at Iowa State University’s Agronomy Farm. A drone equipped with ARA wireless capabilities flies over the field, capturing high-resolution live video. Instead of storing the data for later analysis, the video is streamed in real time to the ICICLE AI pipeline at The Ohio State University, enabling immediate preprocessing and analysis. The live stream is broken into individual frames, geo-tagged, and processed for AI-driven tasks such as weed detection, allowing actionable insights to be generated in near real time. 

ARA Wireless: Enabling Edge Connectivity in Rural Environments

This demonstration is built around the ARA wireless network, which provides the connectivity between the field-deployed drone and downstream computing infrastructure, as illustrated in the figure below.. The drone carries an ARA User Equipment (UE) payload, consisting of an onboard compute node (Raspberry Pi 5 16GB RAM), a high-resolution camera (Arducam OwlSight 64 MP Auto Focus), and a cellular wireless module (Quectel RG530). As the drone flies over the soybean field, it connects to a nearby ARA base station and continuously streams video to the ARA Data Center through ARA x-haul links.

ARA-to-ICICLE architecture.

Rural deployments present unique challenges: unlike urban areas, base stations are often not connected by optical fiber. ARA addresses this limitation by providing long-range, high-capacity wireless backhaul (shown as AraHaul (Wireless Backhaul) in the figure above), leveraging technologies such as millimeter-wave, microwave, and free-space optical communication. This approach demonstrates that even fiber-scarce rural areas can be fully integrated into high-throughput AI pipelines, enabling real-time agricultural insights anywhere.

At the ARA Data Center, a dedicated video streaming server receives the incoming feed. ICICLE then retrieves the stream directly from this server for further processing and analysis.

ICICLE: Turning Remote Data into Actionable Insights

Once the video reaches ICICLE, the system generates individual frames from the live stream and geo-tags them using metadata embedded in the video. These processed frames are then fed into the ICICLE AI pipeline.

In our demo, we focus on weed detection, considering two weed classes. The AI engine analyzes the incoming frames and identifies weed regions, visually highlighting detected weeds for each class. Snapshots of weed detection results at ICICLE are shown in the figure below. This end-to-end workflow, from drone-based sensing to AI inference, demonstrates how remote agricultural environments can be tightly integrated with modern AI systems.

ARA-to-ICICLE: Weed detection snapshots.

Impact and Future Directions: Wireless and Agricultural Innovation Together

The ARA-to-ICICLE pipeline exemplifies a scalable, flexible framework for remote data sourcing in agriculture, providing:

  • Faster insights: Live streaming and real-time processing reduce the time from data capture to actionable recommendations.
  • Precision farming: AI-driven weed detection, crop health monitoring, and yield estimation support targeted interventions such as selective pesticide application.
  • Connectivity solutions for rural areas: Long-range wireless networks overcome infrastructure limitations, enabling AI adoption in remote fields.
  • Extensibility: Beyond weed detection, the system supports a wide range of applications including phenotyping, environmental monitoring, and precision crop management.

By combining ARA’s wireless infrastructure with ICICLE’s AI capabilities, this work illustrates how data-driven, responsive farming is possible anywhere, even in the most remote agricultural landscapes.


Authors: Sarath Babu, Vincent Lee, Joscif Raigne, and Hongwei Zhang (Iowa State University), Yatish S. M., Matt Lieber, and Hari Subramoni (The Ohio State University), and Martin Kandes (San Diego Supercomputer Center)

ARA December 2025 Newsletter Is Now Live!

Author Taimoor Islam Posted on

We’re excited to share that the December 2025 ARA Newsletter is now available! This edition highlights ARA milestones, community insights, and the newest capabilities unlocked through the ARA wireless living lab. Highlights of this edition include:

  • Join us in September 2026 for AraFest 26! Co-located with the 2026 Farm Progress Show, AraFest’26 will showcase breakthroughs in advanced wireless technologies. Please join us to co-shape AraFest’26. Talks, panels, and demos are all welcome!  
  • ARA capability update: A Starlink terminal has been installed next to the OneWeb user terminal. More experiments are supported by ARA , such as outdoor srsRAN experiments and OAI experiments at new outdoor sites.
  • User stories of how ARA can be used for 1) field scouting acceleration in precision agriculture and 2) random access reliability enhancement for NextG. 

“Amiga”: Accelerating Field Scouting with ARA’s FutureG Connectivity 

Author Taimoor Islam Posted on

In the AIIRA national AI institute, Prof. Soumik Sarkar (Mechanical Engineering) and Prof. Asheesh “Danny” Singh (Agronomy) at Iowa State University leverages ARA to investigate how networked agricultural ground vehicles, remote sensing and computing, and machine learning can automate labor-intensive field scouting—transforming hours of manual assessment into near-real-time analysis and decision-making.

Traditional field scouting and crop phenotyping are both time-consuming and labor-intensive. To address this challenge, the team is developing an integrated platform (shown in the picture below) that combines a highly customized, field-deployable “Amiga” platform (Bonsai Robotics, San Jose, CA), which can be equipped with a wide range of sensors such as high-resolution imaging sensors, with an automated navigation system, an ARA communication user endpoint, and a high-performance computing unit. Together, these components enable automated data collection, fast model training, and data-driven actuation decisions directly in the field.

The ARA wireless living lab plays a key role in this initiative. The custom ARA communication endpoint developed for the Amiga robot delivers the throughput, coverage, reliability, and low latency needed for real-time video streaming, data collection, and edge inference, which is in sharp contrast to the unstable and intermittent connectivity commonly experienced in rural farm fields today. Joscif Raigne, a PhD student who manages and operates the Amiga experimental platform, shared his enthusiasm: “The sky is the limit for agricultural applications when unmanned vehicles, field sensors, and computing servers are interconnected through next-generation, reliable, high-performance networks.” 

The Amiga platform was featured as a plenary demo at AraFest’24. Videos of the presentation and demonstration can be found here. Currently, the team is expanding the pilot to include other types of unmanned agricultural vehicles, such as scouting drones and sprayer drones, for precision weed control and other farm applications.

1 2
Iowa State University wordmarkNIFA Logo | PVAMU Home     ARA is funded in part by the NSF award 2130889, NIFA award 2021-67021-33775, and PAWR Industry Consortium.