Noise burst source for transfer function testing

Claim 1

A method of measuring transfer functions of a system under test, the method comprising the steps of: a. generating and applying a transient random stimulus of period T seconds in length to the system under test, this step including the steps of: generating a random stimulus; and applying the random stimulus to the system under test for a period of T2 seconds during the initial portion of the T second period, T2 being a period that is substantially shorter than the T second period, and T being a period that is sufficiently long to allow the transient response of the system under test to the T2 second random stimulus to decay substantially to zero during the T second period; b. recording the transient random stimulus generated in step (a) during its application to the system under test; c. measuring the spectrum of the transient random stimulus during its T second period; d. applying the recorded transient random stimulus to the system under test after the T second period of the transient random stimulus generated in step (a) has been completed; e. measuring the spectrum of the response of the system under test during the T second period of the transient random stimulus applied thereto in step (d); f. computing a first estimate of the transfer function of the system under test from the two measured spectra; g. repeating steps (a) through (f) to obtain an additional estimate of the transfer function of the system under test using a transient random stimulus that is uncorrelated with previous stimuli; h. averaging together each additional estimate and the average of the previously obtained estimates of the transfer function of the system under test; and i. repeating steps (g) and (h) a sufficient number of times to obtain an average estimate of the transfer function of the system under test within a selected accuracy.

Claim 2

A method of measuring transfer functions of a system under test, the method comprising the steps of: a. generating and applying a transient random stimulus of period T seconds in length to the system under test, this step including the steps of: generating a random stimulus; and applying the random stimulus to the system under test for a period of T2 seconds during the initial portion of the T second period, T2 being a period that is substantially shorter than the T second period, and T being a period that is sufficiently long to allow the transient response of the system under test to the T2 second random stimulus to decay substantially to zero during the T second period; b. recording the transient random stimulus generated in step (a) during its application to the system under test; c. measuring the spectrum of the transient random stimulus during its T second period; d. applying the recorded transient random stimulus to the system under test after the T second period of the transient random stimulus generated in step (a) has been completed; e. measuring the spectrum of the response of the system under test during the T second period of the transient random stimulus applied thereto in step (d); f. computing a first estimate of the transfer function of the system under test from the two measured spectra; g. repeating steps (a) through (f) to obtain an additional estimate of the transfer function of the system under test using a transient random stimulus that is uncorrelated with previous stimuli; h. averaging together each additional estimate and the average of the previously obtained estimates of the transfer function of the system under test; and i. repeating steps (g) and (h) a sufficient number of times to obtain an average estimate of the transfer function of the system under test within a selected accuracy.

Claim 4

A method of measuring transfer functions of a system under test, the method comprising the steps of: a. generating and applying a transient random stimulus of period T seconds in length to the system under test, this step including the steps of: generating a random stimulus; and applying the random stimulus to the system under test for a period of T2 seconds during the initial portion of the T second period, T2 being a period that is substantially shorter than the T second period, and T being a period that is sufficiently long to allow the transient response of the system under test to the T2 second random stimulus to decay substantially to zero during the T second period; b. measuring the spectra of the transient random stimulus and the transient response of the system under test simultaneously during the T second period of the transient random stimulus; c. computing a first estimate of the transfer function of the system under test from the two measured specta; d. repeating steps (a) through (c) to obtain an additional estimate of the transfer function of the system under test using a transient random stimulus that is uncorrelated with previous stimuli; e. averaging together each additional estimate and the average of the previously obtained estimates of the transfer function of the system under test; and f. repeating steps (d) and (e) a sufficient number of times to obtain an average estimate of the transfer function of the system under test within selected accuracy.

Claim 5

A method of measuring transfer functions of a system under test, the method comprising the steps of: a. generating and applying a transient random stimulus of period T seconds in length to the system under test, this step including the steps of: generating a random stimulus; and applying the random stimulus to the system under test for a period of T2 seconds during the initial portion of the T second period, T2 being a period that is substantially shorter than the T second period, and T being a period that is sufficiently long to allow the transient response of the system under test to the T2 second random stimulus to decay substantially to zero during the T second period; b. measuring the spectra of the transient random stimulus and the transient response of the system under test simultaneously during the T second period of the transient random stimulus; c. computing a first estimate of the transfer function of the system under test from the two measured specta; d. repeating steps (a) through (c) to obtain an additional estimate of the transfer function of the system under test using a transient random stimulus that is uncorrelated with previous stimuli; e. averaging together each additional estimate and the average of the previously obtained estimates of the transfer function of the system under test; and f. repeating steps (d) and (e) a sufficient number of times to obtain an average estimate of the transfer function of the system under test within selected accuracy.

Claim 7

An apparatus for generating a transient random stimulus for application to a system under test comprising: computational means for generating a series of random sample points to form a random burst stimulus of T2 seconds in length from a kernal during the initial portion of a measurement period T seconds in length, T2 being a period that is substantially shorter than the measurement period T; and an output means responsive to the random burst stimulus for applying that stimulus to the system under test; said computational means also being connected to selectively restore a selected kernal and to initiate the generation of an identical random burst stimulus a single time, after which a next kernal is selected to generate a random burst stimulus that is uncorrelated with all other stimuli after the last preceding T second measurement period is complete, and said measurement period T being sufficiently long to allow the transient response of the system under test to the transient random stimulus to decay substantially zero during the T second measurement period.

Claim 8

An apparatus for generating a transient random stimulus for application to a system under test comprising: computational means for generating a series of random sample points to form a random burst stimulus of T2 seconds in length from a kernal during the initial portion of a measurement period T seconds in length, T2 being a period that is substantially shorter than the measurement period T; and an output means responsive to the random burst stimulus for applying that stimulus to the system under test; said computational means also being connected to selectively restore a selected kernal and to initiate the generation of an identical random burst stimulus a single time, after which a next kernal is selected to generate a random burst stimulus that is uncorrelated with all other stimuli after the last preceding T second measurement period is complete, and said measurement period T being sufficiently long to allow the transient response of the system under test to the transient random stimulus to decay substantially zero during the T second measurement period.