Browse Prior Art Database

System Identification Using Nonsynchronous Sampling

IP.com Disclosure Number: IPCOM000061672D
Original Publication Date: 1986-Sep-01
Included in the Prior Art Database: 2005-Mar-09
Document File: 2 page(s) / 40K

Publishing Venue

IBM

Related People

Ottesen, HH: AUTHOR [+2]

Abstract

Conventional System Identification requires a fixed (synchronous) sampling period. System Identification can be extended to dynamic systems where only nonsynchronous sampling, i.e., time varying sampling period, of input and output variables is available or practical from a cost standpoint. The solution to this time variable sampling period is to approximate the original discrete input and output waveform by using a graphical interpolation between the nonsynchronous sampling points and then sample (in software) the new reconstructed waveform at a constant sampling rate (constant sampling period). A nonsynchronous or time-varying sampling period is typically encountered in electromechanical hardware, where velocity is measured indirectly from the time required to move between two fixed markers of known separation.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 55% of the total text.

Page 1 of 2

System Identification Using Nonsynchronous Sampling

Conventional System Identification requires a fixed (synchronous) sampling period. System Identification can be extended to dynamic systems where only nonsynchronous sampling, i.e., time varying sampling period, of input and output variables is available or practical from a cost standpoint. The solution to this time variable sampling period is to approximate the original discrete input and output waveform by using a graphical interpolation between the nonsynchronous sampling points and then sample (in software) the new reconstructed waveform at a constant sampling rate (constant sampling period). A nonsynchronous or time-varying sampling period is typically encountered in electromechanical hardware, where velocity is measured indirectly from the time required to move between two fixed markers of known separation. The motion could be linear as well as rotary. For example, consider the hard disk file shown in Fig. 1. The circuit depicted in the figure shows how the spindle velocity W(t) is obtained. At the sensing of index 5, an index pulse is generated in the logic 3 which resets the counter 2 and pulses from an oscillator 1 are counted until the next index pulse occurs. Let N be the number of oscillator pulses counted between the occurrence of two index pulses and f be the oscillator frequency. The time T(t) per revolution is then N/f, and the radian velocity is given by:

(Image Omitted)

If the rotational veloci...