In multi-antenna cache-enabled small-cell networks, matched-filter and zero-forcing beamforming increase the advantage of probabilistic and coded caching over most-popular caching, quantified by derived and optimized formulas.
Treating Content Delivery in Multi-Antenna Coded Caching as General Message Sets Transmission: A DoF Region Perspective
1 Pith paper cite this work. Polarity classification is still indexing.
abstract
Coded caching can create coded multicasting thus significantly accelerates content delivery in broadcast channels with receiver caches. While the original delivery scheme in coded caching multicasts each coded message sequentially, it is not optimal for multiple-input multiple-output (MIMO) broadcast channels. This work aims to investigate the full spatial multiplexing gain in multi-antenna coded caching by transmitting all coded messages concurrently. In specific, we propose to treat the content delivery as the transmission problem with general message sets where all possible messages are present, each with different length and intended for different user set. We first obtain inner and outer bounds of the degrees of freedom (DoF) region of a $K$-user $(M,N)$ broadcast channel with general message sets, with $M$ and $N$ being the number of transmit and receive antennas, respectively. Then for any given set of coded messages, we find its minimum normalized delivery time (NDT) by searching the optimal DoF tuple in the DoF regions. The obtained minimum NDT is optimal at antenna configuration $\frac{M}{N} \in (0,1]\cup[K, \infty)$ and is within a multiplicative gap of $\frac{M}{N}$ to optimum at $\frac{M}{N} \in (1,K)$. Our NDT results can be evaluated for any user demand with both centralized and decentralized cache placement.
fields
cs.IT 1years
2019 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Modeling, Analysis, and Optimization of Caching in Multi-Antenna Small-Cell Networks
In multi-antenna cache-enabled small-cell networks, matched-filter and zero-forcing beamforming increase the advantage of probabilistic and coded caching over most-popular caching, quantified by derived and optimized formulas.