ML-Based Real-Time Downlink Performance Prediction in Standalone 5G NR Using Smartphones
2026-04-14T07:23:53Z•7d744a6573eca21b08cbd92e96b0f6be9e3d20e13cad17ae9ea30365e9faa2d5
5GBLERCOTS devicesCayley graphsCoMPHHFLLLM agentsNetAgentBenchO-RANR2E-VIDagentic network managementbenchmarkingdecision-theoretic safetyedge computingfederated learningfinite-state-machinemachine learningmobile measurementsmulti-agent systemspersona-driven agentsreinforcement learning (RL) optimization of graphssrsRANthroughput predictiontopology optimizationvideo inference
What happened
Collection of recent networking and systems research (arXiv) covering: ML-based real-time 5G downlink throughput and BLER prediction using COTS smartphones and srsRAN; NetAgentBench, a deterministic FSM benchmark revealing multi-turn failure modes of LLM-based network agents; Hybrid Hierarchical Federated Learning (HHFL) for exploiting CoMP in 5G/NextG to accelerate convergence; R2E-VID, a two-stage temporal-gating routing framework for elastic edge-cloud video inference; a persona-driven multi-agent framework and decision-theoretic validation for safe O-RAN automation; CayleyTopo, an RL-augme
Why it matters
A reviewed impact interpretation has not been published for this record.
Evidence and limitations
- Source ID
- arxiv_cs_ni
- Record identifier
- 7d744a6573eca21b08cbd92e96b0f6be9e3d20e13cad17ae9ea30365e9faa2d5
- Enrichment time
- 2026-04-14T07:23:53Z
- AI-assisted enrichment
- Yes
This record may overlap with other records. Its enrichment can be incomplete or wrong, and machine assistance was used. Validate consequential decisions against the linked source and your own environment.