Showing papers on "Quadrature mirror filter published in 2007"

ε-Separating Nonlinear Filter Bank and Its Application to Face Image Beautification

Abstract: For the purposes of separating the specific components of nonsteady signals, the e-separating nonlinear filter bank is proposed. Previously, the authors proposed a nonlinear digital filter called the e-filter. This filter effectively eliminates the small signal amplitude random noise superimposed on a signal with a large signal changing component. The e-filter bank proposed in this paper consists of these e-filters and separates the components by taking account of both the signal amplitude and frequency. Further, a filter system for beautifying facial images using this filter bank is proposed. The human facial image consists not only of large-signal components such as the frame, eyes, nose, and mouth, but also of small-signal high-frequency components such as fine wrinkles as well as small-signal medium-frequency components such as large wrinkles and blemishes. Additionally, ultrasmall-amplitude high-frequency components expressing the natural roughness of the skin also exist. The present filter bank can eliminate wrinkles and blemishes while maintaining the natural roughness of the skin, allowing beautification of the skin of facial images to be performed effectively. © 2006 Wiley Periodicals, Inc. Electron Comm Jpn Pt 3, 90(4): 52– 62, 2007; Published online in Wiley InterScience (www.interscience. wiley.com). DOI 10.1002/ecjc.20241

Low-Delay Nonuniform Pseudo-QMF Banks With Application to Speech Enhancement

Abstract: This paper presents a method for designing low-delay nonuniform pseudo quadrature mirror filter (QMF) banks. This method is motivated by the work of Li, Nguyen, and Tantaratana, in which the nonuniform filter bank is realized by combining an appropriate number of adjacent sub-bands of a uniform pseudo-QMF bank. In prior work, the prototype filter of the uniform pseudo-QMF bank was constrained to have linear phase and the overall delay associated with the filter bank was often unacceptably large for filter banks with a large number of sub-bands. This paper proposes a pseudo-QMF filter bank design technique that significantly reduces the delay by relaxing the linear phase constraints. An example in which an oversampled critical-band nonuniform filter bank is designed and applied to a two-state modeling speech enhancement system is presented in this paper. Comparison of the performance of this system to competing methods employing tree-structured, linear phase multiresolution analysis indicates that the approach described in this paper strikes a good balance between system performance and low delay

Direct conversion receiver and method for correcting phase imbalance therein

Abstract: A direct conversion receiver and a method for correcting phase imbalance therein is disclosed. An input signal is applied to an in-phase channel and a quadrature channel of the receiver. The input signal is processed by the direct conversion receiver to obtain an in-phase zero intermediate frequency (IF) signal in the in-phase channel and a quadrature zero-IF signal in the quadrature channel. The in-phase zero-IF signal and the quadrature zero-IF signal are filtered to obtain a fixed band signal. A phase imbalance correction value is obtained for the fixed-band quadrature zero-IF signal as a function of the frequency of the fixed-band in-phase zero-IF signal and the fixed-band quadrature zero-IF signal. The in-phase zero-IF signal and the quadrature zero-IF signal are sampled and the phase imbalance correction value is applied using an interpolation to the sampled quadrature zero-IF signal or to the sampled in-phase zero-IF signal to correct the phase imbalance in the direct conversion receiver.

Parametrization construction of biorthogonal wavelet filter banks for image coding

Abstract: We construct popular biorthogonal wavelet filter banks (BWFBs) having the linear phase and arbitrary multiplicity of vanishing moments (VMs). A novel parametrization construction technique, which is based on the theory of Diophantine equation, is presented and explicit one-parameter expressions of the BWFBs are derived. Using the expressions, any one-parameter family of BWFBs with different VMs can be constructed, and ten families, i.e., 5/7, 6/6, 9/7, 6/10, 5/11, 10/6, 13/7, 6/14, 17/11, and 10/18 families, are constructed here. The free parameter can be used to optimize the resulting BWFBs with respect to other criteria. In particular, in each family, three specific BWFBs with attractive features are obtained by adjusting the free parameter: the first has optimum coding gain and rational coefficients; the second which also has rational coefficients is very close to a quadrature mirror filter (QMF) bank; and the third which has binary coefficients can realize a multiplication-free discrete wavelet transform. In addition, four BWFBs are systematically verified to exhibit performance competitive to several state-of-the-art BWFBs for image compression, and yet require lower computational costs.

