In this paper, we present a hybrid learning framework that integrates two model-driven AI paradigms: Deep unfolding and Variable Projections (VPs). The core idea is to unfold the iterations of VP solvers for separable nonlinear least squares (SNLLS) problems into trainable neural network layers. As a consequence, the network is capable of learning optimal nonlinear VP parameters during inference, which is a form of model-based meta-learning. Furthermore, the architecture incorporates prior knowledge of the underlying SNLLS problem, such as basis function expansions and signal structure, which enhance interpretability, reduce model size, and lower data requirements. As a case study, we adapt the proposed deep unfolded VPNet to learn ECG representations for the classification of five arrhythmias. Experimental results on the MIT-BIH Arrhythmia Database show that VPNet achieves performance comparable to state-of-the-art ECG classifiers, attaining 95% accuracy while maintaining a compact architecture. Its low computational complexity enables efficient training and inference, making it highly suitable for real-time, power-efficient edge computing applications. This is further validated through embedded implementation on STM32 microcontrollers.

Functional link adaptive filters (FLAFs) utilize expansion blocks to nonlinearly augment the input signal to a higher dimensional space, after which an adaptive weight algorithm is applied. These filters are useful for nonlinear system identification tasks, as they can update a large number of coefficients to effectively model the nonlinear system, even when the degree of nonlinearity is not comprehended in advance. However, in many cases, not all of the weights in the nonlinear and linear filter will significantly contribute to the identified model. This paper introduces a novel class of FLAFs based on the nearest Kronecker product (NKP) decomposition. Utilizing the inherent low-rank nature of weight vectors in many scenarios, our approach aims to improve convergence performance and tracking capabilities compared to traditional FLAFs. Additionally, we address noise mitigation challenges, particularly in nonlinear acoustic echo cancellation scenarios. By incorporating NKP decomposition, our proposed FLAFs offer promising solutions for enhancing adaptability and performance in nonlinear system identification, making them valuable tools in practical applications.

State-of-the-art radar sensors for automotive driving applications have to fulfill stringent requirements on reliability and functional safety. Therefore, monitoring concepts that cover various sensor parameters such as gain, phase, noise figure or phase-noise became an active field of research. Another essential sensor parameter is the receiver's linearity. A major design challenge for on-chip linearity testing is the generation of a spectrally pure test signal. In the past, this was mainly tackled by sophisticated analog design of the test signal generator (TSG), such that the overall complexity of the test circuitry was increased. In this work, we propose a monitoring concept that eliminates the need for a pure test signal. This enables accurate on-chip linearity testing by the use of a low-cost TSG. With an imprecise test signal, accurate testing can be achieved by exploiting the test signal generator's repeatability. By means that, instead of injecting a spectrally pure test signal into the device under test (DUT), an imprecise test signal is injected multiple times, but each time scaled with a different, properly chosen factor. With this, all impairments affecting the test signal before the analog scaling can be separated from nonlinear effects after the scaling by digital signal processing. Thus, the requirements on the analog test circuitry are heavily relaxed. Ultimately, we proof our concept by simulation as well as measurement results.

In this paper, 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.

State-of-the-art radio frequency transceivers for mobile communication devices suffer from transmitter-to-receiver (Tx-Rx) leakage in frequency division duplex operation, which, in combination with further non-idealities in the analog frontend, may lead to diverse self-interference (SI) effects. Digital as well as mixed-signal architectures have been proposed for self- interference cancellation. In this work we present a digitally intensive mixed-signal approach, where a low-cost auxiliary receiver senses the leaked Tx-signal. Firstly, the sensed signal is used to adaptively estimate the leakage channel, whereas in a second step a cleaned version of the leaked Tx-signal is reconstructed digitally. This reconstructed Tx-leakage signal is then used as input for a low complex adaptive interference cancellation unit to suppress modulated spurs or intermodulation distortions. We show that this approach allows to significantly relax the analog auxiliary receiver specifications, while different to conventional all-digital solutions being able to deal with multiple different types of SI with minimal configuration overhead.

In order to enable frequency-division duplex operation, radio frequency transceivers usually employ a spectral isolation between transmitter and receiver. Due to nonidealities of the used duplexer filters, the transmit signal leaks into the receive path. Although operating on different frequency bands, nonlinear effects in the transceiver may lead to self-interferences with possibly high power levels. One approach to restore the receiver signal-to-noise ratio in these cases is to apply a digital cancellation of the interference. If perfect model knowledge is available, particularly tailored algorithms can be used for interference cancellation. In this work, we apply a kernel-based universal estimation algorithm, in particular the kernel recursive least squares (KRLS) algorithm, to cancel two different nonlinear interference effects. The transmitter-induced harmonics are explained and studied in detail, while the receiver-induced intermodulation distortion has been treated in a second paper explicitly. The KRLS algorithm is able to cancel both, while two different model-based and particularly tailored methods would be needed to address the two fundamentally different interference effects.

