Patents & Innovations

Patents

• A Method for Mutual Information-Driven Stochastic Federated Learning Policy over Distributed and Scalable evolved O-RAN Architecture in Integrated 6G Networks, European Patent, under review, 2023.
• Continuous network slicing in a 5g cellular communications network via a delayed deep deterministic policy gradient, European Patent, 2021. [Access Link]

Innovations

• Adaptive Standardized Protocol Learning, An adaptive standardized protocol for addressing inter-slice resource contention and conflict in 6G network slicing. [Access Link]
• Explanation-Guided AI (In-hoc Explainability), The core concept behind the proposed approach is the intrinsic interpretability of the reward hypothesis in Deep Reinforcement Learning. [Access Link]
• SliceOps, The proposed approach aligns with this philosophy of segregating the control plane from the user plane in 5G/6G by creating an innovative standalone AI plane to independently provide AI services and functions to the rest of the 6G network. [Access Link]
• Quantum Deep Reinforcement Learning (QDRL) in 6G Slicing, The power of quantum computing pursue a distributed computation and learning can be deemed as a promising prerequisite. [Access Link]
• Massively Large-Scale Distributed Deep Reinforcement Learning (DRL), We explore a novel approach to 6G networks by separating actors from learners and targeting latency cost under a long-term statistical service-level agreement (SLA). [Access Link]
• Brain-inspired Continuous Action in 5G/6G, For solving continuous tasks, this approach adopts and fine-tune an alternative way of updating the actor (policy) in DDPG algorithm to speed-up convergence and fulfill a stable and robust learning process based on TD3 algorithm. [Access Link]