BLAR-3D: Beta-Regulated Learning-Based Adaptive Routing for 3D Network-on-Chip
29th International Symposium on Design and Diagnostics of Electronic Circuits and Systems, DDECS 2026, Bratislava, Slovakia, 27 - 29 April 2026, (Full Text)
- Publication Type: Conference Paper / Full Text
- Doi Number: 10.1109/ddecs69233.2026.11520987
- City: Bratislava
- Country: Slovakia
- Keywords: 3D-NoC, adaptive routing, Network-on-Chip, Q-learning, reinforcement learning
- Hacettepe University Affiliated: Yes
Abstract
Network-on-Chip (NoC) architectures are widely used in many-core systems due to their scalability. Three-dimensional NoCs further improve performance by stacking multiple layers using Through-Silicon Vias (TSVs), at the cost of increased routing complexity. Deterministic routing algorithms often suffer from congestion under non-uniform traffic, motivating adaptive solutions. In this paper, we present a learning-based adaptive routing approach for 3D NoCs that stabilizes Q-learning-based routing decisions using a beta-regulated probabilistic selection mechanism. Direction-based learning is employed to reduce learning complexity. Experimental results show improvements in latency and throughput compared to deterministic and conventional adaptive routing schemes.