Digital-to-time converters (DTCs) are a promising technology for radio frequency (RF) transceivers but are prone to spur generation. A common approach to change the spurious emissions to a spur-free shape is a method called dithering. The power added due to dithering is an important aspect of this approach and gives raise to investigations on additive dither as well as methods for subtractive dithering. This work presents a mathematical model for dithering DTC-based local oscillator (LO) generators. It proposes concepts for the application of subtractive dither and it introduces a novel generalization of quantization-dither to allow for optimal dithering of nonlinear quantizers.

The estimation of the frequencies of multiple complex sinusoids in the presence of noise is required in many applications such as sonar, speech processing, communications, and power systems. According to previous work [1], [2], this problem can be reformulated as a separable nonlinear least squares problem (SNLLS). In this paper, such formulation is derived, and a variable projection (VP) optimization is proposed for solving the SNLLS problem and estimate the frequency parameters. We also apply a lethargy-type theorem for quantifying the difficulty of the optimization. Moreover, an alternative procedure that speeds up the computation of the exact gradient is presented. Simulation results reveal that the proposed algorithm outperforms existing methods in terms of the MSE.

We propose a hybrid method for reconstructing thermographic images by combining the recently developed virtual wave concept with deep neural networks. The method can be used to detect defects inside materials in a non-destructive way. We propose two architectures along with a thorough evaluation that shows a substantial improvement compared to state-of-the-art reconstruction procedures. The virtual waves are invariant of the thermal diffusivity property of the material. Consequently, we can use extremely compact architectures that require relatively little training data, and have very fast loss convergence. As a supplement of the paper [1], we provide the MATLAB and Python implementations along with the data set comprising 40,000 simulated temperature measurement images in total, and their corresponding defect locations. Thus, the presented results are completely reproducible.

Due to its scalability, tuneability, low latency and high linearity, the radio frequency digital to analog converter (RFDAC) is shown to be an attractive choice for the transmitter architecture for the long term evolution (LTE) and new radio (NR) standards, especially for the sub 8 GHz frequency range. When placing several transmitters (TX) and receivers (RX) next to each other on a single die, it is necessary to understand any influence of the components to each other to enable undisturbed functionality and by that to fulfill the tough requirements that are defined by the 3rd Generation Partnership Project (3GPP). One of the unwanted effects generated by the RFDAC structure are harmonics. These harmonics can further lead to effects such as: Inphase Quadrature image generation, TX to TX crosstalk, TX to digitally controlled oscillator (DCO) crosstalk also known as pulling, TX to RX crosstalk etc. In this paper we conduct a thorough analysis of the harmonics that are generated by an RFDAC and show that the widening of the bandwidth does not depend on the harmonic order but rather on the RFDAC specific nonlinearities. Theoretical and mathematical observations are supported by a highly developed MATLAB simulation chain containing an RFDAC model whose behavior is aligned with a physical implementation.

Communication systems have to cope with a variety of non-idealities. Among those, carrier frequency offset (CFO) induced impairments denote a challenging task for any system, but especially for multi-carrier schemes such as orthogonal frequency division multiplexing (OFDM). Unique Word-OFDM (UW-OFDM) is a known variant, which provides various performance benefits over conventional cyclic prefix (CP) based OFDM. In particular, UW-OFDM features excellent spectral sidelobe suppression properties and an outstanding bit error ratio performance. In this work, we expand the investigations to CFO impairments, analyze the varying effects experienced by different UW-OFDM realizations and compare it against conventional OFDM as well as single-carrier systems. Through an error analysis it is shown that compared to CP-OFDM, UW-OFDM has a significantly higher robustness against CFO. This advantage becomes even more dominant with an increasing CFO.

In frequency division duplex transceivers, a part of the transmit signal leaks into the receive path due to the non-ideality of the analog duplexer. Although operating on a different frequency band, non-ideal effects in the receive path lead to self-interferences with potentially higher power level than the wanted receive signal. One option to tackle this problem is the use of adaptive filtering algorithms for interference cancellation. In this work, we discuss support vector machines (SVMs) as an alternative approach. Different to the existing methods, the proposed concept does not need a model of the type of interference. We investigate the cancellation performance of SVMs compared to a recently published nonlinear adaptive filter for a specific type of interference called the second-order intermodulation distortion. It turns out that SVMs clearly outperform the adaptive filtering approach even for the severe case of narrow allocated transmit signals and high transmit power levels.

