Learning High-Frequency Functions Made Easy with Sinusoidal Positional Encoding
Conference presentation, ICML 2024, Vienna, Austria
Presentation of our paper on ICML
Talks and presentations
Towards Effective Anomaly Detection for Open RAN
Conference presentation, Next Generation Networking, Multi-Service Networks workshop, Oxford, UK
I discuss our recent work in Open RAN anomaly detection.
AppShot: A Conditional Deep Generative Model for Synthesizing Service-Level Mobile Traffic Snapshots at City Scale
Conference talk, SCONE – the SCOttish Networking Event, Edinburgh, UK
Service-level mobile traffic data enables research studies and innovative applications with a potential to shape future service-oriented communication systems and beyond. However, real-world datasets reporting measurements at the individual service level are hard to come by due to privacy or industrial secrecy reasons. APPSHOT is a model for producing synthetic high-fidelity snapshots of the traffic generated by mobile services…
Efficient Mobile Network Drive Testing with Deep Generative Modeling
Conference, MobiUK 2023, Management School, Lancaster University, UK
The main goal of drive testing is to assess and optimize mobile network coverage, capacity, and quality of service (QoS) through measurement. This involves collecting field measurements in a controlled manner by driving or walking in a target scenario. Drive testing continues to play a key role in mobile network optimization for operators. The principal concern with traditional drive testing is that it requires manual effort to obtain measurements, making it costly and time-consuming. Several measurement tools are available to perform drive or walk testing…
GenDT: Mobile Network Drive Testing Made Efficient with Generative Modeling
Conference, ACM CoNEXT 2022, Rome, Italy
Drive testing continues to play a key role in mobile network optimization for operators but its high cost is a big concern. Alternative approaches like virtual drive testing (VDT) target device testing in the lab whereas MDT or crowdsourcing based approaches are limited by the incentives users have to participate and contribute measurements. With the aim of augmenting drive testing and significantly reducing its cost, we propose GenDT, a novel deep generative model that synthesizes high-fidelity time series of key radio network key performance indicators (KPIs)…