Fig 1 - uploaded by Ying-Ren Chien
Content may be subject to copyright.
(a) Simplified baseband THP transmission model with imperfect CSI and (b) its equivalent form. The crosstalk interference and the corresponding cancellers are not shown.
Source publication
The interdependence among multiple channels and the interaction between timing and equalization loops bring new challenges to the design of a multi-channel symbol timing recovery (STR) system for 10GBASE-T. In addition, the nonlinear Tomlinson-Harashima precoding (THP) technique used in the 10GBASE-T system is vulnerable to the imperfect channel st...
Contexts in source publication
Context 1
... model of the CSI mismatch between the far-end trans- mitter and local-end receiver of a baseband transmission system using THP is shown in Fig. 1(a). For simplicity, we suppress the subscript, which denotes the channel index, of each subchannel in this section. The data sequences are precoded into the channel input sequences by the THP. There is a feedback filter and a modulo-2M device in a TH precoder. The transfer function of the THP's feedback part is denoted by in the Z-domain ...
Context 2
... is the transfer function of the FFE. There is also another modulo-2M device at the receiver side that precedes the slicer to remove the precoding sequences added by the modulo-2M device at the transmitter side. To explicitly indicate the mismatch term caused by the imperfect CSI, we translate the system model into an equivalent form as shown in Fig. 1(b). Note that the transfer function and are obtained during the training mode. The coefficients of the feedback part of THP are fixed and the coefficients of the FFE may be adaptively updated during the data mode. Ideally, the channel should be perfectly equalized by the THP and FFE, ...
Context 3
... robustness of the proposed ASP scheme and the effec- tiveness of the proposed timing strategy in the data mode are examined as shown in Fig. 10, where case-i, , rep- resents a total of wire pairs suffering from imperfect CSI, from the 1st to the th. The quantity of CSI mismatch , which is de- fined in (1), is varied from 0 to . Some interesting results are summarized as ...
Context 4
... The timing recovery and equalization loop interfere with each other when the DLLs are ON. As the dashed lines show in Fig. 10, the resulting DP-SNR of the proposed ASP scheme is nearly the same as that of the conventional schemes for these four cases. eye diagram of the slicer input (assuming that the CSI is perfect, i.e., = 0). 2) Turning the DLLs ON only benefits the scheme but harms all the other STR schemes in case-1. For the scheme, the higher , the ...
Context 5
... CSI are un- known in advance. For instance, the proposed ASP scheme is more robust than the scheme in general. The av- eraged DP-SNR is improved by about 1.40 dB, 1.22 dB, 0.61 dB, and 0 dB for case-1 to case-4, respectively, when . Similar results are observed in both case-2 and case-3: the ASP scheme is superior to the and schemes shown in Fig. 10(b) and superior to the , and schemes shown in Fig. 10(c). 4) The limitation of the ASP approach can be observed in case-4, in which all wire pairs suffer from imperfect CSI and the performance of all schemes is equally poor. How- ever, the proposed ASP approach provides an alternative. It improves the averaged DP-SNR in general and has ...
Context 6
... ASP scheme is more robust than the scheme in general. The av- eraged DP-SNR is improved by about 1.40 dB, 1.22 dB, 0.61 dB, and 0 dB for case-1 to case-4, respectively, when . Similar results are observed in both case-2 and case-3: the ASP scheme is superior to the and schemes shown in Fig. 10(b) and superior to the , and schemes shown in Fig. 10(c). 4) The limitation of the ASP approach can be observed in case-4, in which all wire pairs suffer from imperfect CSI and the performance of all schemes is equally poor. How- ever, the proposed ASP approach provides an alternative. It improves the averaged DP-SNR in general and has zero improvement only in the worst case. 5) Turning the ...
Citations
... Echo cancellation has been applied in a wide range of applications, such as hands-free telecommunication [1]- [3], assistive hearing devices [4], [5], and full-duplex wireline communications [6], [7]. Generally, an echo cancellation approach involves a finite impulse response (FIR) filter to characterize the echo path of an unknown system. ...
Determining an effective way to reduce computation complexity is an essential task for adaptive echo cancellation applications. Recently, a family of partial update (PU) adaptive algorithms has been proposed to effectively reduce computational complexity. However, because a PU algorithm updates only a portion of the weights of the adaptive filters, the rate of convergence is reduced. To address this issue, this paper proposes an enhanced switching-based variable step-size (ES-VSS) approach to the M-max PU least mean square (LMS) algorithm. The step-size is determined by the correlation between the error signals and their noise-free versions. Noise-free error signals are approximated according to the level of convergence achieved during the adaptation process. The approximation of the noise-free error signals switches among four modes, such that the resulting step-size is as close to its optimal value as possible. Simulation results show that when only a half of all taps are updated in a single iteration, the proposed method significantly enhances the convergence rate of the M-max PU LMS algorithm.
For power-limited mobile devices, how to prolong battery life is an important issue. The voice communication needs to implement high computation complexity echo cancellation to obtain a satisfied voice quality. Partial-update-based adaptive algorithms are known solutions to reduce the computation complexity for long-tap adaptation, such as echo cancellation. However, the convergence rate is decreased due to the partial update algorithm only updates part of weights of the adaptive filters. In this paper, we propose an enhanced switching-based variable step-size approach for the M-max partial update LMS algorithms. During the initial stage, the errors are dominated by the mismatch between the tap weights of the adaptive filter and its ideal weights. We correlate the squared error signals with polynomial of error signal to gain a large step-size; on the other hand, during the steady state, the errors mainly come from the additive noise. Therefore, we switch the correlation to the other mode so that the effect of noise can be eliminated. This can be done by correlating the error signals with a delayed version of error signals and hence a small step-size is obtained during the late stages. Simulation results show that when only one fourth of all taps are updated in one iteration, the proposed method significantly enhances the convergence rate of the M-max least-mean-square (LMS) algorithms.
Partial update (PU) techniques efficiently reduce computational complexity, especially for long-tap applications such as echo cancelation problems. However, periodic signals are known to induce instability for many PU algorithms, but not the stochastic PU (SPU) algorithm. For a small enough step-size, the SPU algorithm guarantees stability. However, it suffers a slow convergence speed. This paper proposes a non-uniformly distributed SPU (NSPU) least-mean-square (LMS) algorithm, which updates the taps in a non-uniform fashion such that a bigger tap gains a higher updating probability. This can be accomplished by randomly combining a “data independent” (SPU) with a “data dependent” (maximum partial output) PU criteria. Our approach not only preserves the stability of the SPU LMS algorithm but also enhances the convergence speed with a lower hardware cost. Simulation results show that our NSPU LMS algorithm demonstrates significant improvements when only one-sixteenths of total taps are updated at each iteration.
