Trustworthy AI-Driven Dynamic Hybrid RIS: Joint Optimization and Reward Poisoning-Resilient Control in Cognitive MISO Networks
2026-04-03T07:24:01Z•2fd6859e6d0d4377f7b3cbe7b62a7213baae50559419e48849abc97b22ec2837
3.4 GHz channel measurementsDRL for resource management (CIVIC)ML for telecomO-RANPareto frontQ2NSQAOAQNOSSDQNUAV A2A channeladversarial ML defensescognitive radio networks (CRN)deep reinforcement learning (DRL)hybrid classical-quantummetaverse resource allocationmulti-objective MDPsns-3physics-informed transformerquantum networkingquantum simulatorquantum walksreconfigurable intelligent surface (RIS)reward poisoningsoft actor-critic (SAC)wideband CFR reconstruction
What happened
Collection of recent research (arXiv 2026-04-03) spanning wireless communications, ML-enabled networking, quantum networking, and networked systems. Key contributions include: (1) a dynamic hybrid active/passive RIS for energy-aware cognitive MISO networks using SAC DRL and the first systematic study of reward-poisoning attacks with a lightweight reward-clipping/anomaly-filter defense; (2) a comprehensive survey of ML integration in O-RAN and associated architectural, performance, and security challenges; (3) Software-Defined Quantum Networking (SDQN) and a Quantum Network OS (QNOS) reference,
Why it matters
A reviewed impact interpretation has not been published for this record.
Evidence and limitations
- Source ID
- arxiv_cs_ni
- Record identifier
- 2fd6859e6d0d4377f7b3cbe7b62a7213baae50559419e48849abc97b22ec2837
- Enrichment time
- 2026-04-03T07:24:01Z
- 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.