Search

This paper presents a computational method for the optimal design of all-pass variable fractional-delay (VFD) filters aiming to minimize the squared error of the fractional group delay subject to a low level of squared error in the phase response. The constrained optimization problem thus formulated is converted to an unconstrained least-squares (LS) optimization problem which is highly nonlinear. However, it can be approximated by a linear LS optimization problem which in turn simply requires the solution of a linear system. The proposed method can efficiently minimize the total error energy of the fractional group delay while maintaining constraints on the level of the error energy of the phase response. To make the error distribution as flat as possible, a weighted LS (WLS) design method is also developed. An error weighting function is obtained according to the solution of the previous constrained LS design. The maximum peak error is then further reduced by an iterative updating of the error weighting function. Numerical examples are included in order to compare the performance of the filters designed using the proposed methods with those designed by several existing methods. Index Terms--All-pass filter, constrained least-squares (LS) design, variable fractional-delay (VFD) filters, weighted LS (WLS) design.

A unified approach is presented to the optimal design of sharp linear-phase finite-impulse-response (FIR) digital filters synthesized using the multistage frequency-response masking technique. The weighted least-squares (WLS) problem is highly nonlinear with respect to the coefficients of the filter and hence it is solved iteratively by alternately solving these linear leas squares (LS) problems with appropriately updated weighting functions.

An efficient computational method for the frequency-domain design of digital finite impulse response (FIR) filters with arbitrary magnitude and group delay responses is presented. The method can deal with both the equiripple design problem and the peak constrained lease squares (PCLS) design problem and the method can also be applied to the equiripple passbands and PCLS stopbands design problem as a special case of the PCLS design.

In this paper, we firstly establish the existence theorem of the global weak solutions of the Cauchy problem for the shallow water wave model of moderate amplitude, then following the idea in Xin and Zhang’s work (see Xin and Zhang, 2002), we prove the uniqueness of global weak solutions using the localization analysis in the transport equation theory. Finally, several travelling wave solutions are derived. [ABSTRACT FROM AUTHOR]

Several computational models have been developed in the literature to describe the dynamics of the cell-cycle for the mammalian cell, in particular for cancer cells, using both traditional and new techniques and yielding some positive results. In this paper, we discuss how to optimise model parameters for these types of models and how this can serve to enhance numerical results. We pose the model parameter selection problem as an optimal parameter selection problem on both a normal cell and cancer cell cycle of growth. Various possible objectives are discussed and we illustrate the process with some numerical results. [ABSTRACT FROM AUTHOR]

In this paper, we address a non-preemptive task scheduling problem with an objective function of minimizing the makespan. We consider online scheduling with release date on identical parallel processing system with centralized and no splitting structure. Multi-steps heuristic algorithms are proposed to solve this non-deterministic scheduling problem. A computational experiment is conducted to examine the effectiveness of the proposed multi-steps method in different size problem. The computational results show that all the proposed heuristics obtain good results with the gap between optimal solutions are less than 10% even for a large date set. The experiment is performed using Microsoft Visual C++ programming software in windows environment. [ABSTRACT FROM AUTHOR]

SUMMARY In this paper, we develop a macro traffic flow model with consideration of varying road conditions. Our analytical and numerical results illustrate that good road condition can enhance the speed and flow of uniform traffic flow whereas bad road condition will reduce the speed and flow. The numerical results also show that good road condition can smooth shock wave and improve the stability of traffic flow whereas bad road condition will lead to steeper shock wave and reduce the stability of traffic flow. Our results are also qualitatively accordant with empirical results, which implies that the proposed model can qualitatively describe the effects of road conditions on traffic flow. These results can guide traffic engineers to improve the road quality in traffic engineering. Copyright © 2012 John Wiley & Sons, Ltd. [ABSTRACT FROM AUTHOR]