Frank Sjöberg

A rank-reduced LMMSE canceller for narrowband interference suppression in OFDM-based systems

We present a narrowband interference (NBI) canceller that suppresses spectral leakage in an orthogonal frequency-division multiplexing (OFDM)-based system caused by a narrowband (NB) signal. We assume that the spectrum of the NB signal is within the spectrum of the OFDM signal. This can be the case, e.g., on digital subscriber lines (DSL) and in new unlicensed frequency bands for radio transmission. The canceller makes linear minimum mean-square error estimates of the spectral leakage by measuring the NBI on a few modulated or unmodulated OFDM subcarriers. It uses a model of the NB signals power spectral density as a priori information. Using a frequency invariant design, it is possible to cancel NBI from signals that are changing their frequency location with significantly reduced complexity overhead. The operational complexity of the canceller can be lowered by using the theory of optimal rank reduction and using the time-bandwidth product of the NB signal. Analytical performance evaluations, as well as Monte Carlo simulations, show that, without perfect a priori information, this canceller can suppress the spectral leakage from a strong NB signal (e.g., with equal power as the OFDM signal) to well below the background noise floor for typical applications where it causes negligible signal-to-noise ratio and symbol error rate degradation.

Zipper: a duplex method for VDSL based on DMT

We present a new duplex scheme, called Zipper, for discrete multitone (DMT)-based very high bit-rate digital subscriber line (VDSL) systems on copper wires. This scheme divides the available bandwidth by assigning different subcarriers for the upstream and downstream directions. It has high flexibility to divide the capacity between the up and downstream, as well as good coexistence possibilities with other systems such as ADSL. Simulation results show the high bit-rate performance in different environments such as mixed ADSL and VDSL traffic under radio frequency interference and with different background noise sources.

Digital RFI suppression in DMT-based VDSL systems

We propose a method for suppressing radio frequency interference (RFI) in discrete multitone (DMT) based very high bit rate digital subscriber line (VDSL) systems. The method operates in the frequency domain of a DMT system. First, we derive a model of how an unknown narrow-band RF signal is mapped onto the DMT carriers. Then, by measuring the RFI on a few unused DMT carriers we are able to subtract RFI estimates from every modulated subcarrier. Simulation results show that this method, applied to an RFI signal with the same average power as the VDSL signal, suppresses the RFI with 40-50 dB, which reduces the average SNR loss from about 20 to less than 0.3 dB.

Asynchronous Zipper [subscriber line duplex method]

Previously the authors presented a novel duplex method for very high bit-rate digital subscriber lines (VDSL) called Zipper. With this method all VDSL-modems on different wires in the same bindergroup have to be time-synchronized to avoid near-end cross-talk (NEXT). We describe a method which enables Zipper to run in a time-asynchronous mode. By introducing pulse-shaping in the transmitter and windowing in the receiver the NEXT is almost completely suppressed even though the synchronization between modems on neighboring lines is skipped. The remaining NEXT and efficiency loss due to pulse-shaping and windowing results in only a small bit-rate performance loss, typically less than 10% compared to the time-synchronized Zipper. However, with the new freedom of optimizing the lengths of the cyclic suffices with asynchronous Zipper, there may even be a small improvement in bit-rate performance for short wires.

A low complexity threshold detector making MLSD decisions in a multiuser environment

We present a pipelined implementation of an iterative threshold detector, which makes the same decisions as a maximum likelihood sequence detector (MLSD) on some, but not necessarily all, bits. We apply this as a first stage detector in a direct sequence code division multiple access (DS-CDMA) interference-limited multiuser system. In combination with the single-user matched filter (MF) detector we obtain improved performance and lower complexity than with the decorrelating receiver for a limited number of simultaneous users, e.g. up to 25 users with a spreading factor of 127.

A noise-tolerant group delay estimator applied to dispersion measurement in gases

In this paper we present a model-based group velocity estimator, that can be used to measure speed of sound in ultrasonic pulse-echo systems as a function of ultrasound frequency. The estimation of group velocities involves numerical differentiation of the phase difference. In the presence of noise, this becomes numerically unstable. The model-based approach presented herein, shows better tolerance to experimental noise. The performance of the estimator is evaluated with simulations as function of pulse bandwidth and SNR. Finally, the estimator applied to real data and compared with other methods for measuring speed of sound in Ethane and Oxygen.

Performance evaluation of the Zipper duplex method

We present performance results for a new duplex scheme, called Zipper, for DMT-based VDSL (very high bit rate digital subscriber lines) systems on copper wires. This scheme divides the available bandwidth by assigning different subcarriers for the upstream and downstream direction. It has a high flexibility to divide the capacity between the up- and downstream as well as good coexistence possibilities with other systems such as ADSL (asymmetrical digital subscriber line). Simulation results show a high bit rate performance in different environments such as mixed ADSL and VDSL traffic, under radio frequency interference, and with different background noise sources.

Self-synchronizing a DMT-based VDSL system

We present an algorithm for self-synchronizing all modems in a discrete multitone (DMT)-based very high-speed digital subscriber line (VDSL) system using the Zipper duplex method. This solves the problem with non-orthogonal near-end crosstalk (NEXT) that appears in systems with unsynchronized modems. The algorithm we present runs autonomously in each VDSL-modem. It uses the auto-correlation of the DMT-signal to determine the frame-offset of other users, and adjusts the own frame-timing to be better aligned with the other users. With our method all modems will be self-synchronized to within a small fraction of the total DMT frame-length. This self-synchronization suppresses the NEXT to a level far below the background noise-floor. This means that our self-synchronized system has the same performance as a system where all modems are perfectly synchronized to a master clock.

Advanced signal processing for digital subscriber lines

The recent deployment of digital subscriber line (DSL) technology around the world is rapidly making broadband access for the mass consumer market a reality. The ever-growing customer demand for higher data rates has been fueled by the popularity of applications like peer-to-peer (P2P) filesharing networks and video-streaming and high-definition television (HDTV). DSL technology allows telephone operators to getmaximum leverage out of their existing infrastructure by delivering broadband access over existing twisted-pair telephone lines. At the heart of DSL lies a plethora of signal processing techniques which enable such high-speed transmission to be achieved over a medium originally designed with only voice-band transmission in mind. These advanced signal processing techniques address many challenges that exist in DSL networks today, such as the near-end and far-end crosstalk (NEXT/FEXT), impulse noise, peak-toaverage-power ratio (PAR), intersymbol and intercarrier interference (ISI/ICI), radio-frequency interference (RFI), and so forth. The goal of this special issue is to discuss the state-of-the-art and recent advances in signal processing techniques for DSL.