Image rejection calibration system

Abstract: Image rejection calibration includes initializing the calibration mode by applying to quadrature mixers, in place of the wanted RF input, an RF source in the frequency range of the wanted RF input, sensing the power output from the poly-phase filter, developing gain adjust and phase adjust correction values in response to the power output and adjusting in accordance with the correction values the gain of the quadrature signals from the quadrature mixers to the poly-phase filter and the phase of local oscillator quadrature signals from the local oscillator to the quadrature mixers to reduce the power output.

A polyphase filter based on CMOS inverters

Abstract: A new idea for generation of quadrature signals on chip is presented The topology is based on a passive RC polyphase filter, where the resistive parts are made active by using inverters The active filter combines quadrature generation, isolation, and gain without losing quadrature performance compared to a regular RC polyphase filter The filter technique is demonstrated in a 10 GHz front-end application where a broadband VCO, having a tuning range of 144 GHz, drives an active polyphase filter to generate quadrature LO signals According to simulations the quadrature phase error shows a typical tuned behavior and stays below 08° for the complete tuning range Since the signal amplitude is high throughout the filter the noise is low, below 160 dBc/Hz at 10 MHz offset The high amplitude also reduces the need for high gain tuned buffers, thereby enabling significant reductions in chip area

Characterization and Sampled-Data Design of Dual-Tree Filter Banks for Hilbert Transform Pairs of Wavelet Bases

Abstract: Characterization and design of dual-tree filter banks for forming Hilbert transform pairs of wavelet bases are studied. The characterization extends the existing results for quadrature mirror filter banks to general prefect reconstruction filter banks that satisfy only a mild technical assumption regarding the ratio of determinants of the two filter banks. We establish equivalent relationships of Hilbert transform pairs on scaling filters, wavelet filters, or scaling functions. The design of scaling filters of a dual filter bank is formulated as a sampled-data Hinfin optimization problem. The wavelet filters are then determined using the relationship on the determinants of the filter banks. We convert the sampled-data problem into an equivalent discrete-time Hinfin control problem, which can be solved by standard Hinfin control theory. An analytical solution to the sampled-data design problem is obtained for a special case. The sampled-data design approach usually gives infinite impulse response filter. In the case where the primal filter bank is of finite impulse response (FIR), we may truncate the impulse responses to get FIR approximations. They also lead to approximate Hilbert transform pairs. Design examples are presented

Method, medium, and system synthesizing a stereo signal

Abstract: A method, medium, and system generating a 3-dimensional (3D) stereo signal in a decoder by using a surround data stream. According to such a method, medium, and system, a head related transfer function (HRTF) is applied in a quadrature mirror filter (QMF) domain, thereby generating a 3D stereo signal by using a surround data stream.

Performance Analysis of Genetic Algorithm for the Design of Linear Phase Digital Filter Banks with CSD Coefficients

Abstract: In this paper genetic algorithm is utilized to design linear phase IIR quadrature mirror filter (QMF) banks with canonical signed digit coefficients (CSD). Subsequently, we present a through study on the performance of GA using different cross-over strategies. It is shown that 2-point cross-over generally works better than 1-point and uniform cross-over for IIR filter design. In the second part, the dependency of genetic algorithm to probability of mutation (Pm) and probability of cross-over (Pc) is analyzed. Experimental results show that with a fixed value for Pc, genetic algorithm performs better with the Pm of 4 to 6 percent, and with a fixed value for Pm, genetic algorithm yields better result with the Pc of around 95 percent.

