Method for reducing computational time for calculating a noise-filtered average approximation of throttle position

Abstract

A method for reducing computational time for calculating a noise-filtered average approximation of a throttle position in an automotive environment. During controller initialization, an initial average value for N samples is established by conventional averaging. The sum obtained is retained for future use as a previously-retained sum. When an updated average value is required, the oldest sample value and the last output calculation are both subtracted from the previously-retained sum, and the newest sample value is added twice to the previously-retained sum, as well as being stored in sequence in the buffer. The new sum is then divided by the number of sample values to obtain a new noise filtered average approximation of throttle position, which again is retained for use in the next update. The new output value obtained by the throttle position sensor is weighed more heavily to decrease the deviation from the average approximation of throttle position.