Sep 2026· Research Journal of Pure Science and Technology· 0 citations
Software-Defined Networks and 5G
TL;DR
Predictive intelligence models serve as the backbone of next-generation 6G network planning and optimization and offer promising directions for delivering intelligent, resilient, and user-aware telecom services, according to a systematic review of current research.
Abstract
The advent of sixth-generation (6G) mobile networks promises to revolutionize
telecommunications through ultra-low latency, massive connectivity, and intelligent
automation. As the industry shifts toward product-centric service delivery—focusing on usercentric applications, dynamic network slicing, and on-demand resource allocation—the role of
predictive intelligence in network planning and optimization becomes increasingly vital. This
systematic review synthesizes current research on predictive intelligence models applied to
product-centric 6G network environments, aiming to evaluate their effectiveness,
methodological diversity, and potential for real-time decision-making in highly complex
systems. The review is conducted following PRISMA guidelines, encompassing studies from
2018 to 2024 across major digital libraries. Selected models include machine learning
algorithms, deep learning networks, Bayesian inference systems, and hybrid approaches used
for traffic forecasting, user behavior prediction, resource provisioning, and failure mitigation.
The findings reveal a growing reliance on supervised and unsupervised learning, particularly
in network traffic analysis and proactive quality-of-service (QoS) assurance. Furthermore, the
review identifies key predictive features utilized in these models, such as historical traffic data,
environmental conditions, device mobility, and service consumption patterns. Emphasis is
placed on models that support intent-based networking and enable autonomous network
reconfiguration. Evaluation metrics, including prediction accuracy, scalability, and
computational overhead, are compared to highlight performance trade-offs in real-world
deployments. The review also uncovers significant challenges, such as data sparsity, high
model complexity, and the need for explainable AI to improve trust and transparency in
automated decisions. Gaps in current literature point to limited research on real-time multidomain orchestration and predictive analytics tailored for vertical-specific applications like
healthcare, autonomous mobility, and immersive media. In conclusion, predictive intelligence
models serve as the backbone of next-generation 6G network planning and optimization. Their
integration into product-centric network management paradigms offers promising directions
for delivering intelligent, resilient, and user-aware telecom services. The paper concludes with
recommendations for future research, including the development of lightweight models,
federated learning frameworks, and integration with digital twin systems for predictive
simulations.
The findings revealed that AI enhances service reliability, reduces operational costs, minimizes latency, improves throughput, and supports proactive network management while ensuring compliance with Quality-of-Service requirements.
Ale Felix, Jude A. Adeleke, A. Abdullahi et al.· International journal of com...· 0 citations
This paper aims to provide references for subsequent academic exploration and industrial application, helping to enhance the efficiency of urban traffic management, alleviate congestion, ensure safety and promote sustainable development.
A detailed review of the evolution of methodologies, from rule-based heuristics to deep reinforcement learning (DRL), federated learning (FL), quantum annealing, and, ultimately, the emerging Agentic AI paradigm is provided.
Nikol Gotseva, Antoni Ivanov, Atanas Vlahov et al.· IEEE Open Journal of the Com...· 0 citations
This article presents the first dedicated and comprehensive survey of DRL for Open AI-RAN, and reviews the foundations of model-free, model-based, offline, safe, multi-agent, federated, and transfer learning, and provides an O-RAN-aware framework for formulating RAN control problems through states, observations, action...
Jie Lu, Peihao Yan, Qijun Wang et al.· 0 citations
A practical AI-enabled operating framework to convert streaming measurements to failure probability, time-to-impact, root-cause ranking, optimization recommendations and governed automation actions is proposed.
M. Imran· Global academic journal of e...· 0 citations
This review examines representative applications and key functionalities within each domain, highlighting how AI techniques—such as deep learning, graph-based models, reinforcement learning, and emerging foundation models—are adapted to diverse transportation contexts.
X. Chen, Jian-Jun Wu, Lu Zhen et al.· Engineering Management· 0 citations
Related blog posts
MIT News · Artificial Intelligence· news.mit.eduNov 17, 2026
By studying the computational foundations of language learning, Associate Professor Jacob Andreas builds intelligent systems that can communicate effectively with humans.
MIT News · Artificial Intelligence· news.mit.eduNov 17, 2026
AI coding agents are increasingly help developers fix bugs and build features largely on their own, but every interaction with their models consumes tokens. In each session, an agent sends instructions, tool descriptions, and session history to its model. Most efforts to reduce this cost focus on individual tasks. Yet developers return to the same […] The post TokenEconomizer: Learning from Experience to Make AI Coding Agents More Efficient appeared first on Microsoft Research.
MIT News · Artificial Intelligence· news.mit.eduNov 9, 2026