Personalized Federated Learning for Cellular VR: Online Learning and Dynamic Caching
Krishnendu S. Tharakan, Hayssam Dahrouj, Nour Kouzayha, Hesham ElSawy, Tareq Y. Al-Naffouri · 2025
Read Personalized Federated Learning for Cellular VR: Online Learning and Dynamic Caching by Krishnendu S. Tharakan, Hayssam Dahrouj, Nour Kouzayha, Hesham ElSawy, Tareq Y. Al-Naffouri (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2501-11745.txt.
Description
Krishnendu Tharakan and four co-authors designed a caching scheme that preloads what virtual reality users are about to see, in a 2025 paper. Wireless VR frees users to move, but the stream must stay fast and sharp in real time. Each user's field of view can be stored ahead of time at nearby base stations.
Their scheme learns what to cache with a federated method that keeps training at each base station and tailors the result to its users. The method carries a correctness guarantee on its average hit rate. A one-bit shortcut for sharing updates cuts traffic, with a proven shrink rate for the learning error. View groups are split into shared or single streams by demand. On real head-tracking data, the scheme cut delay and raised hit rates over baselines.
Catalog record
| CREATOR | Krishnendu S. Tharakan, Hayssam Dahrouj, Nour Kouzayha, Hesham ElSawy, Tareq Y. Al-Naffouri |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 92 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2501-11745