Secrecy Rate Beamforming for Multicell Networks With Information and Energy Harvesting

Ali A. Nasir, Hoang Duong Tuan, Trung Q. Duong, H. Vincent Poor

Abstract

Considering a multicell network for the secure wireless information and power transfer, this paper studies the joint design of transmit beamformers at the base stations (BSs) and receive signal splitting ratios at the end users' equipment (UE). The primary concern in this work is the network internal security, where there may be a single multiantenna eavesdropper or there is a risk that any near user may accidentally eavesdrop on the received signal of any far user. The objective is to maximize the minimum secrecy user rate under BS transmit power and UE minimum harvested energy constraints. New path-following algorithms are proposed for computational solutions of these difficult nonconvex optimization problems. Each iteration involves one simple convex quadratic program. Numerical results confirm that the proposed algorithms converge quickly after few iterations having a low computational complexity.

Introduction

As dense 5G and Internet of Things (IoT) deployments become more prevalent, two critical challenges emerge: energy sustainability and data security. Simultaneous Wireless Information and Power Transfer (SWIPT) addresses the former by allowing devices to split incoming radio signals for both data decoding and battery charging. However, the broadcast nature of wireless signals makes them susceptible to interception. Physical Layer Security (PLS) offers a lightweight alternative to traditional encryption by leveraging channel characteristics to ensure secrecy.

This paper investigates these challenges within a multicell network environment where inter-cell interference is a significant factor. The study focuses on two primary security threats:

  1. External Eavesdropping (Eve Model 1): A dedicated, multi-antenna malicious eavesdropper.
  2. Internal Security (Eve Model 2): A "near" user, typically positioned for high energy harvesting, potentially intercepting signals intended for "far" users.

System Model and Problem Formulation

The network consists of multiple base stations serving users categorized into different zones based on their proximity and signal requirements.

Multicell network model with Zone 1 (near) and Zone 2 (far) users.

The objective is to maximize the worst-case secrecy rate across all users while satisfying constraints on base station transmit power and minimum harvested energy for each device. This results in a complex, non-convex optimization problem that is difficult to solve using standard techniques like Semi-Definite Programming (SDP) relaxation.

Proposed Methodology

The authors propose a novel Path-Following Iterative Algorithm to solve the joint optimization of transmit beamforming and user power-splitting ratios. Key features include:

  • Successive Convex Approximation: The non-convex problem is approximated by a sequence of convex quadratic programs (QPs), which are computationally efficient and scalable.
  • Joint Optimization: Unlike traditional methods that optimize parameters separately, this approach simultaneously tunes base station beamformers and user hardware (splitting ratios).
  • Artificial Noise (AN): Base stations can inject AN specifically designed to degrade the eavesdropper's channel without impacting legitimate users.

Performance Evaluation

The algorithm was evaluated using Monte Carlo simulations in a 3-cell network with 4 users per cell. The channel model incorporated Rician fading to account for line-of-sight components.

Key findings include:

  • Fast Convergence: The algorithm typically converges within 12 to 18 iterations, demonstrating low computational complexity and viability for real-time management.
  • Energy-Security Trade-off: The results highlight a fundamental conflict where stronger signals for energy harvesting increase eavesdropping risks; the algorithm successfully identifies the optimal "sweet spot."
  • Internal vs. External Threats: Interestingly, the "Internal Security" scenario often allows for higher secrecy rates because single-antenna internal users are easier to mitigate via beamforming than multi-antenna external eavesdroppers.

Conclusion

The proposed path-following approach provides a robust and efficient solution for secure SWIPT in multicell networks. Future research directions include relaxing the assumption of perfect Channel State Information (CSI) for passive eavesdroppers and extending the model to accommodate user mobility and massive MIMO configurations.

Reference: Duong, Q. (2017). Secrecy Rate Beamforming for Multicell Networks With Information and Energy Harvesting. IEEE Transactions on Signal Processing, 65(3), 677-689. https://doi.org/10.1109/TSP.2016.2621719

Summary updated

Abstract

Considering a multicell network for the secure wireless information and power transfer, this paper studies the joint design of transmit beamformers at the base stations (BSs) and receive signal splitting ratios at the end users' equipment (UE). The primary concern in this work is the network internal security, where there may be a single multiantenna eavesdropper or there is a risk that any near user may accidentally eavesdrop on the received signal of any far user. The objective is to maximize the minimum secrecy user rate under BS transmit power and UE minimum harvested energy constraints. New path-following algorithms are proposed for computational solutions of these difficult nonconvex optimization problems. Each iteration involves one simple convex quadratic program. Numerical results confirm that the proposed algorithms converge quickly after few iterations having a low computational complexity.
BibTeX
@article{nasir2016secrecy,
  author = {Ali A. Nasir and Hoang Duong Tuan and Trung Q. Duong and H. Vincent Poor},
  title = {{Secrecy Rate Beamforming for Multicell Networks With Information and Energy Harvesting}},
  journal = {IEEE Transactions on Signal Processing},
  year = {2016},
  doi = {10.1109/tsp.2016.2621719},
}
Figure