The ongoing automation of driving functions in cars results in the evolution of advanced driver assistance systems (ADAS) into ones capable of highly automated driving, which will in turn progress into fully autonomous, self-driving cars. To work properly, these functions first must be able to perceive the car's surroundings by such means as radar, lidar, camera, and ultrasound sensors. As the complexity of such systems increases along with the level of automation, the demands on environment sensors, including radar, grow as well. For radar performance to meet the requirements of self-driving cars, straightforward scaling of the radar parameters is not sufficient. To refine radar capabilities to meet more stringent requirements, fundamentally different approaches may be required, including the use of more sophisticated signal processing algorithms as well as alternative radar waveforms and modulation schemes. In addition, since radar is an active sensor (i.e., it operates by transmitting signals and evaluating their reflections) interference becomes a crucial issue as the number of automotive radar sensors increases. This article gives an overview of the challenges that arise for automotive radar from its development as a sensor for ADAS to a core component of self-driving cars. It summarizes the relevant research and discusses the following topics related to highperformance automotive radar systems: 1) shortcomings of the classical signal processing algorithms due to underlying fundamental assumptions and a signal processing framework that overcomes these limitations, 2) use of digital modulations for automotive radar, and 3) interference-mitigation methods that enable multiple radar sensors to coexist in conditions of increasing market penetration. The overview presented in this article shows that new paradigms arise as automotive radar transitions into a more powerful vehicular sensor, which provides a fertile research ground for further investigation.

In this work, we further develop the conformer-based metric generative adversarial network (CMGAN) model ¹ for speech enhancement (SE) in the time-frequency (TF) domain. This paper builds on our previous work but takes a more in-depth look by conducting extensive ablation studies on model inputs and architectural design choices. We rigorously tested the generalization ability of the model to unseen noise types and distortions. We have fortified our claims through DNS-MOS measurements and listening tests. Rather than focusing exclusively on the speech denoising task, we extend this work to address the dereverberation and super-resolution tasks. This necessitated exploring various architectural changes, specifically metric discriminator scores and masking techniques. It is essential to highlight that this is among the earliest works that attempted complex TF-domain super-resolution. Our findings show that CMGAN outperforms existing state-of-the-art methods in the three major speech enhancement tasks: denoising, dereverberation, and super-resolution. For example, in the denoising task using the Voice Bank+DEMAND dataset, CMGAN notably exceeded the performance of prior models, attaining a PESQ score of 3.41 and an SSNR of 11.10 dB. Audio samples and CMGAN implementations are available online ² .

A fundamental drawback of the orthogonal frequency division multiplexing (OFDM), both in communication and radar, is its sensitivity to the Doppler effect. In this paper, we present a novel signal processing approach for OFDM radar and communication systems that overcomes the Doppler sensitivity of OFDM. We propose a scenario independent Doppler correction method that enables an inter-carrier-interference (ICI) free processing for OFDM. The Doppler robustness of the proposed approach opens new perspectives for system parametrization, enabling radar concepts not feasible before. We show in simulations that the proposed Doppler correction method is superior to the classical signal processing in many important aspects. Measurements with an OFDM-MIMO radar prototype are used to validate the proposed approach and to show its effectiveness in real life applications.

The concept of biological age (BA) - although important in clinical practice - is hard to grasp mainly due to the lack of a clearly defined reference standard. For specific applications, especially in pediatrics, medical image data are used for BA estimation in a routine clinical context. Beyond this young age group, BA estimation is mostly restricted to whole-body assessment using non-imaging indicators such as blood biomarkers, genetic and cellular data. However, various organ systems may exhibit different aging characteristics due to lifestyle and genetic factors. Thus, a whole-body assessment of the BA does not reflect the deviations of aging behavior between organs. To this end, we propose a new imaging-based framework for organ-specific BA estimation. In this initial study we focus mainly on brain MRI. As a first step, we introduce a chronological age (CA) estimation framework using deep convolutional neural networks (Age-Net). We quantitatively assess the performance of this framework in comparison to existing state-of-the-art CA estimation approaches. Furthermore, we expand upon Age-Net with a novel iterative data-cleaning algorithm to segregate atypical-aging patients (BA \not ≈ CA) from the given population. We hypothesize that the remaining population should approximate the true BA behavior. We apply the proposed methodology on a brain magnetic resonance image (MRI) dataset containing healthy individuals as well as Alzheimer's patients with different dementia ratings. We demonstrate the correlation between the predicted BAs and the expected cognitive deterioration in Alzheimer's patients. A statistical and visualization-based analysis has provided evidence regarding the potential and current challenges of the proposed methodology.

Physiological motion, such as cardiac and respiratory motion, during Magnetic Resonance (MR) image acquisition can cause image artifacts. Motion correction techniques have been proposed to compensate for these types of motion during thoracic scans, relying on accurate motion estimation from undersampled motion-resolved reconstruction. A particular interest and challenge lie in the derivation of reliable non-rigid motion fields from the undersampled motion-resolved data. Motion estimation is usually formulated in image space via diffusion, parametric-spline, or optical flow methods. However, image-based registration can be impaired by remaining aliasing artifacts due to the undersampled motion-resolved reconstruction. In this work, we describe a formalism to perform non-rigid registration directly in the sampled Fourier space, i.e. k-space. We propose a deep-learning based approach to perform fast and accurate non-rigid registration from the undersampled k-space data. The basic working principle originates from the Local All-Pass (LAP) technique, a recently introduced optical flow-based registration. The proposed LAPNet is compared against traditional and deep learning image-based registrations and tested on fully-sampled and highly-accelerated (with two undersampling strategies) 3D respiratory motion-resolved MR images in a cohort of 40 patients with suspected liver or lung metastases and 25 healthy subjects. The proposed LAPNet provided consistent and superior performance to image-based approaches throughout different sampling trajectories and acceleration factors.