By Topic

# China., 1991 International Conference on Circuits and Systems

## Filter Results

Displaying Results 1 - 25 of 274
• ### China 1991 International Conference on Circuits and Systems. Conference Proceedings (Cat. No.91TH0387-1)

Publication Year: 1991
| PDF (26 KB)
• ### A novel approach for degraded speech recognition

Publication Year: 1991, Page(s):16 - 17 vol.1
| | PDF (88 KB)

Summary form only given. In speech recognition, training and recognition processes can be conducted in different environments. For example, the word pattern may be extracted from clean speech, while the recognition may take place in background noise. Further more, since the vocabulary size is limited, it is possible, in the training phase, to extract the spectral information for all the words to b... View full abstract»

• ### Analysis of spherical-wraparound microstrip antennas by spectral domain method

Publication Year: 1991, Page(s):392 - 394 vol.1
| | PDF (261 KB)

The far field radiation patterns of a wraparound microstrip antenna mounted on a sphere are studied theoretically. The radiator is replaced by an assumed surface current distribution. The effects of the dielectric layer and the metallic sphere are rigorously taken into account by a spectral domain Green's function formulation. Some practical results with different dielectric substrate thickness an... View full abstract»

• ### Neural networks and associative memory

Publication Year: 1991
| | PDF (23 KB)

Summary form only given, as follows. This author presents several design techniques that can be used for continuous-time and discrete-time neural networks in order to realize associative memories. The applicability of these techniques is demonstrated by means of specific examples which illustrate strengths and weaknesses.<> View full abstract»

• ### Signal state and delay calculation in switch-level networks for mixed-level simulation

Publication Year: 1991, Page(s):917 - 920 vol.2
| | PDF (272 KB)

In this paper a robust switch-level simulation algorithm is proposed. It uses a MOSSIM-based algorithm to calculate the steady state. Instead of the unit delay used in the MOSSIM algorithm, it uses nodal delay potential equations, combined with the MOS transistor's equivalent dynamic resistance, as well as the effect of slope of the input stimuli, to calculate the nodal delay. The primary computat... View full abstract»

• ### Bounding the design space in hardware allocation

Publication Year: 1991, Page(s):913 - 916 vol.2
Cited by:  Papers (1)
| | PDF (312 KB)

The authors describe the structure of the FIDIAS system, a high level synthesis tool. They also describe in more detail HARAL, the FIDIAS module accomplishing the hardware allocation subtask. The three main features of HARAL are shown, while the data path model and the algorithm used for exploring the resulting design space are given, as well as the criteria employed for design cost estimation. Th... View full abstract»

• ### Topological cell compaction via transister rotation

Publication Year: 1991, Page(s):909 - 912 vol.2
Cited by:  Patents (1)
| | PDF (304 KB)

The authors propose a new algorithm for layout compaction by modifying the topology of a given layout. Different from most compaction algorithms which move the components of a layout, this algorithm compacts a layout by changing the orientations of transistors. A set of operations including moving, adding, deleting, shrinking, extending, etc., can work on the wires to rebuild and compact the layou... View full abstract»

• ### An improved nonlinear macromodel of OTA

Publication Year: 1991, Page(s):906 - 908 vol.2
| | PDF (168 KB)

A new nonlinear macromodel for operational transconductance amplifier (OTA) is presented. It is developed by the build-up' technique. Typical parameters, such as differential and common mode input impedance, CMRR, slew rate, and the bias current effect, are taken into account. Model parameters can be calculated from the data sheets supplied by the manufacturers or chosen arbitrarily by the user. ... View full abstract»

• ### A concurrent error-detectable module design for FFT processing

Publication Year: 1991, Page(s):839 - 842 vol.2
| | PDF (348 KB)

Most data in digital signal processing belong to the complex number system. Some characteristics of complex number computations can be utilized to reduce hardware overhead for concurrent error detection (CED) scheme. This paper presents a novel design of CED technique based on realization of direct complex computation approach. It is a time-redundant method and the recomputation step is performed ... View full abstract»

