Brian Berscheid

A Fast-Converging Equalizer for Upstream DOCSIS Channels

This paper proposes a method of increasing the efficiency of the CATV plant in the upstream direction. The general approach is to reduce the convergence time of the equalizer to allow a shorter training sequence. Specifically, the upstream channel is modeled as a postdetection filter. A short unique word is inserted near the beginning of the preamble to estimate the coefficients of the postdetection filter. The postdetection filter is then crudely inverted to estimate a few key tap weights of the equalizer. The equalizer is initialized with these estimated tap weights before adaptive updating is turned on. The method is evaluated for the echo-laden DOCSIS CATV upstream channel with a signal-to-noise ratio (SNR) of 25 dB. Plots of the probability of the modulation error ratio exceeding 19- and 22-dB values versus the length of the training sequence show that the performance is comparable to an unseeded equalizer with a training sequence that is about 50 symbols longer.

An Economical, ISI-Immune Frequency Offset Estimator for DOCSIS Upstream Channels

This paper discusses the design and implementation of frequency offset estimation algorithms for DOCSIS upstream channels. A cost-effective estimator which approaches the Cramer-Rao bound for high SNRs is derived. The effect of ISI generated by upstream micro-reflections in typical cable networks is considered, and a condition upon the transmitted preamble sequence which guarantees unbiased estimation is presented. It is shown that the proposed estimator is unbiased for a wide range of preamble sequences, which is generally not the case for burst frequency estimators. This flexibility may be utilized by selecting a broadband preamble which is suitable for performing channel estimation and frequency offset estimation simultaneously, thereby increasing the efficiency of the upstream channels.

A Novel Iterative OFDMA Channel Estimation Technique for DOCSIS 3.1 Uplink Channels

This paper presents an orthogonal frequency division multiple access (OFDMA) channel estimation technique that jointly considers the effects of coarse timing error and multipath propagation. Many conventional approaches only consider an optimistic scenario where timing synchronization is perfect and each of the channel delays is an integer number of system samples. In realistic scenarios timing offsets and echo delays are not integer multiples of the system’s sampling period. This leads to poor estimation and consequently reduces the system’s overall performance. This paper proposes a novel iterative channel estimation technique, which considers the practical scenario of fractional timing error and nonsample-spaced echo delays. The proposed method does not require channel state information (e.g., second-order statistics of the channel impulse responses or the noise power). Moreover, timing error can be conveniently obtained with the proposed technique. Simulation results show that, when comparing OFDMA channel estimation techniques under realistic data over cable service interface specification 3.1 channel conditions, the proposed algorithm significantly outperforms all conventional methods known to the authors.

Digital interpolating peak locator

Locating the position of the peak of an analog signal after it has been sampled is a challenging task with wide-ranging applications. The conventional methods of performing this operation involve either upsampling the digital signal or fitting a parabolic curve to the digital samples. This paper proposes a technique for finding the approximate location of the peak in an analog signal after it has been sampled. The technique involves transforming the sampled data into the logarithm domain and then fitting a parameterized function that is characteristic of the analog signal to the two points neighbouring the peak. The location of the peak is calculated from the parameters obtained in the fitting process. Simulation results show that the proposed estimator significantly outperforms the conventional estimators at a comparable or lower hardware cost.

Iterative Channel Estimation for DOCSIS 3.1 Uplink Channels

This paper presents an OFDMA channel estimation technique that jointly considers the effects of coarse timing error and multipath propagation. Many conventional approaches only consider an optimistic scenario where timing synchronization is perfect and each of the channel delays is an integer number of system samples. In realistic scenarios, timing offsets and echo delays are not integer multiples of the systems sampling period. This threatens sub-carrier orthogonality and causes leakage in the discrete Fourier transform (DFT)-based channel estimation method. Such leakage leads to poor estimation and consequently reduces the systems overall performance. Proposed in this paper is a novel iterative channel estimation technique that considers the practical scenario of fractional timing error and non sample-spaced echo delays. The proposed method does not require channel state information (e.g. second-order statistics of the channel impulse responses or the noise power). Simulation shows that, when comparing OFDMA channel estimation techniques under DOCSIS 3.1 realistic channel conditions, the proposed algorithm significantly outperforms all conventional methods known to the authors.

Optimizing Pulse Shaping Filter for DOCSIS Systems

This paper proposes a cost efficient nearly linear phase approximation to an square root raised cosine (SRRC) pulse shaping filter that satisfies the out-of-band spectral constraints of DOCSIS down-stream channels. A nearly linear phase filter structure is converted to an SRRC filter using a weighted and sampled least-squares criterion to fit the magnitude and phase responses. To ensure stability, the search for optimum coefficients is constrained using the Steigliz-McBride (SM) and Gauss-Newton (GN) methods, which unfortunately also eliminates sets of stable coefficients, one of which could be and probably is the optimum. To expand the sets of coefficients that produce stable filters in the SM and GN methods, the transfer function is parameterized in a special way. The effectiveness of the proposed filter is verified and compared with other approaches.