Technical Report Number
This thesis examines the eﬀects of changing the numerical representation of audio signals in digital hearing aids to minimize power consumption. Within the hearing aid design a majority of the power used is consumed in the many ﬁnite impulse response ﬁlters. The main processing involved in these ﬁlters is a multiply-accumulate function. We examine the power consumption of 12 diﬀerent multiply-accumulate units that use the following numerical representations: a 16-bit linear representation, a 9-bit logarithmic representation, and 10 diﬀerent ﬂoating-point rep-representations ranging from 9 to 13 bits. A selection of the multiply-accumulators are simulated using a continuous-circuit simulator. The power estimates from this are compared with signal transition counts from a discrete event simulator to quantify the relationship between transition counts and power consumption. This relationship is then used to examine other numerical representations.
Hemmeter, Eric E., "Reducing Power Consumption Using Customized Numerical Representations in Digital Hearing Aids" Report Number: WUCSE-2003-30 (2003). All Computer Science and Engineering Research.