.png){kind=icon}
.png){kind=icon}
Return
Hydra-RAN Task 3: A Core-Independent AI Framework for Robust Intra-SRU Switching (ISS) for 6G Networks
.png){kind=icon}
PRE
AI
I
Q
S
D
S
M
K
DOI：10.1109/tcomm.2026.3714355.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Existing 5G mobility schemes fail under severemillimeter-wave blockage due to reactive, radio-only triggers, causing latency spikes and service interruption. This paperintroduces Hydra-RAN Task 3, a core-independent AI frameworkfeaturing a dual-policy intra-SRU switching (ISS) mechanism. The framework uniquely integrates: (i) a sensor-driven proactive(Detective) policy for anticipatory switching; (ii) a measurement-driven reactive (Reactive) policy for robust fallback; and (iii) a lightweight deep reinforcement learning (DRL)-based selector. A sparse multi-task learning (SMTL) engine enables efficientmulti-modal sensing fusion, achieving a 74.4% parameter reduc-tion compared to separate agents. Comprehensive simulationsdemonstrate that the proposed architecture achieves up to 75% reduction in blockage recovery time, maintains URLLC-compliant latency (sub-1 ms) in over 92.5% of events, andprovides robust out-of-distribution (OOD) generalization (89.7% URLLC compliance versus 47.3% for baselines), representing a 17% improvement in mean accuracy over the best-performingbaseline. This work fills a critical gap in 6G literature bypresenting a rigorously evaluated, dual-policy, sensor-RF fusedISS architecture.
Keywords:
Hydra radio access network (Hydra-RAN)
multi-functional networks
integrated sensing and communication (ISAC)
deep reinforcement learning (DRL)
sparse multi-task learning (SMTL)
intra-SRU switching (ISS)
blockage resilience
cooperative perception
out-of-distribution generalization
6G networks
Journal
IF:
8.3
Papers:
1.2W
Citations:
3.6W
Organization
Cited Papers