Abstract: Signal conversion is implemented employing a memory system operating as a look-up table that stores a plurality of sets of output samples associated with each of a plurality of respective input samples. The look-up table thus can generate a corresponding set of output samples in response to a given input sample, thereby emulating desired digital upsampling and delta-sigma modulation. The output samples can be aggregated, such as by multiplexing, to provide an output data stream at a desired sample rate.

Abstract: A unique method for chroma vertical upsampling used, for example, for conversion of the “4:2:0” format chroma information used in many applications of digital video, to the “4:2:2” or “4:4:4” format, is presented This conversion is required so that video encoders can effect the display of this chroma information with a minimum of visible artifacts The present invention carries out chroma vertical upsampling on a pixel by pixel basis This chroma vertical upsampling is performed as a function of the amount of motion associated with each pixel as detected between 2 or more fields, and the field, frame and progressive sequence characteristics of the incoming video signal data

Abstract: An interpolating sample rate converter that provides rate conversion by a factor of M/N comprising: an upsampler that upsamples an input data stream by a factor of M; a filter that shapes the upsampled data stream in accordance with a predetermined power spectrum; and a downsampler that downsamples the upsampled and shaped signal by a factor of N to produce an output data stream; wherein the upsampler and filter are implemented, at least in part, by a cascaded integrator-comb filter; and wherein the upsampling factor M is a natural number and the downsampling factor N is a rational number but not necessarily a natural number.

Abstract: We describe a multiresolution extension to maximum intensity projection (MIP) volume rendering, allowing progressive refinement and perfect reconstruction. The method makes use of morphological adjunction pyramids. The pyramidal analysis and synthesis operators are composed of morphological 3-D erosion and dilation, combined with dyadic downsampling for analysis and dyadic upsampling for synthesis. In this case the MIP operator can be interchanged with the synthesis operator. This fact is the key to an efficient multiresolution MIP algorithm, because it allows the computation of the maxima along the line of sight on a coarse level, before applying a two-dimensional synthesis operator to perform reconstruction of the projection image to a finer level. For interpolation and resampling of volume data, which is required to deal with arbitrary view directions, morphological sampling is used, an interpolation method well adapted to the nonlinear character of MIP. The structure of the resulting multiresolution rendering algorithm is very similar to wavelet splatting, the main differences being that (i) linear summation of voxel values is replaced by maximum computation, and (ii) linear wavelet filters are replaced by nonlinear morphological filters.

Abstract: The intonation of speech is modified by an appropriate combination of resampling and time-domain harmonic scaling. Resampling increases (upsampling) or decreases (downsampling) the number of data points in a signal. Harmonic scaling adds or removes pitch cycles to or from a signal. The pitch of a speech signal can be increased by combining downsampling with harmonic scaling that adds an appropriate number of pitch cycles. Alternatively, pitch can be decreased by combining upsampling with harmonic scaling that removes an appropriate number of pitch cycles. The present invention can be implemented in an automated speech-therapy tool that is able to modify the intonation of prerecorded reference speech signals for playback to a user to emphasize the correct pronunciation by increasing the pitch of selected portions of words or phrases that the user had previously mispronounced.

Abstract: The concept of biorthogonal partners has been introduced recently by the authors. The work presented here is an extension of some of these results to the case where the upsampling and downsampling ratios are not integers but rational numbers, hence, the name fractional biorthogonal partners. The conditions for the existence of stable and of finite impulse response (FIR) fractional biorthogonal partners are derived. It is also shown that the FIR solutions (when they exist) are not unique. This property is further explored in one of the applications of fractional biorthogonal partners, namely, the fractionally spaced equalization in digital communications. The goal is to construct zero-forcing equalizers (ZFEs) that also combat the channel noise. The performance of these equalizers is assessed through computer simulations. Another application considered is the all-FIR interpolation technique with the minimum amount of oversampling required in the input signal. We also consider the extension of the least squares approximation problem to the setting of fractional biorthogonal partners.

