arXiv subjects

Mario Huemer

Publications and source records attributed to Mario Huemer.

At least 19 recordsLinked to original sources

arXiv2026

Fast Time-Domain MLE for Period Estimation of Pulse Trains

Parameter estimation of periodic pulse trains is a critical task in numerous automated sensing and diagnostic applications. While estimation in the time domain provides superior accuracy in low signal-to-noise ratio environments, its high computational complexity frequently precludes its use in real-time systems. This paper investigates algorithmic optimizations to reduce runtime by leveraging recent advancements in computing architectures. Exploiting the inherent sparsity of the signal via sparse matrix multiplication kernels yields a substantial decrease in inference time. Furthermore, by separating the dense matrix projections into sequential cross-correlation and sparse summation steps, we fundamentally reduce both runtime and memory complexity. These optimizations drastically shrink the memory footprint, making time-domain estimation feasible for large datasets in real-time settings.

eess.SP↗

arXiv2026

On the CRLB for Blind Receiver I/Q Imbalance Estimation in OFDM Systems: Efficient Computation and Closed-Form Bounds

Modern mobile communication receivers are often implemented with a direct-conversion architecture, which features a number of advantages over competing designs. A notable limitation of direct-conversion architectures, however, is their sensitivity to amplitude and phase mismatches between the in-phase and quadrature signal paths. Such in-phase and quadrature-phase (I/Q) imbalances introduce undesired image components in the baseband signal, degrading link performance -- most notably by increasing the bit-error ratio. Considerable research effort has therefore been devoted to digital techniques for estimating and mitigating these impairments. Existing approaches generally fall into two categories: data-aided methods that exploit known pilots, preambles, or training sequences, and blind techniques that operate without such prior information. For data-aided estimation, Cram\'{e}r-Rao lower bounds (CRLBs) have been established in the literature. In contrast, the derivation of a CRLB for the blind I/Q-imbalance estimation case is considerably more challenging, since the received data is random and typically non-Gaussian in the frequency domain. This work extends our earlier conference contribution, which introduced a CRLB derivation for the blind estimation of frequency-independent (FID) receiver I/Q imbalance using central limit theorem (CLT) arguments. The extensions include a computationally efficient method for calculating the bound, reducing complexity from cubic in the number of samples to linear in the fast-Fourier transform (FFT) size, along with a simplified closed-form approximation. This approximation provides new insights into the allocation dependent performances of existing estimation methods, motivating a pre-estimation filtering modification that drastically improves their estimation performance in certain scenarios.

eess.SP↗

arXiv2025

Automotive Radar Online Channel Imbalance Estimation via NLMS

Automotive radars are one of the essential enablers of advanced driver assistance systems (ADASs). Continuous monitoring of the functional safety and reliability of automotive radars is a crucial requirement to prevent accidents and increase road safety. One of the most critical aspects to monitor in this context is radar channel imbalances, as they are a key parameter regarding the reliability of the radar. These imbalances may originate from several parameter variations or hardware fatigues, e.g., a solder ball break (SBB), and may affect some radar processing steps, such as the angle of arrival estimation. In this work, a novel method for online estimation of automotive radar channel imbalances is proposed. The proposed method exploits a normalized least mean squares (NLMS) algorithm as a block in the processing chain of the radar to estimate the channel imbalances. The input of this block is the detected targets in the range-Doppler map of the radar on the road without any prior knowledge on the angular parameters of the targets. This property in combination with low computational complexity of the NLMS, makes the proposed method suitable for online channel imbalance estimation, in parallel to the normal operation of the radar. Furthermore, it features reduced dependency on specific targets of interest and faster update rates of the channel imbalance estimation compared to the majority of state-of-the-art methods. This improvement is achieved by allowing for multiple targets in the angular spectrum, whereas most other methods are restricted to only single targets in the angular spectrum. The performance of the proposed method is validated using various simulation scenarios and is supported by measurement results.

