AI + Network Papers19 min read36 citations

AI-RAN: Architecture Design and Performance Evaluation for 5G-Advanced

Dr. Mischa Dohler, Dr. Harish Viswanathan

Ericsson Research

Jan 22, 2026View on arXiv

Abstract

This paper presents the architecture design and comprehensive performance evaluation of AI-RAN for 5G-Advanced networks. We define three deployment models — centralized, distributed, and hybrid — and evaluate each across 15 KPIs including throughput, latency, energy efficiency, and operational complexity. Large-scale simulations with 500 base stations and 50,000 users show that the hybrid model achieves the best balance, improving throughput by 35% and reducing energy by 28% compared to traditional RAN.

AI Summary

AI-Generated Summary
  • Defines three AI-RAN deployment models: centralized, distributed, and hybrid.
  • Hybrid model achieves 35% throughput improvement and 28% energy reduction.
  • Evaluated across 15 KPIs on a 500 base station, 50,000 user simulation.
  • Provides practical deployment guidelines for operators.

Key Findings

  • 1Hybrid AI-RAN offers the best performance-complexity tradeoff for most operators.
  • 2Centralized model is optimal for dense urban but requires high-capacity fronthaul.
  • 3Distributed model excels in edge AI latency but has lower optimization capability.

Industry Implications

Provides a roadmap for operators to adopt AI-RAN incrementally.

Informs 3GPP standardization of AI-RAN interfaces and functions.

Validates AI-RAN as a bridge from 5G-Advanced to 6G.

AI-RAN5G-AdvancedArchitecturePerformance

Read the Original Paper

Access the full paper on arXiv for complete methodology, results, and references.

Open on arXiv

Related Papers