來源36氪•較早收集於 3h
Nokia 與 Nvidia 推出業界首個商用 AI-RAN 平台
#telecom#edge-computing#5g-6gai-ran-platformnokianvidiaai-ran
Nvidia 硬體正深入電信基礎設施，了解 AI-RAN 如何重塑網路容量。
30 秒速覽
有什麼變化
由 Nokia 與 Nvidia 開發的首個商用 AI-RAN 平台
為什麼重要
此整合標誌著電信基礎設施向 AI 優化轉型，有望降低營運成本並提升網路效率。
下一步行動
關注 Nokia 開發者入口網站，了解未來與 AI-RAN 相關的 SDK，探索邊緣運算機會。
誰應關注：Enterprise & Security Teams
關鍵要點
- •由 Nokia 與 Nvidia 開發的首個商用 AI-RAN 平台
- •目標在 2028 年前將無線網路數據容量翻倍
- •預計於 2025 年正式上市
深度解析
本篇為 AI 生成分析，非原文內容。
增強重點摘要
- •The collaboration integrates Nvidia's Aerial AI radio access network software suite with Nokia's AirScale baseband hardware to optimize spectral efficiency.
- •The platform utilizes GPU-accelerated signal processing to replace traditional fixed-function baseband processing, allowing for dynamic resource allocation.
- •Nokia and Nvidia are founding members of the AI-RAN Alliance, an industry body established to integrate AI into mobile network infrastructure.
- •The solution specifically targets the reduction of Total Cost of Ownership (TCO) for operators by consolidating RAN and AI workloads onto a single server architecture.
- •The platform supports multi-tenancy, enabling network operators to run third-party AI applications on the same infrastructure used for radio signal processing.
競品分析
Architecture
- Nokia/Nvidia AI-RAN
- GPU-Accelerated (Nvidia Aerial)
- Ericsson/Intel
- CPU-Accelerated (vRAN)
- Samsung/AMD
- FPGA/ASIC Hybrid
Primary Focus
- Nokia/Nvidia AI-RAN
- AI-Native Signal Processing
- Ericsson/Intel
- Cloud-Native Scalability
- Samsung/AMD
- Energy Efficiency
Market Status
- Nokia/Nvidia AI-RAN
- Commercial Deployment
- Ericsson/Intel
- Commercial Deployment
- Samsung/AMD
- Pilot/Trial Phase
| Feature | Nokia/Nvidia AI-RAN | Ericsson/Intel | Samsung/AMD |
|---|---|---|---|
| Architecture | GPU-Accelerated (Nvidia Aerial) | CPU-Accelerated (vRAN) | FPGA/ASIC Hybrid |
| Primary Focus | AI-Native Signal Processing | Cloud-Native Scalability | Energy Efficiency |
| Market Status | Commercial Deployment | Commercial Deployment | Pilot/Trial Phase |
技術深入
- Utilizes Nvidia Aerial RAN Compute software to perform Layer 1 (L1) signal processing on GPU architectures.
- Implements AI-based beamforming optimization to improve signal-to-interference-plus-noise ratio (SINR) in dense urban environments.
- Leverages Nokia's ReefShark System-on-Chip (SoC) technology in conjunction with Nvidia's Grace Hopper Superchips for hybrid compute tasks.
- Supports O-RAN (Open Radio Access Network) compliant interfaces to ensure interoperability with multi-vendor radio units.
- Employs real-time telemetry data ingestion to train and deploy machine learning models for predictive traffic load balancing.
前景展望基於引用來源的 AI 分析
AI-RAN will become the standard architecture for 6G deployments.
The integration of AI at the physical layer is essential for managing the increased complexity and spectral demands of future 6G networks.
Network operators will transition from hardware-centric to software-defined revenue models.
By hosting third-party AI applications on RAN infrastructure, operators can monetize their network edge as a compute platform.
時間線
2024-02
Nokia and Nvidia announce strategic partnership at MWC Barcelona to develop AI-RAN solutions.
2024-02
Nokia joins the AI-RAN Alliance alongside Nvidia and other industry leaders.
2025-03
Nokia officially launches the commercial AI-RAN platform to the global market.
2025-11
First major tier-1 operator begins commercial deployment of the Nokia-Nvidia AI-RAN solution.
- 2024-02Nokia and Nvidia announce strategic partnership at MWC Barcelona to develop AI-RAN solutions.
- 2024-02Nokia joins the AI-RAN Alliance alongside Nvidia and other industry leaders.
- 2025-03Nokia officially launches the commercial AI-RAN platform to the global market.
- 2025-11First major tier-1 operator begins commercial deployment of the Nokia-Nvidia AI-RAN solution.
AI 週報
閱讀本週精選 AI 大事摘要 →
AI 策展新聞聚合。所有內容版權歸原始發布者所有。
原始來源： 36氪 ↗
每週電子報
每週一封，可隨時退訂。