HyperCell: GSMA Award-Winning Private 5G at 250 km/h - From São Paulo Grand Prix to MWC Barcelona
Jun 4, 2025
Overview
In late 2024, at Cloudnet.ai, through our subsidiary CloudRAN.AI, we deployed a groundbreaking private 5G network at the São Paulo Autódromo Velocitta. This live pilot provided ultra-fast, reliable connectivity for both race cars and spectators – an engineering feat that earned CloudRAN.AI the GSMA Foundry Excellence Award for Outstanding Industry Vertical Use Case at MWC 2025.
The achievement showcases Cloudnet.ai’s leadership in AI-driven networks and CloudRAN.AI’s role as a key innovator simplifying 5G deployment for enterprises. What was proven on the racetrack is a blueprint for high-performance wireless in any extreme environment, from packed stadiums to busy ports.
Highlights:
Seamless Track-Wide Coverage: A single 5G “hyper-cell” spanned the entire ~4 km race circuit by merging signals from three base stations, eliminating handoffs even as cars hit 250 km/h on the straightaways.
Real-Time In-Car Video: The private 5G network reliably streamed 360° HD video from 8 racecars simultaneously, each with multiple on-board cameras, plus telemetry (speed, engine data, etc.). This gave broadcasters and fans an immersive driver’s-eye view with no stutter or drop-outs.
Spectator Connectivity Boost: Using multi-operator core network sharing (MOCN), the private network’s capacity was made available to public mobile users. Even with 60,000 spectators at Autódromo Velocitta, everyone could enjoy solid mobile coverage thanks to the extra 5G “on tap”.
Rapid, Cloud-Native Deployment: The 5G solution was installed in about one hour by a small IT team – as easy as setting up Wi-Fi – thanks to cloud-native design and plug-and-play base stations. Pre-event RF simulations ensured optimal coverage with no interference between the on-site cells.
Award-Winning Impact: Proven first at the Grand Prix, this hyper-cell solution (delivering ~230 Mbps uplink per car with only 20 W radio power) now stands as an award-winning example of how private 5G can transform connectivity at large events and beyond.
Challenge
Major live events like motorsport races push wireless networks to their limits. Tens of thousands of attendees congest the local cellular towers, and at the same time the event itself demands rock-solid connectivity for mission-critical applications (broadcast video, team communications, IoT sensors, etc.).
At the São Paulo Grand Prix, held at the sprawling Autódromo Velocitta circuit, these challenges can be especially acute:
Huge Crowds, Heavy Traffic: With up to 60,000 fans in attendance, the public 4G/5G networks face extreme data loads. Spectators all trying to share videos, use apps, and stay connected, often overwhelming the capacity of traditional mobile infrastructure. Dropped calls and stalled data are a familiar frustration at sold-out stadiums and race venues. The Grand Prix organizers needed a way to boost capacity on race day to ensure a quality experience for attendees and media crews alike.
High-Speed Coverage Gaps: The cars themselves circle a multi-kilometer track at speeds of 250 km/h or more. A standard approach would deploy multiple cell sites around the track, but as cars streak from one cell’s coverage to another, they must perform constant handoffs between base stations. At such high velocities, these handovers happen in split seconds – a recipe for latency spikes, video freezes, or even dropped connections as a camera car moves out of one cell and into the next.
Moreover, the Doppler effect (the shift in radio frequency as the car moves rapidly relative to the antenna) can throw off the radio link tuning. Together, these factors made it incredibly complex to maintain an uninterrupted, high-bandwidth uplink from a fast-moving racecar using conventional network setups.
Logistical Constraints: Deploying a dedicated network for a one-weekend event had to be fast and minimally disruptive. The team had only a short window to set up equipment at the circuit and absolutely no margin for error during the live race. Placing multiple 5G base stations around the track raised concerns about interference between them and with existing macro network cells.