Abstract: A circuit 30 for upsampling and upconverting a high rate signal that is divided into M in phase (I) symbols and M quadrature (Q) symbols. A Nyquist filter 32 upsamples by a factor of k each of the 2M symbols in parallel during one system clock period (CP). The filter 32 has a plurality of 2kM filter components 40, 42 , that each provides a continuous output. A plurality of multipliers 50, 52 each upconverts a filter component output with a carrier wave signal 46, 48 that is output from a numerically controlled oscillator 44 . A plurality of adders 54 each adds the output of two multipliers 50 to recombine corresponding I and Q samples to output kM samples during a CP. For continuous phase modulation, N parallel bits are input into the filter 32 , upsampled in one CP, and accumulated and modulated 82 in parallel in one CP. For analog processing, M (I) and M (Q) symbols are input into an FIR filter 77 a, 77 b for upsampling, and decimated at a MUX/DAC block 78 for subsequent analog upconversion.

Abstract: The invention describes a method and device (4) that compensate sampling instant errors in digital receivers. For improving performance in a cellular communication system, such as WCDMA, information from the path delay estimator (44) is used for improving the robustness of a sampling adjustment. The received signal y(t) is downconverted (42) and sampled, yielding a digital baseband signal yn. The digital baseband signal yn is then, directly or after an oversampling, fed to a path delay estimator (44), which provides information about the power delay profile (PDP). By using knowledge of the receiver filters and the transmitter filters used in the communication system and of the PDP, the optimal sampling instant is computed and sampling instants are adjusted. Thus the receiver performance in high-performance modes such as HSDPA in WCDMA is enhanced, while a moderate sampling rate and power consumption are maintained. According to the invention, a matched-filtering-based peak instant detection improves the robustness and additionally allows computing a metric indication that indicates whether the timing adjustment is advisable in an observed situation. Additionally, the invention is used in sampling timing selection based on either ADC clock adjustment or downsampling phase selection.

Abstract: Electronic imaging systems such as image rendering and printing devices are providing higher quality with improved technologies. Physical hardware limitation has been pushed to a newer level. By utilizing some of these new imaging technologies, three improvements are presented in this document, including banding reduction using pulse width modulation, multitoning using DBS screens, and hybrid screen design. In Ch. 1, we propose a system to reduce electrophotographic laser printer banding artifacts due to optical photoconductor drum velocity fluctuations. The drum velocity fluctuations are sensed with an optical encoder mounted on the drum axis. Based on the line-to-line differential encoder count, we modulate the laser pulse width to compensate fluctuations in development that would otherwise occur. We present an analysis of the system, including the compensation algorithm that determines the desired pulse-width as a function of differential encoder count. Characterization of the system is based on printing, scanning, and processing a special test page that yields information about line-spacing and absorptance fluctuations. This data is synchronized with the encoder count signal that was recorded during the printing of the test page. The experimental results show the efficacy of the system. In Ch. 2, we propose a methodology for multilevel screen design using Direct Binary Search. We define a multitone schedule, which for each absorptance level specifies the fraction of each native tone used in the multitone cell. Based on the multitone schedule, multitone patterns are designed level-by-level by adding native tones under the stacking constraint. At each level, the spatial arrangement of the native tones is determined by a modified DBS search. We explore several different multitone schedules that illustrate the image quality tradeoffs in multitone screen design. In Ch. 4, we propose a screen design method that generates a new hybrid class of halftones---stochastic dispersed-dot in highlights and periodic clustered-dot in midtones. The traditional solution for generating hybrid halftones is the so-called supercell technique. While maintaining the spatial resolution, supercell increases the number of the output levels of a periodic micro-clustered-dot screen by adding dots asynchronously based on the order described in a dispersed-dot macroscreen. The traditional supercell screen employs a Bayer macroscreen and consequently results in visible false textures in highlight area. Simply replacing the Bayer macroscreen with a stochastic macroscreen yields the maze-like artifact due to the embedded upsampling process. In this work, we propose a screen design method which optimizes the employed multiple macroscreens and microscreens simultaneously using DBS. The resulting final screen is decomposable for a memory-efficient implementation. We also propose an adaptive microcell-based edge enhancement algorithm for further improvement. Several results are demonstrated using the proposed screens.

