日本語 
SoftBank Corp. and Ericsson Japan K. K. have together revealed a plan to use a next, generation, AI, driven external control system to manage the coverage patterns of Massive MIMO base stations in 5G networks automatically at large scale event venues throughout Japan. This venture is a big leap for the application of artificial intelligence to radio access network (RAN) operations. It facilitates the improvement of communication performance and adaptability in scenarios where user traffic changes rapidly.
The system builds on successful trials conducted during Expo 2025 Osaka, Kansai, Japan, demonstrating uplifts in coverage performance and responding dynamically to fluctuations in user demand — a capability that conventional preset configurations often fail to manage.
Why AI‑Driven Massive MIMO Coverage Matters
Massive MIMO (Multiple, Input Multiple, Output) is the key technology of 5G. It utilizes large antenna arrays to serve a large number of users simultaneously while increasing throughput and spectral efficiency.
Also Read: Dialpad Introduces Real-Time AI in Japan
However, the distribution of real users is very dynamic, especially in arenas, domes, concert venues, and theme parks, where the demand for traffic can change dramatically in a very short time.
Traditionally, network operation models cover patterns that are manually preset based on the expected usage patterns (i. e. , time of day) which are less responsive to unexpected changes like varied event schedules, weather impacts, or crowd movements. In such instances, mobile performance may deteriorate and users may face congestion, slower data speeds, and a disrupted experience.
In order to be able to cope with these situations, SoftBank and Ericsson have introduced a system that utilizes AI to automatically locate, analyze, and adjust the traffic changes by adjusting the coverage patterns.
How the AI Coverage Optimization System Works
At the heart of this solution is an external control server that collects user distribution data every minute from deployed base stations. An AI model analyzes these data — including beam estimation and spatial distribution signals — to automatically determine whether an event or change in traffic patterns is occurring. From there, it dynamically selects the optimal coverage pattern for the Massive MIMO arrays without human intervention.
This real‑time external control approach allows the system to adapt its radio footprint based on observed usage rather than static scheduling, providing flexibility and sustained network performance under varied conditions.
Proven Performance at Expo 2025
One of the most compelling validations of this technology came from deployment at the Expo 2025 Osaka venue, where SoftBank and Ericsson tested the AI‑driven coverage system in outdoor areas highly susceptible to traffic variability. The results showed an approximate 24 % increase in downlink throughput (from 76.9 Mbps to 95.5 Mbps) during periods of rapid traffic changes compared with conventional operation — a significant improvement in user experience and network efficiency.
This gain not only enhanced peak data speeds but also helped mitigate the phenomenon known as “packet stalling”, where congestion during sudden surges in usage can slow or freeze data traffic altogether.
Deployment Across Event Venues
Based on the success at Expo 2025, SoftBank and Ericsson have expanded their AI, driven system to several large event venues such as arenas, domed stadiums, and theme park environments. In these places, user distribution varies greatly during the day or the occurrence of special events. The aim of these deployments is to provide the attendees with consistent, high, performance 5G connectivity even in situations where the usage patterns are highly unpredictable.
At such sites, traffic distribution may differ dramatically between off‑peak times and event periods, requiring base stations to behave in a highly flexible manner to maintain performance. The AI system’s ability to detect and respond automatically provides significant advantages over static coverage strategies.
Strategic Importance for Japan’s 5G Infrastructure
The partnership between SoftBank and Ericsson is a microcosm of the telecom industry trends where AI and machine learning are becoming a more integral part of network control systems in order to manage the complexity that traditional rule, based systems are not able to handle adequately. RAN optimization with the help of AIsometimes called under the umbrellas of AI, RANis perceived as one of the main facilitators of next, generation cellular networks and also future upgrades to 6G.
By enhancing 5G coverage with dynamic AI‑driven control, operators can improve:
Network efficiency: Optimal resource allocation based on real‑time demand
User experience: less congestion which results in fewer slowdowns and drops.
Operational agility: Less dependence on manual network fine, tuning or pre, planned schedules
Scalability: The ability to handle a wide range of scenarios, from extremely crowded urban areas to sudden surges in utilization.
These advantages are of paramount importance as the demand for mobile broadband and real, time applications, like streaming, AR/VR content, and IoT connectivity, keeps increasing in Japan and worldwide.
Implications for Future 5G and Beyond
The success of this initiative is a signal of how AI is changing the telecom sector not just for coverage optimization but also for broader network automation, predictive maintenance, and energy efficiency. AI, powered devices might thus go further into tasks like spectrum management, interference cancellation, and automated fault detection, thus paving the way for more robust and flexible networks.
Besides, these innovations are in line with worldwide research trends, such as those of the O, RAN (Open Radio Access Network) framework, which points out the importance of intelligence and learning inside next, generation mobile networks to allow closed, loop optimization and autonomous operation.
Looking Ahead
As SoftBank and Ericsson continue to perfect and broaden this AI, powered coverage system, it will likely play a part in delivering high, quality 5G services throughout Japan especially at crowded venues where the traffic load can rise unpredictably. The launch thus also shows Japan’s determination to move forward with AI, augmented network technologies that make user’s connectivity experiences better and help mobile communication innovations of the future.
While global mobile service providers are channeling funds into AI and ML for network optimization, the SoftBank, Ericsson collaboration stands as a testimony to how joint innovation can address performance challenges on the ground and open the gateway to future wireless evolution.
English