State-of-the-art radio frequency transceivers for mobile communication devices suffer from transmitter-to-receiver (Tx-Rx) leakage in frequency division duplex operation. Although the duplexing distance between Tx and Rx carriers prohibits direct down-conversion of this leakage, transmit signal components might fall into the Rx baseband because of further non-idealities in the analog front-end and thus degrade performance. These so called self-interference issues have been addressed in literature and several countermeasures have been proposed. Two of them are digital and mixed-signal mitigation architectures. In this work, we analyze the problem from a system level perspective and enable a fair comparison of the existing solutions by making use of an extensive simulation framework that covers many relevant factors for the task of interference cancellation. Furthermore, the computational complexity of the estimation algorithms and other relevant aspects for real-time processing are discussed. This study reveals that choosing a suitable adaptive algorithm and optimizing it for the application is as crucial as the cancellation architecture itself.

Modern medical science demands sophisticated signal representation methods in order to cope with the increasing amount of data. Important criteria for these methods are mainly low computational and storage costs, whereas the underlying mathematical model should still be interpretable and meaningful for the data analyst. One of the most promising models fulfilling these criteria is based on Hermite functions, however having some important limitations for specific biomedical wave shapes. We extend this model by using weighted Hermite functions and develop a gradient based constrained optimization method to adapt the system for different types of signals. In order to demonstrate the potential of our approach, we consider the problem of electrocardiogram signal compression. The experiments on the MIT/BIH arrhythmia database show a significant improvement compared to the former works using classical Hermite functions.

In this work, we consider a bit-interleaved coded modulation unique word (UW) OFDM system with turbo equalization. For such a system, only a limited performance improvement over the turbo iterations was observed in a previous work. This is because interleaving was limited to one UW-OFDM symbol, which causes dependencies between the log-likelihood ratios exchanged between the equalizer and the decoder. To overcome this issue, we propose interleaving beyond one UW-OFDM symbol. Moreover, through bit error ratio simulations we determine the minimum number of UW-OFDM symbols across which interleaving should be performed to reach the matched filter bound (MFB).

The stringent requirements from 3rd Generation Partnership Project (3GPP) standards pose a big challenge for the design of cellular radio frequency transmitters. Architectural limitations and non-ideal effects of state-of-the art complementary metal-oxide-semiconductor (CMOS) devices need to be carefully addressed to achieve sufficient performance at low costs. In this work, the systematic problem of an IQ image in digital-intensive IQ transmitters is analyzed. The root cause is explained and an efficient cancellation method is presented. A proof of concept for long term evolution (LTE) signals is given via Matlab simulation results and measurements from a test chip implemented in 28 nm Complementary Metal-Oxide-Semiconductor (CMOS). The proposed concept improves the error vector magnitude up to 4% for wideband signals.

This paper presents extensions to the switched state-space modeling approach for capacitive RF-DACs. In contrast to other state-of-the-art behavioral models for nonlinear devices, the state-space representation allows for inclusion of the dominant non-idealities of the capacitive RF-DAC, while maintaining the original benefits of low computational effort and low simulation run times. In this work, the inclusion of capacitor variations, supply voltage variations, and local oscillator phase noise into the switched state-space model is demonstrated and verified.

The limited transmitter-to-receiver stop-band isolation of the duplexers in long term evolution and 5G frequency division duplex transceivers induces leakage signals from the transmitter(s) (Tx) into the receiver(s) (Rx). These leakage signals are the root cause of a multitude of self- interference (SI) problems in the receiver path(s) diminishing a receiver's sensitivity. This work deals with second-order intermodulation distortion, arising from the Tx leakage signal in combination with a coupling between the RF- and local oscillator-ports of the Rx IQ-mixer. We propose a novel adaptive architecture, utilizing neural networks, to cancel this type of interference. In contrast to traditional adaptive filter solutions, the proposed architecture can be used even if there is no model of the system available, making it robust against modeling noise and flexible in terms of interferences that it is able to cancel. The proposed architecture outperforms existing work based on least mean squares (LMS) algorithms and converges as fast as recursive least squares algorithms while maintaining comparably low complexity as the LMS approach.

