Conversational Video Enhancement Based on Intelligent BSR
Conversational video is becoming an increasingly popular way of meeting. In this paper, we apply intelligent buffer status report (BSR) to the uplink scheduling process to improve the resource request scheme currently used at the terminal. Specifically, we deploy an intelligent report scheduler on the terminal that enables dynamic BSR-based on buffer states. According to Table 6.1.3.1 of 3GPP 38.321 [1], we model the BSR index selection as a Markov decision process (MDP) and propose a deep reinforcement learning (DRL)-based intelligent report algorithm to solve the problem. Simulation results show that the proposed strategy can effectively reduce the buffer resource pressure and improve the service experience of conversational video.
Authors
- Lili Wan (ORCID: https://orcid.org/0000-0002-9520-5425)
- Chuanhang Yu
- Hao Wang (ORCID: https://orcid.org/0000-0002-3731-9838)
- Bing Han (ORCID: https://orcid.org/0000-0003-4584-1092)
- Gang Wu (ORCID: https://orcid.org/0000-0001-9595-527X)
- Lei Ji (ORCID: https://orcid.org/0000-0001-7753-3313)
- Kaiwen Yu (ORCID: https://orcid.org/0000-0003-0118-3593)
Institutions
- University of Electronic Science and Technology of China (CN)
- Huawei Technologies (China) (CN)
Publication Details
- Published
- 2024-11-18
- DOI
- https://doi.org/10.1109/fcn64323.2024.10984515
- Primary Topic
- Speech and Audio Processing
- Type
- article
- Field-Weighted Citation Impact
- 0.00