André Neubauer

Differential correlation for Galileo/GPS receivers

Current enhanced sensitivity GPS receivers developed for moderate indoor reception increase the observation period with noncoherent integration of the envelopes of coherently integrated predetection samples. The paper analyzes a novel method, differential correlation, where each current coherently integrated predetection sample is multiplied with the complex conjugate of the previous predetection sample. The products are then accumulated. The deterministic signal and stochastic noise components, together with their probability density functions and moments, are derived algebraically for differential correlation and conventional noncoherent integration, while also considering signal and implementation nonidealities. The paper shows that differential correlation can offer an average sensitivity gain over conventional noncoherent integration of around 1.5 dB.

MLSE based detection for GFSK signals with arbitrary modulation index

Due to its widespread use in cordless indoor systems like DECT, Bluetooth etc., Gaussian frequency-shift keying (GFSK) is an important modulation method. In this paper we present an approach to apply maximum likelihood sequence estimation (MLSE) to GFSK modulated signals. Using a linearized model of the modulation, we develop a pseudo-coherent detection scheme based on the estimation of the transmitter state. With this approach a joint detection and estimation scheme can be realized that allows the application of a Viterbi type equalizer to GFSK signals with arbitrary modulation indices. Simulation results show a gain of up to 7 dB with respect to conventional approaches.

A Digital Receiver Architecture for Bluetooth in 0.25-

A digital receiver architecture for short-range communications systems like Bluetooth is presented. The architecture is tailored to a highly integrated Bluetooth single-chip integrated circuit (IC) and can easily be adapted to other communications systems using a Gaussian frequency-shift keying (GFSK ) modulation scheme. The single-chip IC integrates the complete digital baseband and radio frequency (RF) functionality on a single die and is realized in a 0.25-mum complementary metal-oxide-semiconductor (CMOS) technology targeted for cost efficiency. The superior performance of this digital receiver architecture compared to the state-of-the-art short-range communications receivers is shown. Simulation and measurement results are presented showing a receiver sensitivity of 87 dBm and excellent co-channel and adjacent channel interference performance.

\mu

Location based services with Galileo/GPS satellite receivers integrated in mobile phones are expected to experience a substantial market growth. Satellite signals are usually received on the earth surface with very low signal energy. Attenuation, shadowing, and multipath fading in urban canyons and moderate indoor environments impose challenging acquisition conditions for the state-of-the-art enhanced sensitivity mobile satellite receivers. Residual frequency deviations between the down-conversion frequency and the received signal frequency due to unknown Doppler frequency shifts significantly reduce the signal acquisition sensitivity of current satellite receivers. This paper shows how the residual frequency deviation can be reduced incrementally towards zero much faster than the CDMA code synchronization be detected with sufficient probability. This paper furthermore derives the probability density functions for the signal detection using the adaptive frequency correction method and illustrates the acquisition sensitivity gain of around 4 dB for different Galileo and GPS satellite receiver configurations.

m CMOS Technology and Beyond

Conventional methods for signal detection in wireless receivers use well-established methods of signal processing either in the analog or digital domain. In this contribution a method for digital signal detection is presented. Using this method the detection of the received data is performed by using only the zero crossings of the hard-limited IF-signal. It will be shown that a superior performance compared to conventional zero crossing detectors can be obtained by applying methods of pattern recognition and the respective learning algorithms for signal detection. The method can be made very robust with respect to the nonidealities of the receiver using appropriate methods of digital signal processing. Thus significant improvements with respect to performance and implementation costs are obtained compared to classical solutions by shifting implementation complexity from analogue to the digital domain.

Adaptive frequency correction method for enhanced sensitivity CDMA acquisition

A number of modern wireless systems use modulation schemes like CPFSK, GMSK and DPSK, where the information to be transmitted is exclusively coded in the phase change of the RF-signal. Thus a phase reconstruction receiver can be used, which is based on a simple limiter for digitization. With respect to implementation costs and power consumption this is a very attractive receiver concept. However, due to non-idealities of the RF-front end, the receiver signal can deviate from the ideal PSK-constellation. This can result in performance losses compared to a linear receiver with a high resolution A/D-converter. For the more sophisticated modulation schemes like MPSK, which will be extensively used for high-data rate systems, these performance losses can be severe. Thus, in this contribution, a digital compensation technique for a phase reconstruction receiver is proposed, which reduces the nonlinear distortion introduced by the limiter. Using this method an excellent performance can be obtained with a phase reconstruction receiver also for the more sophisticated modulation schemes.

