Scholarlink

<p>Despite standardization since 3GPP Release 9, multicast and broadcast services (MBS) remain largely undeployed in commercial networks. Artificial intelligence may catalyze adoption by addressing the optimization complexity hindering practical deployment, particularly as 6G targets demanding applications requiring efficient group-oriented transmission. Following PRISMA guidelines, this survey reviews the 5G MBS architecture and analyzes 23 studies applying AI to multicast and broadcast optimization. Surveyed works demonstrate computational complexity reductions from <InlineEquation ID="IEq1"> <EquationSource Format="TEX">\(O(N^3)\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mi>O</mi> <mo stretchy="false">(</mo> <msup> <mi>N</mi> <mn>3</mn> </msup> <mo stretchy="false">)</mo> </mrow> </math></EquationSource> </InlineEquation> to <InlineEquation ID="IEq2"> <EquationSource Format="TEX">\(O(N^2)\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mi>O</mi> <mo stretchy="false">(</mo> <msup> <mi>N</mi> <mn>2</mn> </msup> <mo stretchy="false">)</mo> </mrow> </math></EquationSource> </InlineEquation>, throughput gains of 18–50%, and resource savings up to 33%. Deep reinforcement learning variants dominate resource allocation and scheduling, while unsupervised clustering methods address multicast group formation and federated learning enables privacy-preserving optimization across distributed deployments. We organize findings across six areas: physical layer intelligence, RAN slicing and scheduling, multicast group formation and routing, RIS-assisted transmission, D2D-assisted multicast, and end-to-end optimization. We identify underexplored areas, non-terrestrial networks, ISAC integration, graph neural networks, and foundation models, and provide a research roadmap addressing standardization gaps and deployment barriers.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Survey on AI technologies aiding multicast and broadcast services in 5G and 6G

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年1月28日

查看全文
  • Marcell Szabó,
  • László Toka

摘要

Despite standardization since 3GPP Release 9, multicast and broadcast services (MBS) remain largely undeployed in commercial networks. Artificial intelligence may catalyze adoption by addressing the optimization complexity hindering practical deployment, particularly as 6G targets demanding applications requiring efficient group-oriented transmission. Following PRISMA guidelines, this survey reviews the 5G MBS architecture and analyzes 23 studies applying AI to multicast and broadcast optimization. Surveyed works demonstrate computational complexity reductions from \(O(N^3)\) O ( N 3 ) to \(O(N^2)\) O ( N 2 ) , throughput gains of 18–50%, and resource savings up to 33%. Deep reinforcement learning variants dominate resource allocation and scheduling, while unsupervised clustering methods address multicast group formation and federated learning enables privacy-preserving optimization across distributed deployments. We organize findings across six areas: physical layer intelligence, RAN slicing and scheduling, multicast group formation and routing, RIS-assisted transmission, D2D-assisted multicast, and end-to-end optimization. We identify underexplored areas, non-terrestrial networks, ISAC integration, graph neural networks, and foundation models, and provide a research roadmap addressing standardization gaps and deployment barriers.