[Paper Review] Outage Constrained Robust Transmit Optimization for Multiuser MISO Downlinks: Tractable Approximations by Conic Optimization

Kunyu Wang, Anthony Man–Cho So+3 more|arXiv (Cornell University)|Aug 4, 2011

Advanced MIMO Systems Optimization17 citations

TL;DR

This paper proposes a novel relaxation-restriction (RAR) framework for robust transmit beamforming in multiuser MISO downlinks with imperfect CSI, using semidefinite relaxation and probabilistic approximations to convert chance-constrained SINR problems into tractable convex conic programs. The key contribution is three new conservative approximation methods—sphere bounding, Bernstein-type inequality, and decomposition—that significantly improve solution quality and computational efficiency over existing robust beamforming techniques.

ABSTRACT

In this paper we consider a probabilistic signal-to-interference and-noise ratio (SINR) constrained problem for transmit beamforming design in the presence of imperfect channel state information (CSI), under a multiuser multiple-input single-output (MISO) downlink scenario. In particular, we deal with outage-based quality-of-service constraints, where the probability of each user's SINR not satisfying a service requirement must not fall below a given outage probability specification. The study of solution approaches to the probabilistic SINR constrained problem is important because CSI errors are often present in practical systems and they may cause substantial SINR outages if not handled properly. However, a major technical challenge is how to process the probabilistic SINR constraints. To tackle this, we propose a novel relaxation- restriction (RAR) approach, which consists of two key ingredients-semidefinite relaxation (SDR), and analytic tools for conservatively approximating probabilistic constraints. The underlying goal is to establish approximate probabilistic SINR constrained formulations in the form of convex conic optimization problems, so that they can be readily implemented by available solvers. Using either an intuitive worst-case argument or specialized probabilistic results, we develop various conservative approximation schemes for processing probabilistic constraints with quadratic uncertainties. Consequently, we obtain several RAR alternatives for handling the probabilistic SINR constrained problem. Our techniques apply to both complex Gaussian CSI errors and i.i.d. bounded CSI errors with unknown distribution. Moreover, results obtained from our extensive simulations show that the proposed RAR methods significantly improve upon existing ones, both in terms of solution quality and computational complexity.

Motivation & Objective

• Address the challenge of designing robust transmit beamformers in multiuser MISO downlinks when channel state information (CSI) is imperfect, leading to potential SINR outages.
• Formulate a probabilistic SINR constraint where the outage probability for each user’s QoS must not exceed a specified threshold, ensuring reliable service under CSI uncertainty.
• Overcome the computational intractability of chance-constrained beamforming problems by developing convex approximations that are efficiently solvable via standard conic optimization solvers.
• Develop and compare multiple conservative approximation schemes for quadratic uncertainties arising from CSI errors, applicable to both Gaussian and bounded i.i.d. error models.
• Demonstrate that the proposed RAR methods achieve better trade-offs between solution quality and computational complexity than existing robust beamforming approaches.

Proposed method

• Introduce a relaxation-restriction (RAR) framework that combines semidefinite relaxation (SDR) with analytic tools to approximate probabilistic SINR constraints as convex conic programs.
• Apply three distinct approximation techniques: sphere bounding, Bernstein-type inequality, and decomposition-based methods, to handle quadratic uncertainties in the chance constraints.
• Use worst-case arguments and specialized probabilistic inequalities to conservatively approximate the probability that SINR falls below a threshold, ensuring compliance with outage constraints.
• Formulate the resulting robust beamforming problem as a second-order cone program (SOCP) or semidefinite program (SDP), enabling efficient solution via standard interior-point solvers.
• Integrate a bisection search procedure to fine-tune design parameters and reduce conservatism, improving the actual SINR satisfaction probability toward the target outage level.
• Validate the feasibility and performance of the solutions using Monte Carlo-based procedures to assess the true outage probability of the beamforming designs.

Experimental results

Research questions

• RQ1 How can probabilistic SINR constraints with quadratic uncertainties due to CSI errors be efficiently approximated in a convex optimization framework?
• RQ2 What are the most effective conservative approximation techniques for chance-constrained beamforming problems in MISO downlinks with imperfect CSI?
• RQ3 How do different approximation methods (sphere bounding, Bernstein-type, decomposition) compare in terms of solution quality and computational complexity?
• RQ4 Can the proposed RAR framework achieve better performance than existing worst-case or probabilistic robust beamforming methods in terms of transmit power and outage control?
• RQ5 To what extent can a bisection-based refinement procedure reduce the conservatism of the RAR approximations while maintaining computational tractability?

Key findings

• RAR Method II, based on the Bernstein-type inequality, achieves performance comparable to bisection-aided methods without requiring iterative refinement, significantly reducing computational cost.
• RAR Method IV, designed for i.i.d. bounded CSI errors with unknown distribution, outperforms the probabilistic SOCP method under uniform CSI errors, especially at higher SINR requirements.
• The ratio of rank-one solutions across all RAR methods is consistently high—over 95% for all tested SNR levels—indicating that the SDR relaxation is nearly exact in practice.
• Extensive simulations show that the proposed RAR methods significantly reduce transmit power while maintaining the target outage probability, outperforming existing robust beamforming techniques in both solution quality and computational efficiency.
• The bisection refinement scheme effectively reduces conservatism, bringing the actual SINR outage probability closer to the target $1 - \rho$, though RAR Method II already performs well without it.
• The proposed framework is applicable to both complex Gaussian and i.i.d. bounded CSI errors, demonstrating broad applicability across different practical error models.