Distributed compression amplitude mapping for a cochlear stimulation system
View Patent ↗An implantable neural stimulation system, such as a cochlear implant system, utilizes a Distributed Compression Amplitude Mapping (DCAM) system to distribute signal compression between a pre-bandpass linear mapping function, and a post-bandpass compressive mapping function. The pre-bandpass linear mapping function is implemented, in one exemplary embodiment, as a traditional audio compressor to prevent distortion that might result from a non-linear mapping function. The post-bandpass compressive mapping function is implemented, in another exemplary embodiment, as a logarithmic transform to reflect natural hearing. As a result of the DCAM processing, the differences in amplitudes of components of the acoustic spectrum are maintained. By maintaining these differences, spectral smearing between channels is reduced and speech clues are preserved.
1. An implantable cochlear stimulation (ICS) system comprising:
a microphone adapted to convert an acoustical signal to a broad spectrum electrical signal;
a speech processor connected to the microphone and configured to
subject the broad spectrum electrical signal to a linear compressive mapping function to map the broad spectrum electrical signal to a first mapped signal,
process the first mapped signal with a number of bandpass filters to separate the first mapped signal into a plurality of frequency band output signals,
subject the plurality of frequency band output signals to a non-linear compressive mapping function to map the plurality of frequency band output signals to a multiplicity of second mapped signals; and
an electrode array adapted to electrically stimulate a cochlea with signals derived from the first mapped signal and the second mapped signals;
wherein the speech processor is configured to subject the broad spectrum electrical signal to the linear compressive mapping function to map the broad spectrum electrical signal to the first mapped signal by
detecting an envelope of the broad spectrum electrical signal to produce a detected envelope signal;
generating a processed envelope signal that follows the detected envelope signal at a limited rate; and
combining the processed envelope signal with the broad spectrum electrical signal to produce the first mapped signal.
2. The system of claim 1 wherein the speech processor comprises a Wearable Speech Processor (WP).
3. The system of claim 1 wherein the speech processor comprises a Behind-The-Ear (BTE) speech processor.
4. The system of claim 1 wherein the speech processor comprises an element of a fully implantable cochlear stimulator.
5. The system of claim 1 wherein the speech processor is configured to combine the processed envelope signal with the broad spectrum electrical signal modifying the processed envelope signal with a gain factor to produce a gain output multiplying the gain output by the broad spectrum electrical signal to produce the first mapped signal.
6. The system of claim 1 wherein the speech processor is configured to generate the processed envelope signal by using an
equation that includes
C ( n )= C ( n− 1)+(Env n )− C ( n− 1))*λ;
wherein Env(n) is a current value of the detected signal envelope, C(n−1) is a previous value of the processed envelope signal, and C(n) is a new value of the processed envelope signal, and λ depends on a mode of the speech processor.
7. The system of claim 6 , wherein the mode of the speech processor includes one of an attack mode and a release mode, and wherein the value of λ in the attack mode is greater than the value of λ in the release mode.
8. The system of claim 7 wherein the value of λ in the attack mode is about 0.2, and the value of λ in the release mode is about 0.0018.
9. The system of claim 1 wherein the non-linear compressive mapping function comprises a logarithmic mapping function.