In this paper a state-space approach for modeling radio-frequency digital-to-analog-converters is presented. We explain the root cause of the sign-change error appearing in RF-DAC architectures with local oscillator (LO) switching and present its impact in time domain and on the spectrum of the output RF signal. The main contribution of this paper is a solution for the removal of the sign-change error. The solution is implemented in a linear state-space model, considering static nonlinearities of the switched-capacitor power amplifier cells. We demonstrate the feasibility of our approach by simulation results for LTE5, LTE10 and LTE20 signals and show that the out-of-band noise contribution of the sign-change error is completely removed. Additionally, the impact of static nonlinearities on the system is investigated.

Radar systems are key components for today's advanced driver assistance systems such as adaptive cruise control or emergency brake assistants. Along with the rising utilization of radars in modern cars, the issue of interference amongst themselves arises. It has been shown in previous work that interference between different frequency modulated continuous wave (FMCW) radar systems leads to an increased noise floor. This may severely impact the detectability of objects, especially those with a small radar cross section like pedestrians. In this work we propose a novel concept to mitigate interference in FMCW radar transceivers using digital signal processing. The actual interference cancellation is carried out in frequency domain taking into account a sequence of FMCW chirps. Therewith noise suppression is performed, the interference is cancelled, and the object information is retained in the radar image. In contrast to existing interference cancellation concepts, no threshold needs to be chosen in advance. We prove our method with simulation results, and compare it to existing work.

In this paper, we propose buffer-aided physical-layer network coding (PLNC) techniques for improving data transmission over cooperative networks. In particular, we develop buffer-aided PLNC schemes and relay pair selection algorithms for direct-sequence code-division multiple access (DS-CDMA) systems. We devise PLNC techniques based on optimal linear network coding matrices according to the maximum likelihood and minimum mean-square error design criteria in order to generate the network coded symbols that are sent to the destination. In the proposed buffer-aided PLNC schemes, relay pair selection algorithms are developed to obtain the relay pair and the packets in the buffer entries with the best performance and the associated link combinations are used for the data transmission. An analysis of the computational complexity of the proposed techniques along with their sum-rate analysis is carried out. Simulation results show that the proposed techniques significantly outperform previously reported approaches.

In this paper we develop space-time block codes for unique word - orthogonal frequency division multiplexing (UW-OFDM) systems to fully exploit the diversity gain when the channel state information is not available at the transmitter. To this end, we propose two novel space-time block codes (STBCs) for UW-OFDM systems, namely a frequency domain space-time block code and a time-reversal space-time code (TR-STC). The former one is an extension of the traditional space-frequency block codes (SFBCs) for CP-OFDM systems to UW-OFDM systems while the latter one makes use of the frame structure of the UW-OFDM symbols and has a low complexity decoder. Simulation results show that both of the proposed space-time block codes achieve a significant gain compared to the SFBC based CP-OFDM. Moreover, the frequency domain STBC yields a slightly better performance as compared to the TR-STC but has a higher computational complexity.

The estimation of sparse vectors is an important problem in digital signal processing. Recently, efficient iterative algorithms based on the so-called linearized Bregman iterations have been proposed, combining excellent estimation performance with low implementation complexity. Unfortunately, these algorithms typically use large numerical values, complicating fixed point implementations. To overcome this problem, we propose a modification of these algorithms based on scaling at specific algorithmic steps. We show that with this modification the algorithm still converges to the optimal solution and that it allows to implement linearized Bregman iterations completely in fractional precision fixed point. We show bit true simulation results, as well as synthesis results demonstrating the performance of the implemented algorithms.

The continuous operation of frequency modulated continuous wave (FMCW) radar transceivers leads to permanent leakage from the transmit into the receive path. Besides leakage within FMCW radar systems itself, unwanted reflections from a close object in front of the radar antennas results in so-called short-range (SR) leakage. Specifically the phase noise (PN) contained in the SR leakage causes a severe sensitivity degradation of the radar. In this work, we propose a mixed-signal approach to mitigate the SR leakage, including its contained PN. It makes use of an artificial on-chip target (OCT), whose output signal is modulated on a sinusoidal generated in the digital domain. The concept is based on cross-correlation properties of the residual PN between the OCT and the SR leakage in the intermediate frequency (IF) domain of the transceiver. We prove our analytical investigations with an FMCW radar simulation, wherein almost perfect SR leakage cancelation is achieved.

Modern wireless transmitters supporting Long Term Evolution (LTE) face several issues when employing a radio-frequency-digital-to-analog-converter (RFDAC). One of these issues is the pulling effect between the RFDAC and the digitally controlled oscillator (DCO). This so-called remodulation occurs due to the unavoidable electromagnetic coupling between them. This results in spurious spectral emissions at the transmitter output and causes a severe degradation of the error vector magnitude (EVM). This paper introduces and studies the remodulation problem with a detailed modeling of the LTE transmitter, including an RFDAC and a phase locked loop. Also, the impact of the remodulation effect is analyzed with simulations for various LTE uplink scenarios. To mitigate the unwanted pulling effect, we propose a low-complex digital cancellation architecture that utilizes the inphase and quadrature samples of the transmit signal. Simulations show that the proposed architecture cancels the remodulation and improves the EVM of the transmitter significantly.