• ### Channel routing algorithm by free layer assignment

Publication Year: 1991, Page(s):874 - 877 vol.2
| | PDF (312 KB)

A new channel routing model, which allows horizontal and vertical wire segments to be placed on both layers, and two wires on a different layer to be overlapped, is presented. Based on this model, the basic constraints as well as the rules of layer assignment and routing order in two-layer channel routing are obtained. Then the property of optimal overlapping in a zone is analysed. Experimental re... View full abstract»

• ### A new input method for CAD-automatic read-in technique of paper drawings

Publication Year: 1991, Page(s):431 - 433 vol.1
Cited by:  Papers (2)  |  Patents (2)
| | PDF (196 KB)

The authors present a new technique which can be used to read electronic circuits automatically. Using two prototype systems, they discuss recognition methods for the electronic element symbols. They have applied the syntactic structural approach to recognize the symbols and English characters. The methods described can also be applied to read other engineering drawings View full abstract»

• ### The realization of op amp nonlinear macromodel on SPICE

Publication Year: 1991, Page(s):902 - 905 vol.2
| | PDF (188 KB)

In this paper a comparatively ideal curve fitting technique for a nonlinear function is used to approach the large signal saturation nonlinear characteristics of an op amp input stage, so that the time domain response of circuits containing nonlinear op amp macromodels can be simulated precisely on SPICE. A relevant curve fitting program is realized and two circuits are analysed, giving satisfacto... View full abstract»

• ### Design of distributed hardware preprocessor for realtime processing of telemetry data

Publication Year: 1991, Page(s):835 - 838 vol.2
| | PDF (240 KB)

Even though minicomputers and microcomputers are larger and faster than ever before, they have not kept pace with the demands of telemetry users. Hardware preprocessor of telemetry data has been used in distributed telemetry computer system for processing all data at extremely high rates. According to the requirements and characteristics of telemetry data processing, a distributed multiple data st... View full abstract»

• ### A via minimization channel router for three-layer channel with irregular boundaries

Publication Year: 1991, Page(s):870 - 873 vol.2
| | PDF (268 KB)

In LSI/VLSI chip layout design, channel routing is one of the key steps. Generally, boundaries of the channel are two parallel lines. But sometimes, because of the different sized cells or macro cells, there may be some indentations on boundaries of the channel. In this paper, a channel router for three-layer channel with irregular boundaries is presented. Both the number of vias and the number of... View full abstract»

• ### Address generation of prime factor algorithm in a multiprocessor system

Publication Year: 1991, Page(s):133 - 136 vol.1
| | PDF (396 KB)

The prime factor algorithm is an efficient method for computing DFT and much research has been done in the fast address generation associated with the algorithm. A simple address generation scheme is now proposed which is most suitable for implementation in a multiprocessor system. A tree network is simulated by a transputer network for this study View full abstract»

• ### Design of FIR narrow-band frequency selective filters using sliding DFT array processors

Publication Year: 1991, Page(s):617 - 620 vol.2
Cited by:  Papers (1)
| | PDF (272 KB)

Introduces a kind of DFT array processors. Its weighting factors of processing element's (PE's) are time-invariable. A FIR narrow-band frequency selective filter using DFT array processors has been designed. DFT array processors can easily calculate the frequency spectrum of corresponding sequence of N length by sliding technique. Frequency sampling designs are used for narrow-band frequency selec... View full abstract»

• ### Statistical design of class E tuned power amplifier

Publication Year: 1991, Page(s):746 - 749 vol.2
| | PDF (320 KB)

Because of the existence of unavoidable tolerances in manufacturing, high yield is desirable in mass production. In this paper, the authors focus on the statistical design of class E tuned power amplifier to improve the yield. The results have shown that yield can be improved to a satisfactory extent by statistical design centering. Detailed design data tables are given in the paper View full abstract»

• ### The effect of device configuration on GaAs MESFET negative resistance behaviour

