Interference Removal for Radar/Communication Co-Existence: The Random Scattering Case

Yinchuan Li, Le Zheng, Marco Lops, Xiaodong Wang

TL;DR

This paper tackles the challenge of radar interference in communication systems operating in shared frequency bands. The authors propose two novel algorithms, one based on Atomic Norm minimization and another a two-stage alternating minimization approach, to effectively remove radar interference and improve communication reliability, with the latter showing superior performance and efficiency.

What Problem Does This Paper Address?

As wireless communication systems demand higher data rates, they increasingly share frequency bands with radar systems. This co-existence, especially in un-coordinated scenarios, leads to significant interference at the communication receiver, primarily from radar reverberation off various scatterers. This interference can severely degrade communication performance, even preventing reliable data demodulation. Existing methods often assume coordination or prior knowledge of the radar signal, which isn't always feasible. This paper focuses on mitigating this interference at the communication receiver, specifically addressing the complex scenario where the radar signal is not synchronized and its waveform is unknown, and the interference is caused by random scattering.

Methods

The core of the paper lies in developing algorithms to solve a non-convex problem of joint interference removal and data demodulation. The authors leverage the sparsity of both the radar interference and the communication data demodulation errors.

  • Algorithm 1 (Atomic Norm Minimization): This approach uses a convex relaxation of the problem by employing the Atomic Norm (AN) to model the sparse nature of the radar interference, particularly its continuous delays. It aims to find a tight convex relaxation for enforcing sparsity in the radar signal components.
  • Algorithm 2 (Two-Stage Alternating Minimization): This method adopts a two-stage strategy. The first stage iteratively estimates the radar parameters and communication errors using alternating minimization. The second stage then refines this solution by inferring the global optimum in a higher-dimensional space, using the first stage's output as an initialization. This approach is designed to handle the off-grid nature of delays more effectively and achieve better performance.

Datasets

Not discussed in the paper. The paper focuses on theoretical development and simulation-based evaluation.

Key Insights and Ideas

  • The "off-grid" problem is critical: Traditional compressed sensing techniques struggle with continuous parameters like radar delays. The paper highlights the need for specialized methods like Atomic Norm minimization to handle this.
  • Sparsity is a double-edged sword: While sparsity in radar scatterers and demodulation errors is exploited, the non-convexity of the resulting optimization problem is a major hurdle.
  • Two-stage approach for global optimum: The paper's key insight is that a local optimum found through alternating minimization can be a good starting point for finding the global optimum in a higher-dimensional space, significantly improving performance.
  • Radar interference is a significant challenge: Even with moderate interference-to-signal ratios (ISRs), radar signals can substantially degrade communication performance, underscoring the need for effective mitigation techniques.

Key Results

  • Improved Symbol Error Rate (SER): Both proposed algorithms significantly outperform the baseline "Iteration 0" (direct demodulation) and the conventional on-grid method in terms of SER.
  • 2-AltMin's Superiority: The two-stage alternating minimization (2-AltMin) algorithm consistently achieves better SER performance than the Atomic Norm-based method and the on-grid method across various SNR and ISR conditions.
  • Global Optimum Achievement: The 2-AltMin method demonstrates a high probability of reaching the global optimum, unlike the on-grid and CR methods which often get stuck in local optima.
  • Computational Efficiency: The 2-AltMin method is substantially faster than the CR and on-grid methods, making it more suitable for real-time applications. For example, it takes only 2.3 seconds with 12 iterations, compared to the on-grid method's 1616.3 seconds with 4 iterations.
  • Accuracy in Delay and Code Estimation: The 2-AltMin method also shows much better accuracy in estimating radar delays and codes, with significantly lower mean absolute errors (MAE) and mean-squared errors (MSE).

Limitations and Future Work

  • Computational Complexity: While the 2-AltMin method is computationally efficient, the paper notes that the convex relaxation (CR) method has a higher complexity per iteration. The complexity of the proposed algorithms is analyzed in terms of iterations and operations.
  • Spurious Scatterers: The paper acknowledges that relaxing the original problem can lead to spurious scatterers in the relaxed problem, which the 2-AltMin method aims to mitigate through its two-stage approach.
  • Future Work: The paper implicitly suggests further exploration of the theoretical underpinnings of the algorithms and their application in more complex and dynamic scenarios. The success of the 2-AltMin method in achieving both performance and efficiency opens avenues for its practical deployment.

Summary updated

Abstract

In this paper, we consider an un-cooperative spectrum sharing scenario, where a radar system is to be overlaid to a pre-existing wireless communication system. Given the order of magnitude of the transmitted powers in play, we focus on the issue of interference mitigation at the communication receiver. We explicitly account for the reverberation produced by the (typically high-power) radar transmitter whose signal hits scattering centers (whether targets or clutter) producing interference onto the communication receiver, which is assumed to operate in an un-synchronized and un-coordinated scenario. We first show that the receiver design amounts to solve a joint (non-convex) interference removal and data demodulation problem. Next, we introduce two algorithms exploiting sparsity of a proper representation of the interference and the vector containing demodulation errors of the data block. The first algorithm is basically a relaxed constrained atomic norm minimization, while the latter relies on a two-stage processing structure and is based on alternating minimization. The merits of these algorithms are demonstrated through extensive simulations; interestingly, the two-stage alternating minimization algorithm turns out to achieve satisfactory performance with moderate computational complexity.
BibTeX
@article{li2019interference,
  author = {Yinchuan Li and Le Zheng and Marco Lops and Xiaodong Wang},
  title = {{Interference Removal for Radar/Communication Co-Existence: The Random Scattering Case}},
  journal = {IEEE Trans. Wireless Communications},
  year = {2019},
  url = {https://openreview.net/forum?id=dgdt48Sej9F},
}

Loading related content...

Figure