Biased estimators can outperform unbiased ones in terms of the mean square error (MSE). The best linear unbiased estimator (BLUE) fulfills the so called global conditional unbiased constraint when treated in the Bayesian framework. Recently, the component-wise conditionally unbiased linear minimum mean square error (CWCU LMMSE) estimator has been introduced. This estimator preserves a quite strong (namely the CWCU) unbiased condition which in effect sufficiently represents the intuitive view of unbiasedness. Generally, it is global conditionally biased and outperforms the BLUE in a Bayesian MSE sense. In this work we briefly recapitulate CWCU LMMSE estimation under linear model assumptions, and additionally derive the CWCU LMMSE estimator under the (only) assumption of jointly Gaussian parameters and measurements. The main intent of this work, however, is the extension of the theory of CWCU estimation to CWCU widely linear estimators. We derive the CWCU WLMMSE estimator for different model assumptions and address the analytical relationships between CWCU WLMMSE and WLMMSE estimators. The properties of the CWCU WLMMSE estimator are deduced analytically, and compared by simulation to global conditionally unbiased as well as WLMMSE counterparts with the help of a parameter estimation example and a data estimation/channel equalization application.

In this letter, we investigate the extrinsic log-likelihood ratio (LLR) computation of a soft-input soft-output equalizer used in a turbo equalization system. The optimum LLRs are obtained by a maximum a posteriori -based equalizer, which may be computationally expensive. Thus, several reduced-complexity equalizers have been proposed. The most promising approach first applies linear minimum mean square error filtering to the channel output and then computes the LLRs based on a scalar Gaussian approximation of the filter output. The resulting LLRs can be viewed as an approximation of the optimum LLRs. In order to improve the approximation, we investigate the computation of the LLRs based on a vector Gaussian approximation of the filter output, which incorporates the correlation between the estimated symbols after filtering. Surprisingly, it turns out that both approaches, although their derivation is different, give the same LLRs. We verify this remarkable result through an analytical proof and bit error ratio simulations.

Mobile user equipments supporting Long Term Evolution (LTE) - carrier aggregation (CA) face several new unpredicted issues when operating in frequency division duplex mode. This paper addresses the transmit (Tx) modulated spur problem, a new issue, that appears in LTE-CA transceivers. Due to the presence of several mixers and dividers to support different bands and CA scenarios, many harmonics are generated on the chip. Unavoidable crosstalk between these harmonics can generate an unwanted continuous wave spur around the Tx frequency. Once this spur appears at the chip area where the receive (Rx) local oscillator resides, it downconverts the undesired portion of the Tx signal which leaked through the duplexer to the Rx baseband. This interference, referred to as Tx modulated spur, causes a severe degradation of the signal-to-noise ratio (SNR) of the wanted signal. To mitigate such, we propose an active digital cancellation architecture which utilizes the known information of the Tx signal. An equivalent baseband model of the Tx modulated spur is presented from which the digital cancellation architecture is derived. Simulations show that the proposed architecture cancels the Tx interference and improves the SNR of the wanted signal significantly.

In this paper, a digital implementation of a unified sliding mode controller for DC-DC converters is presented. Although current information is used as a control variable, a pure voltage-mode implementation has been considered and the inductor current ripple is digitally reconstructed. The proposed universal control algorithm can be applied to different converter topologies and operating modes. In particular, experimental results are shown for a prototyping system of a Buck converter operating in Continuous Conduction Mode (CCM) and Discontinuous Conduction Mode (DCM), where only negligible transient effects are present when changing the operation mode. The presented measurements confirm that the proposed universal control algorithm is capable of achieving tight voltage regulation and excellent dynamic performance under different operating conditions. Additionally, a comparison to a state-of-the-art digital Proportional-Integral-Derivative (PID) controller illustrates the performance improvement of the proposed digital control scheme.

This paper presents a switched state-space modeling approach for a switched-capacitor power amplifier. In contrast to state of the art behavioral models for nonlinear devices like power amplifiers, the state-space representation allows a straightforward inclusion of the nonidealities of the applied input sources. Hence, adding noise on a power supply or phase distortions on the carrier signal do not require a redesign of the mathematical model. The derived state-space model (SSM), which can be efficiently implemented in any numerical simulation tool, allows a significant reduction of the required simulation run-time (14x speedup factor) with respect to standard Cadence Spectre simulations. The derived state-space model (SSM) has been implemented in MATLAB/Simulink and its results have been verified by comparison with Cadence Spectre simulations.