eess.SP↗

arXiv2025

Estimators and Performance Bounds for Short Periodic Pulses

In many industrial applications, signals with short periodic pulses, caused by repeated steps in the manufacturing process, are present, and their fundamental frequency or period may be of interest. Fundamental frequency estimation is in many cases performed by describing the periodic signal as a multiharmonic signal and employing the corresponding maximum likelihood estimator. However, since signals with short periodic pulses contain a large number of noise-only samples, the multiharmonic signal model is not optimal to describe them. In this work, two models of short periodic pulses with known and unknown pulse shape are considered. For both models, the corresponding maximum likelihood estimators, Fisher information matrices, and approximate Cram\'er-Rao lower bounds are presented. Numerical results demonstrate that the proposed estimators outperform the maximum likelihood estimator based on the multiharmonic signal model for low signal-to-noise ratios.

eess.SP↗

arXiv2025

Optimum and Adaptive Complex-Valued Bilinear Filters

The identification of nonlinear systems is a frequent task in digital signal processing. Such nonlinear systems may be grouped into many sub-classes, whereby numerous nonlinear real-world systems can be approximated as bilinear (BL) models. Therefore, various optimum and adaptive BL filters have been introduced in recent years. Moreover, in many applications, such as communications and radar, complex-valued (CV) BL systems in combination with CV signals may occur. Hence, in this work, we investigate the extension of real-valued (RV) BL filters to CV BL filters. First, we derive CV BL filters by applying two or four RV BL filters, and compare them with respect to their computational complexity and performance. Second, we introduce novel fully CV BL filters, such as the CV BL Wiener filter (C-BWF), the CV BL least squares (C-BLS) filter, the CV BL least mean squares (C-BLMS) filter, the CV BL normalized least mean squares (C-BNLMS) filter, and the CV BL recursive least squares (C-BRLS) filter. Finally, these filters are applied to identify CV multiple-input-single-output (MISO) systems and CV Hammerstein models.

eess.SP↗

arXiv2025

Tx and Rx IQ Imbalance Compensation for JCAS in 5G NR

Beside traditional communications, joint communications and sensing (JCAS) is gaining increasing relevance as a key enabler for next-generation wireless systems. The ability to accurately transmit and receive data is the basis for high-speed communications and precise sensing, where a fundamental requirement is an accurate in-phase (I) and quadrature-phase (Q) modulation. For sensing, imperfections in IQ modulation lead to two critical issues in the range-Doppler-map (RDM) in form of an increased noise floor and the presence of ghost objects, degrading the accuracy and reliability of the information in the RDM. This paper presents a low-complex estimation and compensation method to mitigate the IQ imbalance effects. This is achieved by utilizing, amongst others, the leakage signal, which is the direct signal from the transmitter to the receiver path, and is typically the strongest signal component in the RDM. The parameters of the IQ imbalance suppression structure are estimated based on a mixed complex-/real-valued bilinear filter approach, that considers IQ imbalance in the transmitter and the receiver of the JCAS-capable user equipment (UE). The UE uses a 5G New Radio (NR)-compliant orthogonal frequency-division multiplexing (OFDM) waveform with the system configuration assumed to be predefined from the communication side. To assess the effectiveness of the proposed approach, simulations are conducted, illustrating the performance in the suppression of IQ imbalance introduced distortions in the RDM.

eess.SP↗

arXiv2024

Relaxed Multi-Tx DDM Online Calibration

In multiple-input and multiple-output (MIMO) radar systems based on Doppler-division multiplexing (DDM), phase shifters are employed in the transmit paths and require calibration strategies to maintain optimal performance all along the radar system's life cycle. In this paper, we propose a novel family of DDM codes that enable an online calibration of the phase shifters that scale realistically to any number of simultaneously activated transmit (Tx)-channels during the calibration frames. To achieve this goal we employ the previously developed odd-DDM (ODDM) sequences to design calibration DDM codes with reduced inter-Tx leakage. The proposed calibration sequence is applied to an automotive radar data set modulated with erroneous phase shifters.

