- Ericsson is introducing an AI-first approach to building networks with the latest RAN hardware, engineered to meet AI-driven network demands, delivering greater uplink performance, improved TCO, and enhanced energy efficiency
- Ericsson’s RAN software enhancements include AI-managed Beamforming, AI-powered Outdoor Positioning, and a best-in-class AI model for instant coverage prediction
- New AI‑ready radios, featuring Ericsson Silicon with neural network accelerators, boost on‑site AI inference capabilities in Massive MIMO radios, enabling real‑time optimization and full stack, fully distributed AI
The accelerating adoption of AI-powered devices around the world underlines the importance of mobile networks that can seamlessly manage the new performance requirements. To meet these evolving demands – driven by multi-modal artificial intelligence (AI) and augmented reality (AR) use cases requiring far greater uplink performance – Ericsson (NASDAQ: ERIC) is introducing a new suite of radio, antenna, and AI RAN software solutions ahead of Mobile World Congress (MWC) 2026 in Barcelona.
The new products will enable communications service providers (CSPs) to monetize AI devices with differentiated connectivity and launch new services. This new lineup includes new AI-ready radios and Ericsson Silicon featuring neural network accelerators that boost AI inference capabilities in Massive MIMO radios. The neural network accelerators are programmable matrix cores that are integrated in the Ericsson Many-Core Architecture and optimized for AI and Machine Learning.
The new portfolio suite is structured around three pillars:
1. Ten AI-ready radios:
Next‑generation Massive MIMO and remote radios engineered for higher downlink efficiency and uplink performance breakthroughs. This includes:
- Expanded performance breakthroughs including: High-power FDD M-MIMO with AIR 3286 and AIR 3211 combining M-MIMO on both TDD and FDD and an expanded 8-receiver portfolio enabled by Radio 4891 and Radio 4458, which will improve uplink performance to support AI and AR applications.
- High‑power triple‑band radios: Radio 4488 and Radio 4464, designed for network consolidation and RAN sharing.
- TDD M‑MIMO advancements including AIR 3267 with 600 MHz instantaneous bandwidth in just 13 kg and AIR 6492 with higher power (480 W) and 256 antenna elements for next-level performance.
2. RAN software:
A new suite of AI-powered functionalities to improve efficiency and reliability, while meeting the latency requirements demanded by AI applications
- The new AI-powered functionalities consist of AI-managed Beamforming, AI-powered Outdoor Positioning, and a state-of-the-art AI model for instant coverage prediction, which complements Ericsson’s market-leading AI-native Link Adaptation software
- To support CSPs meeting the bounded latency and reliable uplink speeds required by AI and AR applications, we have added tools like the new Latency Prioritized Scheduler and Low Latency Mobility, which deliver up to seven times faster response times and a better and more reliable connection.
Together, these AI software enhancements increase efficiency and intelligence and enable new service offerings that were previously not possible.
3. Five high-performing antennas:
Interleaved and passive antennas that enhance spectrum utilisation and simplify site design. The new energy-efficient passive antennas, built on the trio net design, maximise uplink performance, carrier aggregation, and spectral efficiency. The expanded Interleaved AIR portfolio now adds three new configurations to the two already launched, supporting full flexibility for TDD and FDD M‑MIMO deployments.
Mårten Lerner, Head of Networks Strategy and Product Management, Ericsson, says: “As AI transforms traffic patterns and raises consumer expectations, networks must provide precise performance where and when it’s needed most. At Ericsson, we’re committed to delivering that needed performance. We are also embedding AI solutions across our full portfolio, introducing AI RAN software solutions that deliver revolutionary improvements in spectral efficiency. We are now taking the final step to full AI enablement across our portfolio by introducing neural network accelerators in our leading Massive MIMO portfolio.”
Ericsson’s latest portfolio additions integrate AI more deeply into the RAN, enabling real-time optimisation of performance and energy usage. With these AI-powered capabilities and a robust hardware lineup, CSPs can deliver consistent and differentiated user experiences. The portfolio is designed to capitalise on the growing demand for AI-enabled devices by providing superior uplink performance and service differentiation. Additionally, AI-driven efficiency across both hardware and software helps lower TCO and accelerates time to value.
Iain Milligan, Network Development & Infrastructure Director, Vodafone Three, says: “By bringing the Vodafone and Three networks together on a single, high-performing 5G infrastructure powered by Ericsson’s RAN, Vodafone Three is ushering in a new era of connectivity – unlocking next-generation technology and paving the way for AI-driven innovation.” We’re embedding AI throughout our network, reducing operational complexity, boosting energy efficiency, and delivering faster, more reliable connectivity for our customers in every corner of the country.”
Source : www.ericsson.com
