Daniel Warren

Daniel Warren

Auto-generated profile, based on publicly available information

Daniel Warren

Communication Solutions, Samsung R&D Institute U.K. (SRUK), Surrey, U.K.

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
17 January 2024 in IEEE Transactions on Vehicular Technology
Verified
Journal Article
Closed access
ATAndrea Tassi
Andrea Tassi
DWDaniel Warren
Daniel Warren
YWYue Wang
Yue Wang
DBDeval Bhamare
Deval Bhamare
RBRasoul Behravesh
Rasoul Behravesh
Next-generation mobile core networks are required to be scalable and capable of efficiently utilizing heterogeneous bare metal resources that may include edge servers. To this end, microservice-based solutions where control plane procedures are deconstructed in their fundamental building blocks are gaining momentum. This letter proposes an optimization framework delivering the partitioning and mapping of large-scale microservice graphs onto heterogeneous bare metal deployments while minimizing the total network traffic among servers. An efficient heuristic strategy for solving the optimization problem is also provided. Simulation results show that, with the proposed framework, a microservice-based core can consistently support the requested load in heterogeneous bare metal deployments even when alternative architecture fails. Besides, our framework ensures an overall reduction in the control plane-related network traffic if compared to current core architectures.

9 September 2025 in IEEE Transactions on Vehicular Technology
Verified
Journal Article
Closed access
ATAndrea Tassi
Andrea Tassi
OKOluwatayo Yetunde Kolawole
Oluwatayo Yetunde Kolawole
JRJoan Pujol Roig
Joan Pujol Roig
DWDaniel Warren
Daniel Warren
With mobile networks expected to support services with stringent requirements that ensure high-quality user experience, the ability to apply Feed-Forward Neural Network (FFNN) models to User Equipment (UE) use cases has become critical. Given that UEs have limited resources, running FFNNs directly on UEs is an intrinsically challenging problem. This letter proposes an optimization framework for split computing applications where an FFNN model is partitioned into multiple sections, and executed by UEs, edge- and core-located nodes to reduce the required UE computational footprint while containing the inference time. An efficient heuristic strategy for solving the optimization problem is also provided. The proposed framework is shown to be robust in heterogeneous settings, eliminating the need for retraining and reducing the UE's memory (CPU) footprint by over $33.6\%$ ($60\%$).