Robust Beamforming for SWIPT System with Chance Constraints

Yinglei Teng, Wanxin Zhao, Mei Yan, Yong Zhang, Mei SongICC 2018 · OpenReview

TL;DR

This paper tackles the challenge of maximizing data rates in wireless systems that simultaneously transmit information and power (SWIPT). It introduces a novel approach using "chance constraints" to ensure users receive a minimum amount of harvested energy, even with imperfect knowledge of the wireless channel, and proposes an optimization algorithm to achieve these goals.

What Problem Does This Paper Address?

Modern wireless networks face a dual challenge: the ever-increasing demand for data transmission and the growing concern for energy efficiency. Simultaneous Wireless Information and Power Transfer (SWIPT) offers a promising solution by allowing devices to harvest energy from radio frequency (RF) signals while simultaneously receiving data. However, designing systems that effectively balance these two objectives is complex. Specifically, this paper addresses the problem of robust beamforming in a MISO (Multiple-Input Single-Output) SWIPT system. The key challenges are:

  • Imperfect Channel State Information (CSI): In real-world wireless environments, the exact channel conditions are never perfectly known. This uncertainty makes it difficult to design optimal transmission strategies.
  • Energy Harvesting Guarantees: Users need a minimum amount of harvested energy to operate. However, traditional deterministic constraints for energy harvesting can be overly strict and lead to suboptimal data rates, especially when the exact energy requirement isn't critical at every single moment.
  • Maximizing Sum Rate: The ultimate goal is to maximize the total data rate across all users in the system.

The paper focuses on the "time switching" (TS) method for SWIPT, where a portion of time is dedicated to energy harvesting and the remaining portion to information decoding.

Methods

The authors propose a multi-pronged approach to tackle the complex optimization problem:

  • Chance Constraints for Energy Harvesting: Instead of requiring a fixed minimum energy to be harvested at all times, the paper introduces "chance constraints." This means that the system guarantees a minimum energy harvest with a certain probability (e.g., 99% of the time). This is more practical as it acknowledges that occasional dips in harvested energy might be acceptable.
  • Bernstein Inequality: To handle these probabilistic constraints, the paper leverages the Bernstein inequality. This mathematical tool allows them to transform the chance constraints into deterministic (guaranteed) constraints, making the problem amenable to standard optimization techniques.
  • Minimum Mean Square Error (MMSE) for Sum Rate: The problem of maximizing the sum rate with imperfect CSI is inherently non-convex. The paper cleverly transforms this into an equivalent problem of minimizing the expected Minimum Mean Square Error (MMSE) for the receivers. This transformation is based on the understanding that maximizing the sum rate and minimizing the expected MMSE are closely related.
  • Alternative Optimization (AO) Algorithm: To solve the overall non-convex optimization problem, an AO algorithm is proposed. This algorithm iteratively solves two sub-problems:
  • Transmit Beamformer Design: Optimizing the beamforming vectors (which direct the transmitted signals) for both information and energy transmission.
  • Time Switching Ratio Optimization: Determining the optimal fraction of time to allocate for energy harvesting versus information decoding.
  • Semidefinite Relaxation (SDR): For the beamformer design sub-problem, the paper utilizes Semidefinite Relaxation (SDR) to handle the non-convexity introduced by the quadratic terms involving beamforming vectors. This technique relaxes the problem into a convex one that can be solved efficiently.

Datasets

Not discussed in the paper. The evaluation is based on simulated network scenarios.

Key Insights and Ideas

  • The Power of Probabilistic Guarantees: Using "chance constraints" for energy harvesting is a significant improvement over deterministic constraints. It allows for higher data rates by providing flexibility in energy delivery, acknowledging that perfect energy delivery every single moment might not be necessary or optimal.
  • Bernstein Inequality as a Bridge: The application of the Bernstein inequality is crucial for bridging the gap between probabilistic energy harvesting requirements and solvable deterministic optimization problems.
  • MMSE as a Proxy for Sum Rate: The insight that minimizing expected MMSE is equivalent to maximizing the sum rate under imperfect CSI is a powerful simplification that enables the use of well-established techniques.
  • AO Algorithm for Practical Solutions: The proposed AO algorithm effectively decomposes a complex, coupled optimization problem into manageable sub-problems, leading to a practical solution that converges to a locally optimal performance.

Key Results

  • The proposed robust beamforming algorithm with chance constraints consistently achieves higher sum rates compared to schemes using deterministic constraints, especially as the maximum transmission power of the base station increases.
  • The algorithm demonstrates good convergence, with the sum rate gradually increasing and stabilizing over iterations.
  • The paper shows that higher energy outage probabilities (meaning a less strict requirement for energy harvesting) lead to higher achievable sum rates, highlighting the trade-off between energy guarantees and information transmission efficiency.
  • The energy consumption for energy transmission and the time switching ratio increase with the requested minimal energy, while the energy consumption for information transmission decreases.

Limitations and Future Work

  • Rank-One Constraint: While SDR is used, the paper acknowledges that the optimal solution might not inherently satisfy the rank-one constraint required for practical beamforming. However, they cite previous work and simulation results suggesting that the solution is often rank-one, and if not, approximate methods like Gaussian randomization could be employed.
  • Single-User Energy Harvesting Constraint: The paper assumes all users have the same energy requirements. Future work could explore scenarios with heterogeneous energy demands.
  • MISO System: The current work focuses on a MISO system. Extending the framework to MIMO (Multiple-Input Multiple-Output) systems would be a natural next step.
  • Channel Estimation: The paper assumes imperfect CSI but doesn't delve into the channel estimation process itself. Investigating the impact of different channel estimation techniques on the proposed robust beamforming would be valuable.

Summary updated

Abstract

The robust beamforming problem in multiple-input single-output (MISO) downlink networks of simultaneous wireless information and power transfer (SWIPT) is studied in this paper. Adopting the time switching fashion to perform energy harvesting and information decoding respectively, we aim at maximizing the sum rate under imperfect channel state information (CSI) and the chance constraints of users' harvested energy. In view of the fact that the constraints for minimal harvested energy is not necessary to meet from time to time, this paper adopts chance constraint to model it and uses the Bernstein inequality to transform it into deterministic constraints equivalently. Recognizing the maximum sum rate problem of imperfect CSI as nonconvex problem, we transform it into finding the expectation of minimum mean square error (MMSE) equivalently in this paper, and an alternative optimization (AO) algorithm is proposed to decompose the optimization problem into two sub- problems: the transmit beamformer design and the division of switching time. The simulation results show the performance gains compared to non-robust state of the art schemes.
BibTeX
@inproceedings{teng2018robust,
  author = {Yinglei Teng and Wanxin Zhao and Mei Yan and Yong Zhang and Mei Song},
  title = {{Robust Beamforming for SWIPT System with Chance Constraints}},
  booktitle = {ICC},
  year = {2018},
  url = {https://openreview.net/forum?id=250Q1k4ILD},
}
Figure