Optimal Design of a Class of M-Channel Uniform Filter Bank using Genetic Algorithms

Abstract: This paper deals with the optimization of a class of M-channel uniform multirate filter bank. The specific case of a cosine modulated pseudo quadrature mirror filter (QMF) bank is considered for which all the filter bank channels are of equal width on the frequency scale. The design and optimization of the filter bank is based on the use of a single prototype filter. The first stage optimization process is based on perturbing the bandwidth and roll-off factor of a raised cosine prototype filter by using a genetic algorithm (GA) technique. Further improvements are subsequently derived by using a downhill Simplex optimization method.

Frequency domain linear prediction for QMF sub-bands and applications to audio coding

Abstract: This paper proposes an analysis technique for wide-band audio applications based on the predictability of the temporal evolution of Quadrature Mirror Filter (QMF) sub-band signals. The input audio signal is first decomposed into 64 sub-band signals using QMF decomposition. The temporal envelopes in critically sampled QMF sub-bands are approximated using frequency domain linear prediction applied over relatively long time segments (e.g. 1000 ms). Line Spectral Frequency parameters related to autoregressive models are computed and quantized in each frequency sub-band. The sub-band residuals are quantized in the frequency domain using a combination of split Vector Quantization (VQ) (for magnitudes) and uniform scalar quantization (for phases). In the decoder, the sub-band signal is reconstructed using the quantized residual and the corresponding quantized envelope. Finally, application of inverse QMF reconstructs the audio signal. Even with simple quantization techniques and without any sophisticated modules, the proposed audio coder provides encouraging results in objective quality tests. Also, the proposed coder is easily scalable across a wide range of bit-rates.

Low-Power, Low-Complexity Bit-Serial VLSI Architecture for 1D Discrete Wavelet Transform

Abstract: The discrete wavelet transform (DWT) is an upcoming compression technique that has been selected for MPEG-4 and JEPG 2000, because it has no blocking effects and it efficiently determines the frequency property of the temporary signals. In this paper, we propose a low-complexity, low-power bit-serial DWT architecture, employing a two-channel lattice-based quadrature mirror filter (QMF). The filter consists of four lattices (filter length = 8), and we determine the quantization bit for the coefficients using a fixed-length peak signal-to-noise ratio analysis and propose the architecture of the bit-serial multiplier with a fixed coefficient. The canonical signed digit encoding for the coefficients is applied to minimize the number of nonzero bits, thus reducing the hardware complexity. The proposed folded one-dimensional DWT architecture processes the other resolution levels during idle periods by decimations, and it provides efficient scheduling. The proposed architecture requires only flip-flops and full adders. This architecture has been designed and verified by the Verilog HDL and synthesized using the Synopsys Design Compiler with the DongbuAnam 0.18 μm Standard Cell Library. The maximum throughput is 393 Mbps at 450 MHz with a latency of 16 clocks, and the gate count is about 5K in equivalent two-input NAND gates. The dynamic power is 7.02 mW at 1.8 V. The data scheduling using a data dependency graph, and the performance, power, and required hardware cost are discussed.

Method, medium, and system decoding compressed multi-channel signals into 2-channel binaural signals

Abstract: A decoding method, medium, and system decoding an input compressed multi-channel signal, as a mono or stereo signal, into 2-channel binaural signals. Channel signals making up the multi-channel signals may be reconstructed from the input compressed signal in the quadrature mirror filter (QMF) domain, and head related transfer functions (HRTFs) for localizing channel signals in the frequency domain, represented as values in the time domain, may be transformed into spatial parameters in the QMF domain. Accordingly, channel signals may be localized in the QMF domain in directions corresponding to the channels, thereby decoding the input compressed signal as 2-channel binaural signals.