eess.SP↗

arXiv2024

SICNN: Soft Interference Cancellation Inspired Neural Network Equalizers

In recent years data-driven machine learning approaches have been extensively studied to replace or enhance traditionally model-based processing in digital communication systems. In this work, we focus on equalization and propose a novel neural network (NN-)based approach, referred to as SICNN. SICNN is designed by deep unfolding a model-based iterative soft interference cancellation (SIC) method. It eliminates the main disadvantages of its model-based counterpart, which suffers from high computational complexity and performance degradation due to required approximations. We present different variants of SICNN. SICNNv1 is specifically tailored to single carrier frequency domain equalization (SC-FDE) systems, the communication system mainly regarded in this work. SICNNv2 is more universal and is applicable as an equalizer in any communication system with a block-based data transmission scheme. Moreover, for both SICNNv1 and SICNNv2, we present versions with highly reduced numbers of learnable parameters. Another contribution of this work is a novel approach for generating training datasets for NN-based equalizers, which significantly improves their performance at high signal-to-noise ratios. We compare the bit error ratio performance of the proposed NN-based equalizers with state-of-the-art model-based and NN-based approaches, highlighting the superiority of SICNNv1 over all other methods for SC-FDE. Exemplarily, to emphasize its universality, SICNNv2 is additionally applied to a unique word orthogonal frequency division multiplexing (UW-OFDM) system, where it achieves state-of-the-art performance. Furthermore, we present a thorough complexity analysis of the proposed NN-based equalization approaches, and we investigate the influence of the training set size on the performance of NN-based equalizers.

eess.SP↗

arXiv2023

OFDM-based Waveforms for Joint Sensing and Communications Robust to Frequency Selective IQ Imbalance

Orthogonal frequency-division multiplexing (OFDM) is a promising waveform candidate for future joint sensing and communication systems. It is well known that the OFDM waveform is vulnerable to in-phase and quadrature-phase (IQ) imbalance, which increases the noise floor in a range-Doppler map (RDM). A state-of-the-art method for robustifying the OFDM waveform against IQ imbalance avoids an increased noise floor, but it generates additional ghost objects in the RDM [1]. A consequence of these additional ghost objects is a reduction of the maximum unambiguous range. In this work, a novel OFDM-based waveform robust to IQ imbalance is proposed, which neither increases the noise floor nor reduces the maximum unambiguous range. The latter is achieved by shifting the ghost objects in the RDM to different velocities such that their range variations observed over several consecutive RDMs do not correspond to the observed velocity. This allows tracking algorithms to identify them as ghost objects and eliminate them for the follow-up processing steps. Moreover, we propose complete communication systems for both the proposed waveform as well as for the state-of-the-art waveform, including methods for channel estimation, synchronization, and data estimation that are specifically designed to deal with frequency selective IQ imbalance which occurs in wideband systems. The effectiveness of these communication systems is demonstrated by means of bit error ratio (BER) simulations.

eess.SP↗

arXiv2023

Bi-Linear Homogeneity Enforced Calibration for Pipelined ADCs

Pipelined analog-to-digital converters (ADCs) are key enablers in many state-of-the-art signal processing systems with high sampling rates. In addition to high sampling rates, such systems often demand a high linearity. To meet these challenging linearity requirements, ADC calibration techniques were heavily investigated throughout the past decades. One limitation in ADC calibration is the need for a precisely known test signal. In our previous work, we proposed the homogeneity enforced calibration (HEC) approach, which circumvents this need by consecutively feeding a test signal and a scaled version of it into the ADC. The calibration itself is performed using only the corresponding output samples, such that the test signal can remain unknown. On the downside, the HEC approach requires to accurately scale the test signal, impeding an on-chip implementation. In this work, we provide a thorough analysis of the HEC approach, including limitations such as the effects of an inaccurately scaled test signal. Furthermore, the bi-linear homogeneity enforced calibration (BL-HEC) approach is introduced and suggested to account for an inaccurate scaling and, therefore, to facilitate an on-chip implementation. In addition, a comprehensive stability analysis of the BL-HEC approach is carried out. Finally, we verify our concept with behavioral Matlab simulations and measurements conducted on 24 integrated ADCs.

