Chau-Yun Hsu

A study of feature-mapped approach to the multiple travelling salesmen problem

A neural network approach based on the self-organized feature map model is proposed to solve the multiple traveling salesmen problem (MTSP). Instead of studying the MTSP theoretically, the authors try to solve the MTSP. The process of evolving populations of cells, featuring duplication and selection, iteratively organizes a quasi-optimal solution for each TSP. The example of Wacholders demonstration is examined.<<ETX>>

Comparative performance of fast cosine transform with fixed-point roundoff error analysis

Suitable scaling schemes are chosen for the Lees and the Hous (1984) fast DCT algorithms, and the relative fixed-point roundoff error analyses are carried out, respectively. The average output signal-to-noise ratio are then calculated, and it is shown that in DCT and for N>16 stage-by-stage scaling of Hous algorithm has the best performance, whereas in inverse DCT, the global scaling of either algorithms has the best performance. >

Fixed-point round-off error analysis for the discrete cosine transform

Abstract In this paper, a fixed-point round-off error analysis of the discrete cosine transform (DCT) has been carried out. A relationship between the range of the twiddle factor and the dimension of the DCT is first derived, whence a suitable scaling model is chosen for the DCT algorithm and the average output signal-to-noise ratio is calculated.

Wave-induced flow in meridians demonstrated using photoluminescent bioceramic material on acupuncture points

The mechanisms of acupuncture remain poorly understood, but it is generally assumed that measuring the electrical conductivity at various meridians provides data representing various meridian energies. In the past, noninvasive methods have been used to stimulate the acupuncture points at meridians, such as heat, electricity, magnets, and lasers. Photoluminescent bioceramic (PLB) material has been proven to weaken hydrogen bonds and alter the characteristics of liquid water. In this study, we applied the noninvasive PLB technique to acupuncture point irradiation, attempting to detect its effects by using electrical conductivity measurements. We reviewed relevant literature, searching for information on meridians including their wave-induced flow characteristics.

Systematic reducing the PAPR of OFDM by cyclic coding

This paper presents a new systematic method that combines parity check and cyclic coding technology to reduce the peak to mean envelope power ratio (PMEPR) of multicarrier transmission system. According our study, this new systematic method is much efficient than detail search method to find a set of good code word. The first procedure of the proposed method is to use parity check technology to reduce the peak power by more than half and generate a set of M-ary code words. The second procedure is to create a polynomial generator matrix from Galois fields, and than using a linear code set, multiplied by the matrix, to generate an m-ary cyclic codes set. Finally, the m-ary set of cyclic code is mapped into an M-ary set of cyclic codes to produce final PEP distribution. The results of this mapping reveal that PMEPR is reduced. We simulate this new method by the specific example of an eight-carriers multicarrier transmission system.

Fast discrete extrapolation via the fast Hartley transform

A fast algorithm for A. Papouliss (1975) and R.W. Gerchbergs (1974) iterative extrapolation based on the fast Hartley transform (FHT) is presented. The low-pass filtering in the iterative procedure can be implemented by the FHT directly instead of by the fast Fourier transform (FFT) and the inverse FFT. M.S. Sabri and W. Steenarts (1978) example demonstrates that the FHT approach is simple to use. >

Low complexity radix-4 butterfly design for the soft-decision Viterbi decoder

This study presents a new radix-4 butterfly design for Viterbi decoders. The branch symmetry of the proposed radix-4 butterfly is exploited to design a low-complexity radix-4 butterfly module to simplify the implementation of the soft-decision Viterbi decoder. By exploiting the branch symmetry, only a half of branch metrics need to be computed, while other metrics can be derived from the computed branches. Therefore, the branch metric computation of the radix-4 butterfly can be reduced by a factor of 2. Considering the convolutional code in the DAB system as an example, experimental results indicate that the proposed radix-4 butterfly design can reduce the number of FPGA slices of the radix-4 butterfly module by 24% over the conventional design.

The one‐time learning hierarchical CMAC and the memory limited CA‐CMAC for image data compression

Abstract Two methods to compress transmitted image data are proposed in this paper. The first method is the one‐time learning hierarchical CMAC method and the second is the memory limited CA‐CMAC method for image data compression and reconstruction. The one‐time learning hierarchical CMAC method is used when a coarse image needs to be sent to the receiver initially and then the image quality is gradually improved at the request of the receiver. But, when the transmitting channel data is limited, the memory limited CA‐CMAC method can be used to decrease the bit rate per pixel. Both proposed methods, unlike conventional compression methods, use no filtering technique in either compression or reconstruction. CMAC networks use a few hypercubes to learn the characteristics of the original image, so image data can be compressed without suffering from blocking effect or boundary effect. Onetime learning is good enough for compressing image data, and it has a high SNR after reconstruction.

Discrete interpolation using the discrete cosine transform with the mapping of the boundary conditions

Discrete interpolation between successive samples of a sequence of real numbers by using the sinusoidal transforms have been an interesting topic in digital signal processing. The accuracy of discrete interpolation by using the discrete cosine transform of type I can be greatly improved with the aid of the mapping of the input data sequence to the one that satisfies the boundary conditions. In this article, the critical boundary conditions is set, and the corresponding mapping for the input data is also developed. The simulation of two typical examples shows that the accuracy of this proposed algorithm is more accurate than any other existing increased accuracy algorithms by using the sinusoidal transforms. >

An improved algorithm for Kohonen's self-organizing feature maps

A modified algorithm is presented for the learning by self-organizing topology-preserving maps to improve the piecewise-correct problem that arose frequently with the original self-organizing maps. The problem is generally caused by two dominant factors existing in the learning procedure of the original algorithm. One is the initial-sequence-order problem. The present algorithm efficiently reduces the influence of these two factors and successfully guides the network to form a topologically correct map. The proposed algorithm adopts a dynamic network that allows cells to be inserted and deleted, and it adds the Coulomb effect to the learning factor. Simulation results indicate that the modified algorithm performs well in learning the mapping of a two-dimensional input vector distribution using a one-dimensional network.<<ETX>>