Scalable Wide-band Audio Codec based on Frequency Domain Linear Prediction

Abstract: This paper proposes a technique for wide-band audio applications based on the predictability of the temporal evolution of Quadrature Mirror Filter (QMF) sub-band signals. An input audio signal is first decomposed into 64 frequency sub-band signals using QMF decomposition. The temporal envelopes in critically sampled QMF sub-bands are approximated using frequency domain linear prediction applied over relatively long time segments (e.g. $1000$ ms). Line Spectral Frequency parameters related to autoregressive models are computed and quantized in each frequency sub-band. The sub-band residual signals are quantized in the frequency domain using a split Vector Quantization (VQ) technique. In the decoder, the sub-band signal is reconstructed using the quantized residual and the corresponding quantized envelope. Finally, application of inverse QMF reconstructs the audio signal. Even with simple quantization techniques and without any psychoacoustic model, the proposed audio coder provides encouraging results on objective quality tests.

Method and apparatus for decoding input signal which encoding multi-channel to mono or stereo signal to 2 channel binaural signal

Abstract: A method and an apparatus for decoding an input signal, that a multi-channel signal is compressed as a mono or stereo signal, into a binaural signal of 2 channels are provided to simplify an operating process by synthesizing and outputting the multi-channel signal as the binaural signal of the 2 channels without a process for recovering the input signal, that the multi-channel signal is compressed as the mono or stereo signal, is recovered as the multi-channel signal in a QMF(Quadrature Mirror Filter) domain. A channel level analyzing unit(204) calculates an FBCLs(Full Band Channel Levels) of respective channels composing a multi-channel from the CLD(Channel Level Difference) of the multi-channel. A two-channel synthesizing unit(208) performs the sound image localization of channel-classified data included in an input signal in directions corresponding to the channels on the basis of the calculated FBCLs of the respective channels.

Method, medium, and apparatus decoding an input signal including compressed multi-channel signals as a mono or stereo signal into 2-channel binaural signals

Abstract: A decoding method, medium, and device decoding an input signal, including compressed multi-channel signals as a mono or stereo signal, into 2-channel binaural signals. A full band channel level of each channel in the multi-channel system is calculated from channel level differences between the channels, and data of each channel included in the input signal is localized in directions corresponding to the channels based on the calculated full band channel levels of the channels. Accordingly, the input signal can be output as the 2-channel binaural signals by using simple operations without having to reconstruct multi-channel signals from the input signal in a quadrature mirror filter (QMF) domain.

3-band symmetric bi-orthogonal filter banks with compact support and their applications in signal processing

Abstract: Multi-band wavelets are newly emerging branch in wavelet family and could have better properties than dyadic wavelets in terms of symmetry, orthogonality, compact support and smoothness. The purpose of this paper is to present a new method for constructing the filter banks of 3-band symmetric bi-orthogonal wavelet using a scaling function of linear spline function. To construct such 3-band wavelet with desirable properties, a set of linear algebra equations can be listed according to the requirements of the bi-orthogonal multi-resolution analysis. And these equations are then solved to obtain the filter coefficients. The properties of the filters and the multi-resolution analysis (MRA) in signal processing are discussed. Experiments show that the 3-band filter banks could be potentially better in signal processing than dyadic wavelets.

Genetic algorithm and its application for the design of QMF banks with Canonical Signed Digit coefficients: A comparative study and new results

Abstract: In this paper genetic algorithm (GA) is used for designing quadrature mirror filter (QMF) banks with canonical signed digit (CSD) coefficients. The performance of genetic algorithm is investigated through a novel study on dependency of population size and number of generations to mean square error (MSE). Two optimization techniques are also compared for retaining the CSD constraints of coefficients. The first method offers a restoration operation for each loop of genetic algorithm but the second one can easily keep the CSD constraints through a new chromosome coding scheme. It is shown that for both IIR QMF bank and FIR QMF bank; the second technique yields better performance.

