Secrecy Rate Performance of Cache-enabled Millimeter Wave Cellular Networks

Abstract

Secure transmission of cached files in wireless networks is crucial to protect the transmitted files from interception by illegitimate nodes. This paper analyzes the secrecy rate performance of a cache-enabled millimeter wave (mmWave) cellular network in the presence of colluding eavesdroppers. The average cache hit probability is computed and the stochastic geometry framework is used to evaluate the average secrecy rate of the network, taking into consideration directional beamforming and Nakagami-m fading. Moreover, the size of the available files is modeled by a Pareto distribution based on existing studies, while a memory and file size-aware caching (MFC) scheme is proposed. The MFC scheme incorporates an optimized memory allocation algorithm that reserves memory blocks in the cache of a base station (BS) based on the frequency of the requested files, the parameters of the Pareto file size distribution, and the network design parameters. The numerical results show that the proposed MFC scheme achieves up to two-fold gain in the average secrecy rate compared to the state-of-the-art probabilistic and most-popular-content (MPC) caching schemes. Furthermore, the impact of the skew exponent and the colluding eavesdroppers’ intensity on the average cache hit probability and secrecy rate, respectively, is investigated.