Joint Optimization of Transmit Waveform and Receive Filter for Target Detection in MIMO Radar

Hao Zheng, Bo Jiu, Hongwei LiuIEEE Access 2019 · DOI

Abstract

Effective target detection in Multiple-Input Multiple-Output (MIMO) radar systems is often hampered by signal-dependent interference and hardware-imposed peak-to-average ratio (PAR) constraints. This paper presents a novel framework for the joint optimization of transmit waveforms and receive filters. By maximizing the signal-to-interference-plus-noise ratio (SINR) and minimizing the integrated sidelobe level (ISL), the proposed method enhances detection performance in cluttered environments. The authors utilize a minorization-maximization (MM) approach to solve the resulting non-convex optimization problem, demonstrating significant improvements in sidelobe suppression and computational efficiency.

Introduction

In modern radar applications, the ability to distinguish a target from environmental clutter—such as reflections from the ground or sea—is paramount. This "signal-dependent interference" can obscure targets and lead to false alarms. MIMO radar systems offer increased degrees of freedom to combat this, but they require sophisticated design of both the transmitted signal and the receive processing chain. Two primary metrics guide this design: SINR, which determines detection sensitivity, and ISL, which measures the precision of the radar pulse by quantifying energy leakage into adjacent range cells.

Problem Statement

A significant challenge in radar engineering is that SINR and ISL are often optimized in isolation, or without regard for the practical limitations of transmit amplifiers. High-power amplifiers operate most efficiently when the transmit signal has a low Peak-to-Average Ratio (PAR). This paper addresses these challenges by formulating a joint optimization problem that balances SINR, ISL, and PAR constraints within a unified framework.

Proposed Methodology

The researchers propose a specialized signal processing architecture that decouples the space-time receive filter into two distinct components: a spatial receive beamformer and a temporal filter. This structure is illustrated in Figure 1.

Architecture of the proposed space-time receive filter for MIMO radar target detection.

To navigate the non-convex optimization landscape, the authors developed two main algorithms:

  • Joint Optimization Algorithm (JOA): An iterative scheme that sequentially updates the waveform and filter components. It leverages the Minorization-Maximization (MM) framework, which simplifies complex global optimization by solving a series of easier surrogate sub-problems.
  • JOA for Special Case (JOASC): A streamlined version designed for speed, where the temporal filter is fixed to match the mainlobe signal.

A key conceptual breakthrough in their approach is the focus on "Mainlobe Dominance," which simplifies the mathematical requirements for ISL optimization without sacrificing detection performance.

Results and Discussion

The proposed methods were validated through numerical simulations of a 6x6 MIMO radar system. The results highlighted several key advantages:

  • Superior Sidelobe Suppression: The JOA method achieved peak sidelobe levels below -20 dB, significantly outperforming baseline methods by fully exploiting the temporal filter's degrees of freedom.
  • The PAR Trade-off: The study demonstrated that allowing for slight variations in waveform amplitude (relaxing the PAR constraint) provides the algorithm with more flexibility to suppress interference and maximize SINR.
  • Computational Efficiency: By incorporating the SQUAREM acceleration scheme, the A-JOASC algorithm achieved performance comparable to state-of-the-art semi-definite programming (SDP) methods while being orders of magnitude faster.
  • Robustness: The algorithms proved resilient to minor errors in the estimated spatial location of interference sources, a critical requirement for real-world deployment.

Conclusion

This research provides a robust framework for the joint design of MIMO radar components, offering a path toward more sensitive and precise target detection in complex environments. While the current focus is on Integrated Sidelobe Levels, future work will address Peak Sidelobe Level (PSL) constraints to further reduce false alarm rates. The transition from simulated environments to real-world experimental validation remains the next major milestone for this technology.

Summary updated

BibTeX
@article{zheng2019joint,
  author = {Hao Zheng and Bo Jiu and Hongwei Liu},
  title = {{Joint Optimization of Transmit Waveform and Receive Filter for Target Detection in MIMO Radar}},
  journal = {IEEE Access},
  year = {2019},
  doi = {10.1109/access.2019.2960865},
}
Figure