High Quality, Low Power QMF Bank Design for SBR, Parametric Coding, and MPEG Surround Decoders

Abstract: Due to the aliasing-free properties, the complex quadrature mirror filter (QMF) bank has been used in MPEG-4 audio standard on SBR, parametric, and surrounds coding. The high complexity overhead from the complex QMF bank and the complex data processing in the decoder leads to the development of low power decoder which adopts the real QMF bank as the basic building module to reduce the complexity. However the artifacts from the aliasing in the real QMF bank are the major concern. This paper studies the artifacts from the real QMF bank and proposes a novel QMF bank design to achieve both low complexity and high quality. Also, this paper applies the novel QMF bank to develop the high-quality and low-power SBR, parametric, and MPEG surround decoders and shows the merits in complexity and quality.

Quadrature Sampling of Wideband IF Signal

Abstract: On the basis of reviewing and contrasting the three processing methods for quadrature sampling of IF(Intermediate Frequency) signal: the LPF(Low Pass Filter) method,the bessel interpolation method and the polyphase filter methodThis paper deals with the applicability of the three methods with wideband signals and discusses the appropriate method fit for wideband signal and its implementation in projectThese are also simulated on computer and the results show that the LPF method is more suitable for the quadrature sampling of wideband IF signal

FPGA-Based Quadrature Mirror Filters for DSP Applications

Abstract: One of the most efficient multi-rate digital signal decomposition structures is the quadrature mirror filter (QMF) bank.It has been used successfully to implement subband coding applications such as speech coding and image compression. In this paper, we report on several efficient FPGA-based implementations of the basic quadrature mirror filter bank. The implementations are based on transforming the basic filter bank structure using ployphase decomposition and distributed arithmetic techniques. The two techniques reduce computational time dramatically by enabling maximum exploitation of the ample parallelism inherent in the filter bank. Performance results demonstrate the effectiveness of FPGA implementations of ployphase and distributed arithmetic quadrature mirror filter bank.

New Second Generation Current Conveyor-Based Current-Mode First Order All-Pass Filter And Quadrature Oscillator

Abstract: A new current-mode first order all-pass filter configuration is proposed and as it’s an application a current-mode quadrature oscillator circuit is also presented in this paper. The proposed circuit consists of a second generation current conveyor (CCII), two different type resistors and one type capacitors. Since the output of the filter exhibits low output impedance, the synthesized filter can be cascaded without additional buffers. Furthermore, it does not impose any component matching constraint in analog signal processing circuits. To explore the performance of the proposed all-pass filter section, a new currentmode quadrature oscillator is implemented. The performances of the proposed all-pass filter as well as the quadrature oscillator are also investigated by applying sensitivity analysis to the both proposed circuits. Finally the theoretical results are verified with PSPICE simulations using a CMOS realization of CCII.

Bidimensional PR QMF with FIR Filters

Abstract: Multidimensional perfect reconstruction (PR) quadrature mirror filter (QMF) banks with finite impulse response (FIR) filters induced from systems of biorthogonal multivariate scaling functions and wavelets are investigated. In particular, bivariate scaling functions and wavelets with dilation as an expansive integer matrix whose determinant is two in absolute value are considered. Demonstrative quincunxial examples are explicitly given and new FIR filters are constructed.

Method and apparatus for synthesizing stereo signal

Abstract: A method and an apparatus for generating a stereo signal are provided to reduce a system complexity for generating a 3D signal by using a surround data stream by applying an HRTF(Head Related Transfer Function) on a QMF domain. A demultiplexer(300) receives a surround data stream, which contains a down-mixed signal and a spatial parameter, from an encoder, and outputs a demultiplexed result. A domain converter(310) converts the down-mixed signal from the demultiplexer to a QMF(Quadrature Mirror Filter) domain signal. An up-mixer(320) receives the down-mixed signal from the domain converter, decodes the received signal, and up-mixes the decoded result to a multi-channel signal. A stereo signal generator(330) generates a 3D stereo signal on the QMF domain from the up-mixed multi-channel signal. A domain inverse-converter(340) inverse-converts the 3D stereo signal from a QMF domain to a time domain and outputs L-channel and R-channel signals.