eess.SP↗

arXiv2023

Analysis and Compensation of Carrier Frequency Offset Impairments in Unique Word OFDM

Unique Word-orthogonal frequency division multiplexing (UW-OFDM) is known to provide various performance benefits over conventional cyclic prefix (CP) based OFDM. Most important, UW-OFDM features excellent spectral sidelobe suppression properties and an outstanding bit error ratio (BER) performance. Carrier frequency offset (CFO) induced impairments denote a challenging task for OFDM systems of any kind. In this work we investigate the CFO effects on UW-OFDM and compare it to conventional multi-carrier and single-carrier systems. Different CFO compensation approaches with different computational complexity are considered throughout this work and assessed against each other. A mean squared error analysis carried out after data estimation reveals a significant higher robustness of UW-OFDM over CP-OFDM against CFO effects. Additionally, the conducted BER simulations generally support this conclusion for various scenarios, ranging from uncoded to coded transmission in a frequency selective environment.

eess.SP↗

arXiv2023

Complex-valued Adaptive System Identification via Low-Rank Tensor Decomposition

Machine learning (ML) and tensor-based methods have been of significant interest for the scientific community for the last few decades. In a previous work we presented a novel tensor-based system identification framework to ease the computational burden of tensor-only architectures while still being able to achieve exceptionally good performance. However, the derived approach only allows to process real-valued problems and is therefore not directly applicable on a wide range of signal processing and communications problems, which often deal with complex-valued systems. In this work we therefore derive two new architectures to allow the processing of complex-valued signals, and show that these extensions are able to surpass the trivial, complex-valued extension of the original architecture in terms of performance, while only requiring a slight overhead in computational resources to allow for complex-valued operations.

cs.LG↗

arXiv2023

Doppler-Division Multiplexing for MIMO OFDM Joint Sensing and Communications

A promising waveform candidate for future joint sensing and communication systems is orthogonal frequencydivision multiplexing (OFDM). For such systems, supporting multiple transmit antennas requires multiplexing methods for the generation of orthogonal transmit signals, where equidistant subcarrier interleaving (ESI) is the most popular multiplexing method. In this work, we analyze a multiplexing method called Doppler-division multiplexing (DDM). This method applies a phase shift from OFDM symbol to OFDM symbol to separate signals transmitted by different Tx antennas along the velocity axis of the range-Doppler map. While general properties of DDM for the task of radar sensing are analyzed in this work, the main focus lies on the implications of DDM on the communication task. It will be shown that for DDM, the channels observed in the communication receiver are heavily timevarying, preventing any meaningful transmission of data when not taken into account. In this work, a communication system designed to combat these time-varying channels is proposed, which includes methods for data estimation, synchronization, and channel estimation. Bit error ratio (BER) simulations demonstrate the superiority of this communications system compared to a system utilizing ESI.

cs.IT↗

arXiv2023

Enhanced Nonlinear System Identification by Interpolating Low-Rank Tensors

Function approximation from input and output data is one of the most investigated problems in signal processing. This problem has been tackled with various signal processing and machine learning methods. Although tensors have a rich history upon numerous disciplines, tensor-based estimation has recently become of particular interest in system identification. In this paper we focus on the problem of adaptive nonlinear system identification solved with interpolated tensor methods. We introduce three novel approaches where we combine the existing tensor-based estimation techniques with multidimensional linear interpolation. To keep the reduced complexity, we stick to the concept where the algorithms employ a Wiener or Hammerstein structure and the tensors are combined with the well-known LMS algorithm. The update of the tensor is based on a stochastic gradient decent concept. Moreover, an appropriate step size normalization for the update of the tensors and the LMS supports the convergence. Finally, in several experiments we show that the proposed algorithms almost always clearly outperform the state-of-the-art methods with lower or comparable complexity.