In this paper we consider a unique word OFDM (UW-OFDM) system with iterative detection, for which we explore two soft-input soft-output (SISO) detection algorithms: A LMMSE detector and a detector based on the generalized approximate message passing (GAMP) algorithm. For the GAMP based detector we propose additional simplifications suitable for UW-OFDM to further reduce the complexity while exhibiting only a small BER performance loss. As verified by computer simulations, both algorithms show a significant BER performance improvement over the iterations. Moreover, the GAMP based detector with the proposed simplifications outperforms the computational more complex LMMSE detector.

We present a novel iterative algorithm for approximating the linear least squares solution with low complexity. After a motivation of the algorithm we discuss the algorithm's properties including its complexity, and we present theoretical results as well as simulation based performance results. We describe the analysis of its convergence behavior and show that in the noise free case the algorithm converges to the least squares solution.

A reliable knowledge of cell parameters like the state-of-charge (SoC) is essential for the optimization of battery-powered applications. Usually, during relaxation (the phase of no or low loads) the SoC is determined based on the measurement of the battery's electro-motive force (EMF). To obtain a reliable measurment, it is required that the battery voltage transient is in a well-relaxed state, which is rarely reached in practice (e.g. due to periodic discharge activities). In this paper, a predictive methodology is presented which is able to forecast the EMF and therewith the SoC already during a not well-relaxed state of the voltage transient. A nonlinear relaxation voltage model is reformulated such that the problem can be treated as a linear least squares estimation problem. Based on this estimation, the performance is evaluated with respect to the following aspects: prediction time, current rate influence, SoC influence, cell-to-cell deviation, or rather aging and temperature effects. Experimental results are presented for a fixed-point implementation of the estimation scheme on a CY8CKIT-050 PSOC5 programmable system on chip. For validation, measurements of 2.25Ah Sanyo UR18650A lithium cells have been used. It is shown that the presented approach offers an improved re-initialization methodology for the Coulomb counting method, and that it clearly outperforms the usual EMF-measurement based SoC determination method.

Unique word-orthogonal frequency division multiplexing (UW-OFDM) is known to feature an excellent bit error ratio performance when compared to conventional OFDM using cyclic prefixes (CP). In recent papers classical and Bayesian linear data estimators as well as the non-linear sphere decoding (SD) approach have been investigated in the UW-OFDM context. In general, the remaining error samples after a linear data estimation in UW-OFDM are correlated. In this work, noise interpolation (NI) is proposed, which uses preliminary data decisions together with Wiener interpolation to exploit these correlations. It turns out that the selection of samples to be used for NI is crucial. With the sample selection policy suggested in this work, NI clearly outperforms LMMSE data estimation, and due to its lower complexity compared to SD, it can be regarded as an attractive compromise for UW-OFDM reception.

Unique word orthogonal frequency division multiplexing (UW-OFDM) inherently introduces a complex number Reed Solomon (RS) code. Originally, the code generator matrix of systematic coded UW-OFDM had been designed rather intuitively by minimizing the mean redundant energy. In this work we justify this approach by applying a cost function that incorporates the overall transceiver chain including a linear minimum mean square error (LMMSE) data estimator. In addition to the LMMSE estimator we investigate a nonlinear sphere detection (SD) receiver for both systematic and nonsystematic coded UW-OFDM. We study and interpret the estimators' performance and their diverse ability to exploit the redundant energy.

The recently presented UW-OFDM (Unique Word - orthogonal frequency division multiplexing) signaling scheme [1] uses certain subcarriers in frequency domain for redundant symbols instead of data, in order to generate a zero word in the DFT (discrete Fourier transform) interval in time domain. These redundant symbols depend on the data loaded on the other carriers and thus introduce correlation. The resulting linear system model enables sophisticated detectors for data recovery. As the best known maximum likelihood detector for this case, we applied the Sphere Decoding (SD) algorithm to a single antenna UW-OFDM system and evaluated its bit error performance in AWGN (additive white Gaussian noise) and frequency selective environments. Compared to linear receivers, the SD is able to take the correlations on the redundant subcarriers optimally into account and shows an enormous gain. A reduction of the redundant energy by increasing the number of redundant subcarriers improved the bit error performance of UW-OFDM systems with linear data estimators [2] at the price of a lower bandwidth efficiency. In contrast it is found, that a receiver based on Sphere Decoding is able to optimally exploit the excess redundant energy. The SD based system therefore shows its best performance at maximum bandwidth efficiency.