Signal detection based on pattern classification for use in wireless CPFSK receivers

The representation of signals in spline signal spaces has several advantages compared to the general approach of working with band-limited functions. Among them are the finite support of B-splines, simple manipulations like differentiation, integration, etc. In this paper, the deterministic treatment of spline signals is extended to a stochastic analysis. The main focus is on the analysis of second-order characteristics of spline signals, i.e. auto-correlation and cross-correlation functions and sequences are calculated. The signal analysis as well as synthesis steps are analyzed and an algorithm for estimating the auto-correlation and cross-correlation functions and sequences is derived.

Phase reconstruction receiver with compensation of nonlinearities

Wir sind nun — endlich — am eigentlichen Ziel unserer Bemuhungen angelangt, namlich der Signaltheorie der irregularen Abtastung. Durch die bisherigen Betrachtungen haben wir das notwendige Rustzeug erworben, um die in diesem Kapitel besprochene Signaltheorie der irregularen Abtastung verstehen zu konnen. Zu diesem Zweck haben wir bereits eine Fulle von Ergebnissen in einer fur die Ubertragung auf den Fall der irregularen Abtastung gunstigen Form zusammengestellt. Im Folgenden wollen wir uns die eingefuhrte Theorie der Rahmen und insbesondere die bereits kennen gelernten Rekonstruktionsalgorithmen fur unsere Zwecke der signaltheoretischen Analyse sowie der Signalverarbeitung auf Basis der irregularen Abtastung zunutze machen. Wir beginnen wie bei der Behandlung der regularen Abtastung mit dem HILBERTRaum L2(ℝ) und speziell dem Paley-Wiener-Raum PWΩ ⊂ L2(ℝ) bevor wir auf periodisch fortgesetzte zeitbegrenzte Signale ubergehen. Da historisch in der mathematischen Literatur in diesem Zusammenhang so genannte ganze Funktionen vom exponentiellen Typ eine Rolle spielen, wagen wir vorher jedoch noch einen kleinen Ausflug in die Funktionentheorie.

Second-order statistics of stochastic spline signals

Wenn Signale und ihre zugehorigen Reprasentationen betrachtet werden, so ist zunachst die Festlegung der wesentlichen Signaleigenschaften notwendig, die bei der Signalreprasentation berucksichtigt werden mussen. Beispiele fur derartige Signaleigenschaften sind Frequenzband-Begrenztheit, Periodizitat et cetera. Hierzu werden Signale als Elemente eines Signalraums S aufgefasst; die fur die Reprasentation der Signale innerhalb des Signalraums S zu berucksichtigenden Eigenschaften lassen sich dann durch die geeignet gewahlten Eigenschaften von S fassen.

Die Signaltheorie der irregulären Abtastung

In diesem Abschnitt werden wir die grundlegenden Eigenschaften der FOURIER-Transformation zusammenstellen. Wir setzen dabei voraus, dass wir uns im HILBERT-Raum beziehungsweise LEBESGUE-Raum L2(ℝ) befinden, so dass alle Skalarprodukte und Normen entsprechend den Gleichungen 2.45 beziehungsweise 2.47 auf Seite 41 berechnet werden. Wie schon im letzten Kapitel bei der Diskussion der Distributionentheorie praktiziert werden wir zeitkontinuierliche Signale mit f bezeichnen, da diese Signale Funktionen (function) des LEBESGUE-Raums L2(ℝ) darstellen. Koeffizientenfolgen (coefficient) im HILBERTschen Folgenraum l2(ℝ) werden entsprechend mit c = |c j } bezeichnet. Wir verlassen somit die im letzten Kapitel verwendete allgemeine Bezeichnungsweise mit dem Symbol xum das Verstandnis zu erleichtern