@article{3024004,
    title = "Capacity analysis of power beacon-assisted energy harvesting mimo system over κ - μ shadowed fading channels",
    author = "Yang, J. and Wu, X. and Peppas, K.P. and Takis Mathiopoulos, P.",
    journal = "IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY",
    year = "2021",
    volume = "70",
    number = "11",
    pages = "11869-11880",
    publisher = "Institute of Electrical and Electronics Engineers, Inc. (IEEE)",
    issn = "0018-9545",
    doi = "10.1109/TVT.2021.3116190",
    keywords = "Channel capacity;  Eigenvalues and eigenfunctions;  Energy efficiency;  Energy harvesting;  Energy transfer;  Fading (radio);  Intelligent systems;  MIMO systems;  Monte Carlo methods;  Multipath fading;  Probability density function;  Signal to noise ratio, Channel-state information;  Energy;  Ergodic capacity;  Evaluation results;  Multi-input multi-output;  No channel state information;  Performances evaluation;  Power;  Power beacon-assisted;  Κ-μ shadowed channel, Channel state information",
    abstract = "In this paper, novel ergodic capacity (EC) performance evaluation results of a power beacon (PB)-assisted multiinput multi-output (MIMO) wireless powered communication network are presented. In the considered system, the energy harvesting node harvests energy from the radio-frequency signals sent by the dedicated PB and uses this energy to communicate with the destination node. To accurately model the combined effect of multi-path fading and shadowing, it is assumed that the energy transfer link is subject to κ-μ shadowed fading. Performance evaluation results are presented for two cases, depending upon the availability of channel state information (CSI) at the PB, namely, no CSI and full CSI. In the former case, equal power allocation is assumed, whereas, in the later case, energy beamforming is employed to increase energy transfer efficiency. For the performance evaluation of EC under full CSI, a closed-form approximation for the probability density function of the maximum eigenvalue of a κ-μ shadowed distributed random matrix is derived. For both no CSI and full CSI cases, lower and upper bounds on the achievable EC are derived in closed-form. Moreover, in order to obtain further insights on the impact of key parameters on the system performance, asymptotic EC expressions which become very tight at low- and high-signal-to-noise ratio regimes, are obtained. Using the proposed EC lower bound as well as these asymptotic results, simple closed-form expressions for the optimal time split that maximize the achievable EC are derived. Numerically evaluated results accompanied with Monte-Carlo simulations are further presented to corroborate the theoretical analysis. © 2021 Institute of Electrical and Electronics Engineers Inc.. All rights reserved."
}