Signal Adapted Filter BankDesign UsingMarkovParameters

Abstract: In this paper a method for designing principal component filter banks (PCFBs) based on a truncated sequence of estimated Markov parameters is described. This technique yields filters that are an excellent approximation to the ideal PCFB channel filters without nearly the computational complexity of traditional numerical optimization methods.

A New Family of PR Two Channel Filter Banks

Abstract: A new family of multidimensional dimensional (MD) perfect reconstruction (PR) two channel filter banks with finite impulse response (FIR) filters induced from systems of biorthogonal MD scaling functions and wavelets are introduced. One of the advantages of this construction is that the biorthogonal scaling functions and wavelets are easy to establish due to the interpolatory property of the scaling functions to start with. The other advantage is that all filters can be centrosymmetric or bi-linear phase. Examples of two dimensional (2D) bi-linear phase PR twochannel FIR filter banks will be demonstrated.

Optimaldesignofa classofm-channel uniform filterbank usinggeneticalgorithms

Abstract: Thispaperdeals withtheoptimization ofaclass ofMchannel uniform multirate filter bank.Thespecific case ofacosine modulated pseudo quadrature mirror filter (QMF)bankisconsidered forwhichallthefilter bank channels areofequal width onthefrequency scale. The design andoptimization ofthefilter bankisbased onthe useofa single prototype filter. Thefirst stage optimization process isbasedon perturbing the bandwidth androll-off factor ofaraised cosine prototype filter byusinga genetic algorithm (GA)technique. Further improvements aresubsequently derived byusing adownhill Simplex optimization method.

Perfect-Reconstruction Four-Tap Size-limited Filter Banks for Image Fusion Application

Abstract: Perfect reconstruction of the size-limited filter banks constructed with 4-tap wavelet filters and linear boundary extension is studied. The analytic results on the perfect reconstruction of these size-limited filter banks are presented. The analytic results show that, except some extreme cases, the size-limited filter banks constructed with 4-tap orthogonal wavelets and linear boundary extension have perfect reconstruction structure without considering the real values of the wavelet filters. The proposed size-limited bank was applied to image fusion and the results are impressive.

Low-complexity bit-serial vlsi architecture for 1d discrete wave let transform*

Abstract: The discrete wavelet transform (DWT) is an upcoming compression technique that has been selected for MPEG-4 and JEPG 2000, because it has no blocking effects and it efficiently determines the frequency property of the temporary signals. In this paper, we propose a low-complexity, low-power bit-serial DWT architecture, employing a two- channel lattice-based quadrature mirror filter (QMF). The filter consists of four lattices (filter length = 8), and we determine the quantization bit for the coefficients using a fixed- length peak signal-to-noise ratio analysis and propose the architecture of the bit-serial multiplier with a fixed coefficient. The canonical signed digit encoding for the coefficients is applied to minimize the number of nonzero bits, thus reducing the hardware complexity. The proposed folded one-dimensional DWT architecture processes the other resolution lev- els during idle periods by decimations, and it provides efficient scheduling. The proposed architecture requires only flip-flops and full adders. This architecture has been designed and verified by the Verilog HDL and synthesized using the Synopsys Design Compiler with the DongbuAnam 0.18µm Standard Cell Library. The maximum throughput is 393 Mbps at 450 MHz with a latency of 16 clocks, and the gate count is about 5K in equivalent two-input NAND gates. The dynamic power is 7.02 mW at 1.8 V. The data scheduling using a data dependency graph, and the performance, power, and required hardware cost are discussed. Ke yw ords: DWT, bit-serial, folding, pipeline, QMF lattice.