Interest is apparently growing in the four-switch noninverting Buck-Boost converter, which can be effectively used as a power supply for mobile devices. A typical application is a dynamic power supply for third-generation (3G) Wideband Code Division Multiple Access (WCDMA) Power Amplifiers (PA). While several control strategies based on the linear control theory have been proposed recently, there is still room for investigation of nonlinear control techniques. In this paper, a fixed-frequency nonlinear controller based on the sliding mode theory is presented and compared to a classical linear Proportional-Integral-Derivative (PID) implementation. The main differences from the design perspective and in terms of performance are then pointed out and commented. The main advantages of the proposed technique generally include an improvement of the dynamic performance and increased energy efficiency of the conversion. Simulation results are provided in order to evaluate and compare the different control strategies.

Power efficiency is an important issue in mobile communication systems. Especially for mobile user equipments (UE), the energy budget, limited by a battery, has to be treated carefully. In fact, though, quite an amount of energy is wasted in todays UEs, as analog and digital frontend (AFE/DFE) in communication systems are designed for worst case scenarios. Therefore, for most real transmission processes those are over-engineered concerning e.g. amplifier linearity and filter slope steepnesses. Gaining information about the instantaneous spectral allocation of the UE's environment highly supports optimal frontend configuration for adequate energy consumption. This paper introduces an efficient concept for spectrally sensing a Long Term Evolution (LTE) UE's environment leading to an intelligent DFE. Low complex filter chains for different channel bandwidths based on a dyadic structure, containing mainly infinite impulse response (IIR) filters will be discussed. Furthermore, simulation results for the 20MHz test case will be given.

This paper discusses the application of iterative versus adaptive equalizers to a universal mobile telecommunications system (UMTS) high speed downlink packet access (HSDPA) receiver. Among the investigated algorithms are the least mean square (LMS) and recursive least square (RLS) algorithms and their adaptations to code division multiple access (CDMA), such as Griffith's algorithm, as well as the conjugate gradient (CG) algorithm. We show that adaptive algorithms do not perform well in HSDPA scenarios at medium to high mobile speed, and that the conjugate gradient algorithm achieves higher performance at the same or even lower complexity than any of the investigated adaptive algorithms. We also discuss the necessity to consider oversampling.

In this work, we present a new framework for developing decoding algorithms for linear block codes. We define a new graphical model, called configuration graph and show that maximum likelihood (ML) sequence decoding can be performed by finding a maximum vertex packing on a configuration graph. We present examples of low-cost decoding algorithms developed using this graphical model and show that remarkable performance improvements can be achieved especially by combining these algorithms with belief propagation (BP) decoding for Low Density Parity Check (LDPC) codes.

Quality of Service (QoS) is a prevalent issue in communication systems, but an often neglected aspect of local positioning. In this paper, we present a QoS architecture for the high-accuracy local positioning system RESOLUTION. Different backend configurations of this system are compared in terms of impact on accuracy and latency of the system. Quality metrics relevant to local positioning requirements are defined and tested with several protocol and algorithmic options by means of simulations. Results indicate parameter tradeoffs which call for optimization of the backend architecture for specific deployment scenarios. These insights are easy to transfer to any positioning system which has to deal with channel contention situations.

The DNL performance of the last generation of Analog-to-Digital Converter (A/D Converter, ADC) based on the principle of SAR (successive-approximation register) was limited by a small amount of spikes at particular codes that changed with varying conversion rate, so that these errors were called 'dynamic errors'. These errors always appeared in pairs: a long code followed by a short code at very particular transitions. This paper is discussing the root source and techniques that fix the problem. These techniques were implemented and verified on silicon.

In this paper we present a signal processing methodology for sharing symmetric keys in personal area networks. Symmetric encryption and decryption are commonly used because of limitations in computing power and energy consumption. However, key sharing still imposes challenges regarding usability, computational complexity of algebraic key exchange algorithms, and security. Our approach is that keys are generated locally on devices by shaking them, and that the keys are equal if and only if the devices are shaken together. Based on practical assessments, we show that the key generation algorithm is able to generate keys from acceleration data with an average entropy of 13bit/key in 70% of the cases.

