Digital quasi-RMS detector

1. A method of approximating a root-mean-square (RMS) value for an audio signal in a digital hearing aid, comprising the steps of:

receiving the audio signal;

rectifying the audio signal to generate a rectified signal;

determining a difference between the rectified signal and an estimated present energy value of the audio signal;

if the difference is not greater than zero, then multiplying the rectified signal by a release time constant to generate a first filtered signal and summing the first filtered signal with the estimated present energy value to determine the approximate RMS value; and

if the difference is greater than zero, then multiplying the rectified signal by an attack time constant to generate a second filtered signal and summing the second filtered signal with the estimated present energy value to determine the approximate RMS value; wherein the attack time constant is less than the release time constant.

2. The method of claim 1 , wherein the attack time constant is a fraction of the release time constant.

3. The method of claim 1 , comprising the additional step of:

setting the estimated present energy value to the approximate RMS value.

4. The method of claim 1 , comprising the additional step of:

prior to the rectifying step, sampling the audio signal to generate a sampled audio signal, wherein the sampled audio signal is rectified to generate the rectified signal.

5. A method of approximating a root-mean-square (RMS) value for an audio signal in a digital hearing instrument, comprising the steps of:

receiving the audio signal;

rectifying the audio signal to generate a rectified signal;

determining a difference between the rectified signal and an estimated present energy value of the audio signal;

generating a filtered audio signal by shifting a binary representation of the rectified signal in relation to the difference between the rectified signal and the estimated present energy value of the audio signal; and

summing the filtered audio signal with the estimated present energy value to determine the approximate RMS value.

6. The method of claim 5 , comprising the additional step of:

setting the estimated present energy value to the approximate RMS value.

7. The method of claim 5 , wherein if the difference is greater than zero (0), then generating the filtered audio signal by shifting the binary representation of the rectified signal to the right by a predetermined number of register positions.

8. The method of claim 7 , wherein the binary representation of the rectified signal is shifted right by two register positions.

9. The method of claim 5 , wherein:

if the difference is greater than zero (0), then generating the filtered audio signal by shifting the binary representation of the rectified signal to the right by a first number of register positions; and

if the difference is not greater than zero (0), then generating the filtered audio signal by shifting the binary representation of the rectified signal to the right by a second number of register positions.

10. The method of claim 9 , wherein the second number of register positions is greater than the first number of register positions.

11. The method of claim 10 , wherein the first number of register positions is five (5) and the second number of register positions is seven (7).

12. A digital hearing instrument, comprising:

a microphone that receives an acoustical signal and generates a wideband audio signal;

a band-split filter coupled to the microphone that filters the wideband audio signal into a plurality of channel audio signals;

means for rectifying each channel audio signal to generate a rectified signal;

means for determining a difference between each rectified signal and an estimated present energy value for each corresponding channel audio signal;

means for shifling a binary representation of each rectified signal in relation to the difference between the rectified signal and the estimated present energy value of the corresponding channel audio signal to generate a filtered audio signal for each rectified signal;

means for summing each filtered audio signal with the corresponding estimated present energy value to determine an approximate energy value for each channel audio signal;

a plurality of channel processors coupled to the band-split filter that amplify each of the channel audio signals by a channel gain to generate a conditioned channel signal, wherein the channel gain for each channel audio signal is determined as a function of the approximate energy value for the channel audio signal;

a summation circuit coupled to the plurality of channel processors that sums the conditioned channel signals from the channel processors and generates a composite signal; and

a speaker coupled to the summation circuit that receives the composite signal and generates an acoustical output signal.

13. The digital hearing instrument of claim 12 , wherein the channel gain for at least one channel audio signal is determined as a function of the approximate energy value for the one channel audio signal and the approximate energy value for one other channel audio signal.

14. The digital hearing instrument of claim 12 , further comprising:

means for setting the estimated present energy value for each channel audio signal to the approximate energy value for the corresponding channel audio signal.

15. The digital hearing instrument of claim 12 , wherein if the difference between any rectified signal and the estimated present energy value for the corresponding channel audio signal is greater than zero (0), then the shifting means shifts the binary representation of the rectified signal to the right by a predetermined number of register positions.

16. The digital hearing instrument of claim 15 , wherein the binary representation of the rectified signal is shifted to the right by two resister positions.

17. The digital hearing instrument of claim 12 , wherein:

if the difference between any rectified signal and the estimated present energy value for the corresponding channel audio signal is greater than zero (0), then the shifting means shifts the binary representation of the rectified signal to the right by a first number of register positions; and

if the difference between any rectified signal and the estimated present energy value for the corresponding channel audio signal is not greater than zero (0), then the shifting means shifts the binary representation of the rectified signal to the right by a second number of register positions.

18. The digital hearing instrument of claim 17 , wherein the second number of register positions is greater than the first number of register positions.

19. The digital hearing instrument of claim 18 , wherein the first number of register positions is five (5) and the second number of register positions is seven (7).

20. A digital hearing instrument, comprising:

a microphone that receives an acoustical signal and generates a wideband audio signal;

a band-split filter coupled to the microphone that filters the wideband audio signal into a plurality of channel audio signals;

means for rectifying each channel audio signal to generate a rectified signal;

means for determining a difference between each rectified signal and an estimated present energy value for each corresponding channel audio signal;

means for multiplying each rectified signal by a release time constant if the difference between the rectified signal and the estimated present energy value for the corresponding channel audio signal is not greater than zero to generate a filtered channel signal, and for multiplying each rectified signal by an attack time constant if the difference between the rectified signal and the estimated present energy value for the corresponding channel audio signal is greater than zero to generate the filtered channel signal;

means for summing each filtered channel signal with the corresponding estimated present energy value to determine an approximate energy value for each channel audio signal;

a plurality of channel processors coupled to the band-split filter that amplify each of the channel audio signals by a channel gain to generate a conditioned channel signal, wherein the channel gain for each channel audio signal is determined as a function of the approximate energy value for the channel audio signal;

a summation circuit coupled to the plurality of channel processors that sums the conditioned channel signals from the channel processors and generates a composite signal; and

a speaker coupled to the summation circuit that receives the composite signal and generates an acoustical output signal.

21. A digital hearing instrument, comprising:

a microphone that receives an acoustical signal and generates a wideband audio signal;

a band-split filter coupled to the microphone that filters the wideband audio signal into a plurality of channel audio signals;

a plurality of twin detectors that determine an estimated energy level for each of the channel audio signals by rectifying each channel audio signal to generate a rectified signal, determining a difference between each rectified signal and an estimated a present energy value for each corresponding channel audio signal, generating a filtered audio signal for each rectified signal by shifting a binary representation of the rectified signal in relation to the difference between the rectified signal and the estimated present energy value of the corresponding channel audio signal, and summing each filtered audio signal with the corresponding estimated present energy value;

a plurality of channel processors coupled to the band-split filter and the twin detectors that amplify each of the channel audio signals by a channel gain to generate a conditioned channel signal, wherein the channel gain for each channel audio signal is determined as a function of the approximate energy value for the channel audio signal;

a summation circuit coupled to the plurality of channel processors that sums the conditioned channel signals from the channel processors and generates a composite signal; and

a speaker coupled to the summation circuit that receives the composite signal and generates an acoustical output signal.