Issuu

International Research Journal of Engineering and Technology (IRJET)

e-ISSN: 2395-0056

Volume: 12 Issue: 10 | Oct 2025

p-ISSN: 2395-0072

www.irjet.net

Collective Agentic Systems in Multi-Domain Telco Clouds: Toward Cognitive 6G Networks Ahmed Awwad1 1Independant researcher, Telecommunications, Colorado, USA

---------------------------------------------------------------------***--------------------------------------------------------------------transformations have evolved largely in isolation, with Abstract - The rapid fusion of artificial intelligence (AI), cloud-native architecture, and telecommunications is reshaping how next-generation networks progress toward autonomy. This paper introduces Collective Agentic Systems, a paradigm in which AI agents distributed across RAN, Core, and Edge collaborate over a unified telco-cloud substrate. The Collective Agentic Intelligence Framework (CAIF) is proposed to coordinate domain agents via multi-agent and federated learning, governed by an intent-aware meta-agent with explainability and sustainability controls. A hybrid emulation testbed-combining srsRAN, Open5GS, Kubernetes/Nephio, Kafka, and MATLAB analytics-quantifies CAIF’s impact. Compared with traditional domain-isolated automation, CAIF achieved 28% lower latency, 21% lower energy use, 25% faster policy convergence, and +6 pp SLA adherence under controlled scenarios. The paper outlines a six-stage fusion roadmap (2010–2035) from rule-based automation to cognitive 6G autonomy, highlighting design principles, governance requirements, and transition milestones. Results indicate that the defining capability of 6G will be collective cognition-networks that sense, reason, and evolve collaboratively, rather than merely higher raw throughput.

limited coordination or knowledge exchange between automation layers [9].

Key Words: AI-Native Networks; Agentic Intelligence; Telco Cloud; Collective Agentic Systems; Cognitive 6G; Federated Learning; Intent-Based Networking; Autonomous Operations.

1. INTRODUCTION

2. LITERATURE REVIEW

Telecommunications networks are entering a phase where software-defined design, cloud-native deployment, and embedded AI must converge to meet stringent performance, agility, and sustainability targets. The evolution from 4G to 5G demonstrated that network function disaggregation and virtualization could enhance flexibility, yet full autonomy across operational layers remains unrealized as in open RAN networks [1], [2], [3]. Existing automation frameworks are largely rule-based and domain-specific, resulting in fragmented and reactive management that limits end-to-end optimization [4]. The emergence of AI-native networks introduces intelligence as an intrinsic property of the network, enabling systems to sense, reason, and act autonomously [5], [6]. In parallel, the telco cloud-built on Kubernetes, microservices, and CI/CD pipelines-has matured into a programmable environment capable of dynamically orchestrating workloads across RAN, transport, and core domains [7], [8]. Despite these advancements, AI and cloud

The evolution of automation in telecommunications has followed a progressive trajectory, from static, rule-based systems toward learning-enabled, adaptive, and intent-driven architectures. This trajectory can be broadly categorized into three epochs: (i) rule-based automation, (ii) AI-assisted orchestration, and (iii) the emerging phase of collective, AInative coordination. Each stage reflects a shift in the relationship between control, intelligence, and cloud infrastructure.

Impact Factor value: 8.315

To bridge this gap, the concept of agentic AI has emerged. These agents are autonomous software entities that can perceive the environment, set goals, and act within their respective domains [10]. However, they typically lack crossdomain coordination. The next evolution, therefore, lies in collective agentic systems, where multiple domain agents collaborate through federated or multi-agent reinforcement learning to achieve global network optimization [11], [12]. This study introduces the Collective Agentic Intelligence Framework (CAIF)-a unified model for multi-domain orchestration and cross-agent collaboration. The framework builds on AI-native and cloud-native foundations to create an intelligent, distributed system capable of self-optimization and ethical governance [13]. The objectives of this research are threefold: 1.

To analyze the historical fusion of AI, cloud, and telecom technologies from rule-based automation to cognitive orchestration. To design and validate the CAIF model for multi-domain coordination. To define a roadmap toward self-evolving, AI-governed 6G networks.

2.1 Rule-Based and Domain-Centric Automation The foundation of network automation was established during the 4G era with Self-Organizing Networks (SON), which introduced localized optimization functions such as automatic neighbor relations (ANR) , coverage optimization and PCI conflict resolution [14]. Although SON reduced human intervention, it relied on deterministic (pre-defined)

ISO 9001:2008 Certified Journal

Page 766