Several efforts are currently underway in investigating multiple-input multiple-output (MIMO) communication systems. MIMO technology promises high data rate wireless communications and/or the potentiality of exploiting transmit and receive diversity schemes. For WLAN (wireless local area networks) scenarios most investigations focus on OFDM (orthogonal frequency division multiplexing) transmission. In this paper we discuss spatial multiplexing in conjunction with optimal matched filtering and MMSE-equalization for SC/FDE (single carrier transmission with frequency domain equalization), a single carrier technology closely related to OFDM due to its frequency domain signal processing characteristics. We show that our MMSE equalizer derived in S. Reinhardt et al. (2006) can be further optimized by combining it with successive interference cancellation known from multiuser detection, which yields a novel structure for MIMO-SC/FDE systems similar to VBLAST (G.D. Golden et al., 1999), but adapted to SC/FDE and suited for transmission over frequency selective fading channels

In this paper we present a simple non-recursive frequency modulator which avoids the inherited memory and complexity of linearized continuous phase modulation (CPM). This novel approach decreases the system complexity and it improves the bit error performance compared to differentially decoded CPM signals. Furthermore we combine this method advantageously with a multiple input multiple output (MIMO) approach and we apply it to the single carrier transmission with frequency domain equalization (SC/FDE) technology. The overall system features constant envelope transmit signals, which are immune to the nonlinearity of the power amplifiers, a high spectral efficiency and a low complex equalization process at the receiver. After a description of the concepts we show simulation results for the non-block transmission single input single output (SISO) as well as for the MIMO case

Several efforts are currently underway in investigating multiple-input multiple-output (MIMO) communication systems. MIMO technology promises high data rate wireless communications and/or the potentiality of exploiting transmit and receive diversity schemes. For WLAN (wireless local area networks) scenarios most investigations focus on OFDM (orthogonal frequency division multiplexing) transmission. In this paper we discuss spatial multiplexing in conjunction with optimal matched filtering and MMSE-equalization for SC/FDE (single carrier transmission with frequency domain equalization), a single carrier technology closely related to OFDM due to its frequency domain signal processing characteristics. We show that the conventional MMSE-equalizer equation is not suited in conjunction with optimal matched filtering and we derive a novel MMSE structure

This paper investigates the behavior of two different slow fading models in cellular mobile radio systems. A one-dimensional and a two-dimensional model are compared with respect to their influence on the results of system level simulations for cellular systems. A disadvantage of most known slow fading models is that they cannot handle the correlation behavior of shadowing in two dimensions. Recently a new model, which provides a two-dimensional correlation property, was proposed by Cai and Giannakis (2003). We show that this model gives different results for system capacity and handover rates than the well known one-dimensional models.

In this paper we consider phase tracking algorithms for single carrier systems with frequency domain equalizers (SC/FDE), which make use of the concept of unique word (UW) blockwise extension instead of the classical concept of cyclic prefix (CP) like it is used for example in OFDM (orthogonal frequency division multiplexing). Very similar to IEEE 802.11a like OFDM systems the overall baseband processing performance of SC/FDE systems largely depends on the design of channel estimation and synchronization algorithms. Some of the synchronization tasks are very straight forward, but the algorithm developer has a lot of freedom for the design of the channel estimator, and for the design of the residual carrier frequency offset tracking procedure (which we call phase tracking). In this paper we focus on the design and performance of unique word based phase tracking algorithms, which are being compared with pilot carrier based phase tracking algorithms for IEEE 802.11a like OFDM systems.

In this work we investigate the effect of convolutional coding, interleaving and Viterbi decoding, as they are used for OFDM (orthogonal frequency division multiplexing) in the IEEE 802.11a standard, for a single carrier system with frequency domain equalization (SC/FDE). We will characterize the effect of decision length (maximal length of the trellis) and code rate in terms of bit error behavior, reliability of soft decisions and implementation effort for the classical Viterbi algorithm as well as for the soft output Viterbi algorithm (SOVA). The (coding) parameters of the simulated system are adapted to the IEEE 802.1 Ia standard for WLAN applications. In detail we will elaborate on the advantages of pilot symbols, that are multiplexed into the data stream as a method to reduce the necessary decision length as well as to reduce the implementation effort and to improve the SOVA output. It will be demonstrated that the use of pilot symbols shows significant advantages that dominate the disadvantages.

We investigate the use of a known symbol sequence - a so-called unique word (UW) - instead of the well-known cyclic prefix (as it is used in OFDM, orthogonal frequency division multiplexing, systems) for a single carrier system with frequency domain equalization (SC/FDE). The considered SC/FDE system is similar to the one proposed in the IEEE 802.16.3c target group. It is shown that the UW fulfils the theorem of cyclic convolution required for efficient implementation of the SC/FDE structure. We elaborate on the advantages that result for equalization, channel estimation, and synchronisation. Compared to the advantages, the disadvantages in terms of BER behaviour and bandwidth efficiency are rather modest.