Thorough RF planning and testing would be required to ensure three adjacent cell sites could operate as one harmonious network without hurting performance. Additionally, any solution had to comply with spectrum regulations and coordinate with local mobile operators – all on a tight timeline leading up to race day.
In summary, our team faced a dual challenge: technical – delivering gigabit-class, low-latency coverage to fast-moving devices across a large area – and operational – implementing this network quickly and safely in a live event environment.
The stakes were high: success would mean unprecedented live motorsport content and happy connected fans; failure would mean blacked-out onboard footage or racetrack stands full of disconnected, unhappy attendees.
Solution
Cloudnet.ai’s answer to these challenges was an innovative CloudRAN.AI “HyperCell” private 5G network – essentially a single supersized 5G cell covering the entire race track and audience, powered by cloud-native intelligence.
The solution blended cutting-edge radio engineering with pragmatic deployment ease:
HyperCell Architecture (No More Handoffs): Instead of multiple independent cells, CloudRAN.ai created one continuous cell across the venue by deploying three 5G base stations that acted in unison. All base stations were controlled by a centralized unit (CU) using standard 5G interfaces (via the nFAPI protocol) to coordinate their distributed radios as if they were a single cell. This meant a racecar could drive the entire length of the circuit without a single handover – the car’s onboard 5G device saw one consistent network the whole time.
By merging coverage zones into one “hyper-cell,” CloudRAN.AI eliminated the latency and packet loss normally triggered by cell boundaries. The pesky dropouts at cell edges vanished, ensuring that live video feeds from the cars never froze as they raced from sector to sector.
Doppler Compensation Algorithm: The CloudRAN.AI team incorporated an AI-driven algorithm to tackle the Doppler effect head-on. As cars approached 250 km/h, the relative motion shifts the frequency of radio signals.
The HyperCell system intelligently anticipated and corrected for these frequency shifts in real time, so that the link budget and signal quality remained optimal even at extreme speeds. This predictive tuning kept the high-bandwidth uplink from each car locked in and stable, further guaranteeing that on-board cameras and sensors stayed connected throughout high-speed laps.
Private 5G + Public Network Integration: Cloudnet.ai’s solution didn’t just help the race teams – it also augmented the public mobile network for fans on-site. Using a Multi-Operator Core Network (MOCN) approach, the private 5G deployment was linked with the spectator’s regular mobile carriers. In practice, this meant that on race day, spectator devices could automatically connect through the CloudRAN hyper-cell as additional capacity, as if it were just another part of their carrier’s network.
This network-sharing model would instantly boost throughput for thousands of users in the stands without them needing to switch SIMs or even know it was happening. The CloudRAN hyper-cell operated in the 3.3–3.8 GHz mid-band spectrum, providing an ideal mix of coverage and capacity. By configuring the network for a strong uplink (given the heavy upload demands of live video from cars), Cloudnet.ai ensured that both the broadcast uplinks and consumer downlinks were well supported in parallel.
Cloud-Native, Rapid Deployment: A hallmark of Cloudnet.ai’s approach is its enterprise-friendly, plug-and-play deployment model. The entire private 5G system was delivered as a pre-integrated, cloud-managed solution. Setup was incredibly quick – roughly an hour to install and configure the three base stations on-site. “The cloud-native technologies make the private 5G solution very simple to deploy, just like setting up a traditional Wi-Fi network,” notes Lion Li, CloudRAN.AI’s Director of Product Management.
With minimal tuning and no complex on-site core network installation (the core was run in the cloud), a small IT team was able to bring the network online before the first practice laps. CloudRAN’s software automation handled most of the optimization.
Pre-deployment simulations were run to fine-tune coverage overlap and prevent any interference between the three radios , so by the time hardware was in place, the network performed as expected out-of-the-box. This rapid, software-defined deployment illustrates Cloudnet.ai’s broader mission to remove the complexity from private 5G through AI and automation – making advanced networks accessible to enterprises without deep RF expertise.