math.ST↗

arXiv2022

Neural Network Approaches for Data Estimation in Unique Word OFDM Systems

Data estimation is conducted with model-based estimation methods since the beginning of digital communications. However, motivated by the growing success of machine learning, current research focuses on replacing model-based data estimation methods by data-driven approaches, mainly neural networks (NNs). In this work, we particularly investigate the incorporation of existing model knowledge into data-driven approaches, which is expected to lead to complexity reduction and / or performance enhancement. We describe three different options, namely "model-inspired'' pre-processing, choosing an NN architecture motivated by the properties of the underlying communication system, and inferring the layer structure of an NN with the help of model knowledge. Most of the current publications on NN-based data estimation deal with general multiple-input multiple-output communication (MIMO) systems. In this work, we investigate NN-based data estimation for so-called unique word orthogonal frequency division multiplexing (UW-OFDM) systems. We highlight differences between UW-OFDM systems and general MIMO systems one has to be aware of when using NNs for data estimation, and we introduce measures for successful utilization of NN-based data estimators in UW-OFDM systems. Further, we investigate the use of NNs for data estimation when channel coded data transmission is conducted, and we present adaptions to be made, such that NN-based data estimators provide satisfying performance for this case. We compare the presented NNs concerning achieved bit error ratio performance and computational complexity, we show the peculiar distributions of their data estimates, and we also point out their downsides compared to model-based equalizers.

cs.IT↗

arXiv2022

Diagnostic Quality Assessment for Low-Dimensional ECG Representations

There have been several attempts to quantify the diagnostic distortion caused by algorithms that perform low-dimensional electrocardiogram (ECG) representation. However, there is no universally accepted quantitative measure that allows the diagnostic distortion arising from denoising, compression, and ECG beat representation algorithms to be determined. Hence, the main objective of this work was to develop a framework to enable biomedical engineers to efficiently and reliably assess diagnostic distortion resulting from ECG processing algorithms. We propose a semiautomatic framework for quantifying the diagnostic resemblance between original and denoised/reconstructed ECGs. Evaluation of the ECG must be done manually, but is kept simple and does not require medical training. In a case study, we quantified the agreement between raw and reconstructed (denoised) ECG recordings by means of kappa-based statistical tests. The proposed methodology takes into account that the observers may agree by chance alone. Consequently, for the case study, our statistical analysis reports the "true", beyond-chance agreement in contrast to other, less robust measures, such as simple percent agreement calculations. Our framework allows efficient assessment of clinically important diagnostic distortion, a potential side effect of ECG (pre-)processing algorithms. Accurate quantification of a possible diagnostic loss is critical to any subsequent ECG signal analysis, for instance, the detection of ischemic ST episodes in long-term ECG recordings.

eess.SP↗

arXiv2022

Homogeneity Enforced Calibration of Stage Nonidealities for Pipelined ADCs

Pipelined analog-to-digital converters (ADCs) are fundamental components of various signal processing systems requiring high sampling rates and a high linearity. Over the past years, calibration techniques have been intensively investigated to increase the linearity. In this work, we propose an equalization-based calibration technique which does not require knowledge of the ADC input signal for calibration. For that, a test signal and a scaled version of it are fed into the ADC sequentially, while only the corresponding output samples are used for calibration. Several test signal sources are possible, such as a signal generator (SG) or the system application (SA) itself. For the latter case, the presented method corresponds to a background calibration technique. Thus, slowly changing errors are tracked and calibrated continuously. Because of the low computational complexity of the calibration technique, it is suitable for an on-chip implementation. Ultimately, this work contains an analysis of the stability and convergence behavior as well as simulation results.

eess.SP↗

arXiv2021

VPNet: Variable Projection Networks

We introduce VPNet, a novel model-driven neural network architecture based on variable projection (VP). Applying VP operators to neural networks results in learnable features, interpretable parameters, and compact network structures. This paper discusses the motivation and mathematical background of VPNet and presents experiments. The VPNet approach was evaluated in the context of signal processing, where we classified a synthetic dataset and real electrocardiogram (ECG) signals. Compared to fully connected and one-dimensional convolutional networks, VPNet offers fast learning ability and good accuracy at a low computational cost of both training and inference. Based on these advantages and the promising results obtained, we anticipate a profound impact on the broader field of signal processing, in particular on classification, regression and clustering problems.

cs.LG↗