Advanced equalization techniques for DMT-based systems

Digital subscriber line (DSL) technology is one of the fastest growing broadband internet access media. Whereas asymmetric DSL (ADSL) already offers data rates of a few megabits per second, next-generation ADSL2+ and VDSL promise even higher bit rates to support so-called triple play (high-quality video, voice and high-speed data). The use of a large bandwidth over the phone line (up to 12 MHz for VDSL) induces impairments, such as severe channel distortion, echo, narrow-band radiofrequency interference (RFI) and crosstalk from other DSL systems. DSL communication makes use of so-called discrete multitone (DMT) modulation, supplemented with advanced digital signal processing algorithms, to tackle these impairments and serve a maximum number of customers. In this thesis, we focus on channel equalization and RFI mitigation algorithms that outperform existing algorithms in terms of bit rate. DMT equalization is typically done by means of a channel-shortening timedomain equalizer (TEQ) and a one-tap frequency-domain equalizer per used tone. We first derive a TEQ design criterion that corresponds to true bit rate maximization, an unsolved problem so far. The bit rate maximizing TEQ (BMTEQ) can also be posted as the solution to an iteratively-reweighted separable non-linear least-squares (LS) problem. This alternative formulation gives rise to an LS-based framework for DMT equalization. It also leads to an adaptive optimization algorithm that allows to track changing channel conditions and, as such, integrates well with the so-called seamless rate adaptation functionality of ADSL2/ADSL2+. A complete range of adaptive bit rate maximizing equalizers then opens up, offering a flexible trade-off between computational and memory cost while keeping bit rate performance at the same level. The memory-efficient adaptive BM-TEQ and the computationally efficient so-called adaptive per-tone equalizer (PTEQ) form extreme cases of the set of bit rate maximizing per-group equalizers, which equalize groups of tones in parallel. A tone-grouping solution with as few as 4 equally-sized tone groups gives close-to-PTEQ performance in harsh, RFI-dominated scenarios. To further enhance the RFI mitigation capabilities, receiver windowing offers a cheap and effective solution. We extend the BM-TEQ design criterion and optimization algorithm to incorporate an optimal receiver window and perform joint equalization and RFI mitigation. Several important special cases are also discussed, such as the bit rate maximizing windowing-only case, the per-group and per-tone combined equalizer and windowing case and the BM-TEQ design in combination with a fixed window. Finally, we provide low-cost linear and decision-feedback BM-TEQ and PTEQ extensions that exploit so-called frequency-domain transmit redundancy from pilot, unused and/or feedback tones. These extensions accommodate infinite impulse response channels as well as a reduced DMT cyclic prefix length