Projects per year
Abstract
We consider a hybrid streaming scheme based on cache-enabled orthogonal multipoint multicast (OMPMC) and on-demand single-point unicast (SPUC) transmission. The network contains two types of nodes, cache-equipped helper nodes (HNs) handling content-centric OMPMC, and cellular base stations (BSs) handling user-centric SPUC. The OMPMC service streams cached files across the network. Users whose demands cannot be satisfied by OMPMC, either because of poor signal quality or because the requested file is not cached at HNs, are served by SPUC; requested files are fetched from the core network and unicast to users using group-specific beamforming transmissions. We consider the overall network radio resource consumption to satisfy the users' requests for a given outage probability. This yields a parametric constrained optimization problem over the cache and resource allocations of the OMPMC component, as well as the multi-user beamforming scheme of the SPUC component. We devise a surface-following approach on the basis of path-following method to find the optimal traffic streaming solution. Simulation results show that the hybrid scheme provides a more promising trade-off between resource consumption and service outage probability, compared to OMPMC-only and SPUC-only alternatives.
Original language | English |
---|---|
Pages (from-to) | 7833-7848 |
Number of pages | 16 |
Journal | IEEE Transactions on Wireless Communications |
Volume | 23 |
Issue number | 7 |
Early online date | 28 Dec 2023 |
DOIs | |
Publication status | Published - Jul 2024 |
MoE publication type | A1 Journal article-refereed |
Keywords
- Array signal processing
- Hybrid content streaming
- multipoint multicasting
- Power system reliability
- Probabilistic logic
- Probability
- Resource management
- single-point unicasting
- Streams
- surface following approach
- Unicast
- user group-specific beamforming
- wireless caching
Fingerprint
Dive into the research topics of 'Optimal Multicast-Cache-Aided On-demand Streaming in Heterogeneous Wireless Networks via a Path/Surface Following Approach'. Together they form a unique fingerprint.Projects
- 1 Finished
-
RILREW: Reinforcement Learning for Real-time Wireless Scheduling and Edge Caching: Theory and Algorithm Design
Tirkkonen, O. (Principal investigator)
01/01/2022 → 31/12/2024
Project: Academy of Finland: Other research funding