Methods for the transfer function design of multiple narrow-band bandpass sigma-delta modulators

This paper presents the design analysis of novel tunable narrow-band bandpass sigma-delta modulators, which can achieve concurrent multiple noise-shaping for multi-tone input signals. Four different design methodologies based on the noise transfer functions of comb filters, slink filters, multi-notch filters and fractional delay comb filters are applied for the design of these multiple-band sigma-delta modulators. The latter approach utilises conventional comb filters in conjunction with FIR, or allpass IIR fractional delay filters, to deliver the desired nulls for the quantisation noise transfer function. Detailed simulation results show that FIR fractional delay comb filter-based sigma-delta modulators tune accurately to most centre frequencies, but suffer from degraded resolution at frequencies close to Nyquist. However, superior accuracies are obtained from their allpass IIR fractional delay counterpart at the expense of a slight shift in noise-shaping bands at very high frequencies. The merits and drawbacks of each technique for the various sigma-delta topologies are assessed in terms of in-band signal-to-noise ratios, accuracy of tunability and coefficient complexity for ease of implementation.

Increasing the variability of centre frequency locations in multiple-band sigma-delta modulators via the use of fractional delay filters

This paper presents the design analysis of novel tunable narrow-band bandpass sigma-delta modulators, that can achieve concurrent multiple noise-shaping for multi-tone input signals. This approach utilises conventional comb filters in conjunction with FIR, or allpass IIR fractional delay filters, to deliver the desired nulls for the quantisation noise transfer function. Detailed simulation results show that FIR fractional delay comb filter based sigma-delta modulators tune accurately to most centre frequencies, but suffer from degraded resolution at frequencies close to Nyquist. However, superior accuracies are obtained from their allpass IIR fractional delay counterpart at the expense of a slight shift in noise-shaping bands at very high frequencies.

A Single-Frame Super-Resolution Innovative Approach

Super-resolution refers to the process of obtaining a high resolution image from one or more low resolution images. In this work, we present a novel method for the super-resolution problem for the limited case, where only one image of low resolution is given as an input. The proposed method is based on statistical learning for inferring the high frequencies regions which helps to distinguish a high resolution image from a low resolution one. These inferences are obtained from the correlation between regions of low and high resolution that come exclusively from the image to be super-resolved, in term of small neighborhoods. The Markov random fields are used as a model to capture the local statistics of high and low resolution data when they are analyzed at different scales and resolutions. Experimental results show the viability of the method.