Interview: Freshwave CTO Offers AI-RAN Walkthrough

Why It Matters

The insight highlights a pragmatic path for operators to adopt AI‑RAN without costly hardware overhauls, accelerating network efficiency and energy savings while meeting security standards.

Key Takeaways

• Freshwave relies on fibre to link RAN sites to data centres
• CTO stresses data sharing over GPU deployment for AI‑RAN
• Common data language needed for signal metrics transmission
• Trust and audit trails essential for AI‑driven closed‑loop control

Pulse Analysis

Artificial intelligence is reshaping how mobile operators manage radio access networks, but the technology’s success hinges on data, not raw compute power at the edge. Freshwave’s Tom Bennett points out that large language models require vast, high‑quality datasets, and the most efficient way to supply those datasets is through robust fibre connections that stream RAN telemetry to centralized AI engines. By keeping the heavy processing in data centres, operators avoid the logistical nightmare of installing GPUs on thousands of small cells and DAS sites, while still leveraging the predictive power of AI.

The infrastructure shift Bennett describes aligns with broader industry trends toward cloud‑native RAN and Open RAN architectures. Fibre’s low latency—approximately one millisecond per 100 km—provides the bandwidth needed for real‑time analytics, enabling functions such as dynamic spectrum allocation, power‑saving mode activation, and automated capacity scaling. A standardized data model that captures signal strength, fade levels, and environmental factors is critical; without it, AI cannot translate raw measurements into actionable insights. This software‑centric approach reduces capital expenditures and accelerates deployment timelines, making AI‑RAN attractive to both incumbent MNOs and neutral‑host providers.

Nevertheless, the move to AI‑driven, closed‑loop control raises governance challenges. Operators must establish transparent audit trails that record every AI‑initiated adjustment, satisfying both internal risk management and external regulators like the UK Telecommunications Security Act. Security of the data in transit and at rest is paramount, and existing telecom security frameworks already provide a solid foundation. As AI models mature and ingest richer datasets—including building material characteristics and urban topology—operators can expect finer‑grained optimisation, lower energy consumption, and improved user experience. Freshwave’s fibre‑first strategy therefore offers a realistic roadmap for the industry to harness AI’s benefits while maintaining trust and compliance.