Multirate Approach Research Articles

Low Power Digital Frequency Conversion Architectures

State-of-the-art data communication systems make extensive use of digital hardware. Besides baseband modulation functions, also the frequency tuning functions are now being shifted from analog to digital implementation. Integration, cost and ease of programming are the primary motivations for doing this. This paper presents an alternative to the traditional digital frequency conversion architectures. The proposed architecture achieves low power as well as high speed operation, and achieves this dual goal by reducing programmability. A multi-rate filtering approach is used, which is applicable for both upconversion and downconversion of quadrature modulated data.

Multirate Predictive Control with Nonlinear Optimization Application to a Thermal Process for Batch Dyeing

In this paper, we present a new approach to multirate predictive control with a multiple reference model in order to control commercial processes with nonlinear models, such as batch dyeing thermal processes. The time delay problem is resolved successfully by using multirate predictive control, and we introduce nonlinear optimization based on an augmented Lagrangien objective function to minimize the multirate predictive control quadratic criterion in the case of the nonlinear plant model. Moreover, the multirate approach improves predictive control robustness. The results of our study are used to simulate nonlinear thermal plant control with very good results. For this reason, our future work will use this method to control a batch dyeing process.

Efficient multirate adaptive beamforming technique

An efficient beamforming technique based on the recently proposed quadrature mirror filter based multirate sub-band adaptive beamformer (QMF-MSAB), is presented. The improved technique uses only one adaptive processor for all sub-bands, resulting in a considerable reduction in hardware requirements. In addition, the new structure maintains the overall superiority of the multirate approach over full-band beamforming.

On the design of a broad class of 2-D recursive digital filters with fan, diamond and elliptically-symmetric responses

This paper deals with the design of a broad class of 2-D recursive digital filters including general fan, diamond and elliptically-symmetric filters. At first, the explicit design of quadrant and /spl plusmn/45/spl deg/ fan filters is reviewed. A full-plane diamond filter (both axes are nearly of length 2 /spl pi/) can easily be obtained through an appropriate transformation of a /spl plusmn/45/spl deg/ fan filter. Using this diamond filter as a prototype, a very simple spectral transformation is then proposed to design full-plane circularly-symmetric filters (radius /spl ap//spl pi/). These filters share the same lattice wave digital structure and employ only first or second order all-pass sections which can be realized by two-port adaptors. They provide very desirable properties such as low sensitivity, high dynamic range and good stability even under finite arithmetic conditions. Using the above mentioned filters as prototypes, a multirate approach is then introduced to deal with the design of fan filters with arbitrary angle and orientation, diamond and elliptically-symmetric filters with arbitrary length of axes. Finally, some design examples are given. >

A multirate processing approach for microwave imaging

In this paper, a multirate processing approach for microwave imaging is presented. This approach has the advantages of largely compressing the raw spectral data for imaging, greatly reducing the storage requirement and enhancing the processing efficiency. To demonstrate its applicability, the proposed approach is tested on both simulated and experimental data.

Accounting for Different Time Scales in Activated Sludge Process Control

A multirate approach is presented for the integration and discrete linearization of the activated sludge dynamics, that can be readily incorporated in optimal control and uncertainty propagation algorithms. This procedure fully accounts for the existence of multiple time-scale clusters in the activated sludge process and lends itself to a computationally efficient implementation. The method is verified in a computational case study involving a reduced-order version of the IAWPRC model.