Publication Year: 1991, Page(s):427 - 430 vol.1
Cited by:  Papers (3)
| | PDF (248 KB)

A new analysis of the effect of device configuration on negative resistance behaviour of a GaAs MESFET with series feedback has been undertaken. The analysis is based on a simplified equivalent circuit representation of the active device, and gives explicit results for the frequency range over which negative resistance behaviour can be expected to occur. It is shown that even with S12=0... View full abstract»

• ### 8 Kb/s trellis speech coding with and without prediction

Publication Year: 1991, Page(s):33 - 36 vol.1
| | PDF (224 KB)

The basic principle of trellis source coding is introduced and the design method for trellis source coding system is described, with an emphasis on the design of the trellis codebook using a training sequence. Speech is used as the source, and the computer simulation results show that trellis speech coding can reach fairly good performance at the bit rate of 8 kb/s View full abstract»

• ### m3DI: a dynamic configurable multilayer router in the automatic cell layout generation environment

Publication Year: 1991, Page(s):898 - 901 vol.2
Cited by:  Papers (2)
| | PDF (436 KB)

The authors have implemented the 3-2 route scheme, successfully constructing a multilayer router, m3DI. The system is intended to solve the routing problems encountered in the automatic generation of custom cell layouts targeted at GaAs and CMOS technology which require three or more metal layers. The router is a dynamically configurable router which not only can handle the routing at the leaf cel... View full abstract»

• ### An ASIC design for adaptive beamforming

Publication Year: 1991, Page(s):832 - 834 vol.2
Cited by:  Papers (2)
| | PDF (244 KB)

Adaptive beamforming is based on the RLS (recursive least squares) algorithm realized by the RLS systolic array (RLSSA) proposed by McWhirter in 1983. The authors use CORDIC (CO-ordinate Rotation Digital Computer) algorithm to implement RLSSA. Only shifting and adding operations are employed to realize all the primitive array's complex procedure. The authors have specially designed some proposed l... View full abstract»

• ### Analysis of stereophonic image vibration in AM stereophonic broadcasting

Publication Year: 1991, Page(s):385 - 387 vol.1
| | PDF (236 KB)

A computer analysis shows that under the influence of interfering signals, the stereophonic vibration angle is up to ±25° in the Motorola system. The stereophonic images show left' and `right' swing after the interference. The vibrating angle is smaller in the Kahn system. Generally, it is below ±5°. However, there are vibrations in both L and R signals so that the stereopho... View full abstract»

• ### The hidden nodes of BP neutral nets and the activation functions

Publication Year: 1991, Page(s):77 - 79
| | PDF (196 KB)

Proposes a new approach for the estimation of the number of hidden nodes in a BP neural net, and a semi-experimental expression is given. Along with the comparison of the recent results concerned, the authors discuss the effect of input patterns. A new activation function f(x)=exp(-x2) is also proposed, which may decrease the nodes of the net for XOR problem View full abstract»

• ### Analysis of switched-capacitor network with nonideal op amps using analog circuit simulation program

Publication Year: 1991, Page(s):248 - 251 vol.1
| | PDF (240 KB)

Two new methods for analyzing the effects of the nonideal op amp on the frequency-domain response of switched-capacitor networks using an analog circuit simulation program are presented. The main idea is based on dealing with the problem in the s-domain or in the transformed domain (p-domain) rather than in the z-domain. By incorporating the indefinite admittance matrix technique, these methods pr... View full abstract»

• ### Multirate switched-capacitor N-path filtering systems using bandpass path filters

Publication Year: 1991, Page(s):230 - 233 vol.1
Cited by:  Papers (1)
| | PDF (464 KB)

The author investigates new types of multirate switched-capacitor filtering systems whose operating principles combine the ideas of N-path filters using bandpass path filters and single-path frequency-translated systems, and which are capable of realising high complexity bandpass responses which are fully immune to in-band clock feedthrough and mirror frequency components. Alternative system archi... View full abstract»