Together, these solution components formed an end-to-end platform that delivered fiber-like wireless performance in one of the most demanding scenarios imaginable. CloudRAN.AI’s HyperCell design not only solved the immediate technical hurdles (coverage, handovers, Doppler), but did so in a way that was operator-friendly (integrating with carriers) and enterprise-friendly (fast and simple deployment). It’s the fusion of deep technology with practical usability.
Outcome
The results at the São Paulo Grand Prix were nothing short of transformational:
Unbroken High-Bandwidth Streams: Throughout the race weekend, the private 5G network delivered as promised – eight racecars streaming 360° HD video in real time without a glitch. Each car’s three onboard cameras fed live video to the cloud, where it was relayed to broadcasters and jumbo screens.
Observers noted that even as cars whipped past at top speed, their video feeds never stuttered or dropped. The average uplink per car clocked around 20 Mbps, well within the network’s tested capacity (~230 Mbps total uplink). This gave the production teams ample headroom to push new angles and rich telemetry to the broadcast. In essence, CloudNet.ai proved that a properly engineered 5G network can outpace a race car, handling huge data rates under extreme motion.
Happy Fans, Connected Venue: On the spectator side, the typically spotty cell coverage at big events was noticeably improved. Fans at Autódromo Velocitta could share selfies, live-stream from the grandstands, and use data-heavy apps with ease, even during peak usage moments like the race start. By integrating with carriers, the private network offloaded a significant amount of traffic from the public system, meaning visitors experienced faster speeds and more reliable service than in past events. This success did not go unnoticed by the telecom analysts and the enterprise clients in attendance – it demonstrated a clear path to providing VIP-level connectivity at mass gatherings without over-investing in permanent infrastructure.
Industry Accolades: The impact of this deployment reverberated beyond just one event. In March 2025, the project earned CloudRAN.AI the prestigious GSMA Foundry Excellence Award (Outstanding Industry Vertical Use Case) at MWC Barcelona 2025.
This award, given to just a handful of top projects globally, recognized the solution’s innovative approach to a real-world problem and its potential to be replicated across industries. The hyper-cell project was hailed as a “revolution in how connectivity is delivered during large-scale events, ensuring seamless, reliable, and high-performance experiences for all.” CloudRAN.AI’s achievement featured prominently in MWC coverage, positioning the company as a thought leader in the private 5G arena.
Path to Broader Adoption: Fresh off this success, CloudRAN.AI (backed by Cloudnet.ai) is already gearing up to scale this solution to new venues and markets. Following the successful pilot, the start-up is aiming to secure two or three partnerships in 2025, working with both event organizers and mobile network operators to deploy the solution commercially in Brazil and elsewhere, says Lion Li.
The Grand Prix proved the technology; next comes broad adoption. Stadium concerts, championship games, and other motorsport series are natural next steps – anywhere large crowds or fast-moving devices strain conventional networks.
Beyond sports, high-density industrial sites like ports and logistics hubs are also on the radar, where fleets of vehicles or robots need uninterrupted connectivity across vast facilities. These environments share the key challenge addressed at São Paulo: a need for massive, reliable wireless capacity in a concentrated area. This case study shows a viable template to meet that need.
This motorsport deployment underscores a broader strategic point: by leveraging cloud-native design and AI-driven optimization, previously “impossible” connectivity problems can be solved in practical, business-friendly ways. In the spirit of Geoffrey Moore’s playbook, Cloudnet.ai crossed a chasm here – turning an advanced telecom concept into a working solution that delivers clear business value to event organizers, broadcasters, and mobile operators.
CloudRAN.AI’s leadership in this space is now proven. The company has demonstrated not just a one-off triumph, but a replicable model for the future of private 5G networks. From the pit lanes of São Paulo to the smart factories and smart cities of tomorrow, here at Cloudnet.ai we're setting out to prove how cutting-edge network technology can be strategically deployed for transformative outcomes. The race at Velocitta was just the beginning – and it ended with Cloudnet.ai and CloudRAN.AI ahead, carrying the checkered flag for 5G innovation.
