Robust Secrecy Energy Efficiency Optimization for Wireless Powered Heterogeneous Networks Using Distributed ADMM Algorithm

Bo Zhang,Zhou Zhong,Bin Li,Kaizhi Huang,Lijian Zhang,Zesong Fei

doi:10.1109/access.2019.2935798

Abstract

Under the case of imperfect channel state information (CSI), this paper investigates the robust secrecy energy efficiency (SEE) optimization for the heterogeneous networks (HeNets) supported by simultaneous wireless information and power transfer. Specifically, we first consider a two-tier HeNet composed of a macrocell base station (MBS) and several femtocell base stations (FBSs), where the MBS serves multiple macrocell users (MUs) while each FBS serves an information receiver (IR) and an multiple-antenna energy receiver (ER). Meanwhile, a malicious multiple-antenna eavesdropper (Eve) attempts to wiretap the downlink information of MUs and the ER acts as a potential Eve to eavesdrop the confidential information for IR in the same femtocell. To enhance the secrecy performance, artificial noise (AN) is injected into the downlink information beams of the MBS and FBSs. By considering the user fairness, the problem of SEE maximization of the whole network is formulated via a cross-tier multi-cell coordinated beamforming design. The resulting problem contains infinite constraints caused by CSI error, which is nonconvex and cannot be solved directly. To this regard, we resort to the successive convex approximation, S-procedure and semi-definite relaxation techniques to obtain a solvable form of it. Furthermore, to reduce the overhead of information exchange among coordinated BSs, we develop a distributed solution based on alternative direction multiplier method (ADMM) that can achieve a good approximation performance. Finally, simulation results verify the validity of the proposed AN-aided cross-tier multi-cell coordinated beamforming design and distributed ADMM-based design.

Highlights

We consider that two coordinated femtocell base stations (FBSs) are deployed under the coverage of the macrocell base station (MBS), where two macrocell users (MUs) and an Eve are randomly distributed in a macrocell with a radius of 300m while an information receiver (IR) and an energy receiver (ER) are scattered randomly in each femtocell with a radius of 100m
With the increase of ε, the secrecy energy efficiency (SEE) performance of them all tends to decrease. This can be explained that the deviation of downlink beamforming and artificial noise (AN) at the MBS and FBSs occurs due to the existence of channel state information (CSI) error, which will surely bring about the waste of transmit power and the decrease of secrecy performance
Under the same parameters setting, we find that our proposed design always has the most SEE performance while zeroforce AN design is second to it

Summary

Introduction

A. BACKGROUND Since the dramatic increase of Internet-enabled smart devices (e.g., smart phones and electronic tablets) has spurred the explosive growth of high-rate multimedia wireless services, it becomes difficult for mobile operators to utilize conventional homogeneous networks to achieve higher capacity and. The associate editor coordinating the review of this article and approving it for publication was Yuan Gao. coverage for next-generation 5G wireless communications. Increasing cell density for higher spatial spectrum reuse is one viable solution to do so. In this regard, heterogeneous networks (HeNets) have been proposed as a promising network densification architecture, where different types of overlaid small cells are deployed under the coverage of conventional macrocell and share the same spectrum resource [1], [2].

Objectives

Results

Conclusion