Abstract: PROBLEM TO BE SOLVED: To provide a delta sigma type multi-bit A/D converter 11 for oversampling an inputted analog signal with a frequency higher than a desired sampling frequency in a ΔΣ modulating part 12 first and converting the signal into a 1 bit signal, then converting the 1 bit signal into a multibit signal of the desired sampling frequency in a digital filter part 13, which can cope with a wideband signal. SOLUTION: The digital filter part 13 is constituted by a 2 bit decoder 18 and a low pass filter 19, and the 2 bit decoder 18 converts the two of the 1 bit signals into multi-bit signals at a time. Therefore, a clock CK2 to an LPF 19 enables a clock CK1 to the ΔΣmodulating part 12 to become a signal divided into 1/2 in a divider circuit 20 and, in dealing with the wideband signal, a clock frequency can be reduced to 1/2, and the converter 11 can easily cope with the wideband signal. COPYRIGHT: (C)2004,JPO&NCIPI

Abstract: A design procedure for data compression and equalization for digital communication systems is developed based on the multirate sampled-data H∞ control theory. The procedure provides transmitting/receiving filters so as to minimize the error between the original signal and the received signal with a time delay, and to reduce the noise added to the channel. While the system is multirate and infinite-dimensional due to up-and downsampling and a delay, the design problem can be reduced to a finite-dimensional discrete-time problem using the lifting and the FSFH (fast-sample and fast-hold) approximation. Numerical examples are presented to illustrate the effectiveness of the proposed method.

Abstract: In this paper we discussed issues related to resampling speech signal at arbitrary frequency by using interpolation methods. The implementation of four resampling methods, 1. direct interpolation, 2. Lagrange interpolation, 3. sine interpolation and, 4. Taylor series method, is presented. These methods have been tested with some speech data and various resampling frequencies. The quality of the resampled speech signals is analyzed and evaluated by human listening. The experiment results showed that either for upsampling and downsampling, sine and Lagrange methods generate additive high-frequency noise like metal sounds, but the direct and Taylor methods do not have some problem. The resampled speech by the direct and Taylor methods sounds more natural than that by sine and Lagrange methods.

Abstract: Audio processing using a video rasterizer allows simultaneous display of both audio and video data on the same raster display. Audio samples are clocked into a buffer memory, such as a first-in, first-out (FIFO) buffer, by an audio clock and clocked out of the buffer memory by a rasterizer (system) clock to allow the audio data to cross over from the audio clock domain to the video clock domain. The audio data is then upsampled by a sample ratio converter/interpolator before being processed by a video display engine. The video display engine includes a polyphase filter for upsampling display data for input to a rasterizer from which the data is read for display.

Abstract: PROBLEM TO BE SOLVED: To provide an image encoder wherein processing of reducing production of noise is performed on the encoder side without using a mask image so as to realize a transmission color function even in the case of irreversible encoding and to provide an image decoder SOLUTION: A gradation shift section 35 shifts the gradation of a transmission color of an original image, and an encoding section 36 applies irreversible encoding to the image whose gradation is shifted A transmission color discrimination section 39 discriminates the transmission color of the decoded image on the basis of only luminance information and restores the contour of an object with fidelity without causing reduction in the resolution by subsampling in an upsampling section 37 so as to obtain an image with high quality and less noise COPYRIGHT: (C)2003,JPO

Abstract: Programmable transceivers are essential for efficient implementation of communication systems. This calls for implementation of various flexible signal processing blocks in the physical layer of a network. Rate conversion is an important functional block in programmable digital modems. In this paper, the implementation of an asynchronous interpolator for a digital modem is discussed and performances of three different types of interpolation filters are compared.

Abstract: A sample rate conversion system developed to implement a rate change of M7N using a very efficient design implementation. The sample rate conversion system of the present invention is implemented as a CIC-based interpolating sample rate convertor with noise-shaped control of the N value. For a decimator, noise-shaped control of the M value is utilized. In the interpolator, the N value is the correct value on average, but demonstrates instantaneous errors ('non-uniform' resampling) that are corrected through noise-shapping. The CIC SRC implementation capitalizes on the fact that the outputs of the CIC that are discarded during downsampling need not be calculated by the CIC in the first instance. The combination of the computational simplicity of CIC SRC with noise-shaped, non-uniform resampling performs the sample rate conversion very economically and facilitates conversion between a plethora of sample rates at the input and output without requiring the various filters to be explicitly formulated. A method for sample rate conversion is also described.