Tone detector and method therefor
View Patent ↗A tone detector able to detect multicomponent tones for various tone formats is disclosed. An input signal is received and frequency and amplitude components, if any, of each sample of the incoming signal are estimated. In estimating the components, a current digit value is used to appropriately filter the input signal and estimate frequency and amplitude components. Using the component estimates, a new digit value is determined using a table in storage corresponding to the current tone format. The new digit value corresponds to the symbol whose components are closest in value to the estimated components corresponding to the current digit value. The new digit value is fed back as the current digit value and used in combination with a next sample of the input signal to determine a new digit value. A decision unit is used to process the digit values to determine when a valid symbol is detected.
1. A method for performing tone detection, comprising:
receiving an input signal;
selecting a first tone format, wherein a first tone corresponds to a first symbol;
providing a first table of values corresponding to the first tone format, wherein at least one entry in the first table of values relates the first tone to the first symbol;
using a predetermined algorithm and the first table of values to determined if the input signal comprises the first symbol;
selecting a second tone format, wherein a second tone corresponds to the first symbol;
providing a second table of values corresponding to the second tone format, wherein at least one entry in the second table of values relates the second tone to the first symbol; and
using the same predetermined algorithm and the second table of values to determined if the input signal comprises the first symbol,
wherein the first tone format and the second tone format are different, and wherein the first table of values and the second table of values have at least one value that is different.
2. A method as in claim 1 , wherein the first tone format is dual-tone multiple frequency (DTMF).
3. A method as in claim 1 , wherein the first tone format comprises a plurality of components having overlapping frequency bands.
4. A method as in claim 1 , wherein the algorithm is non-linear and does not utilize Fourier transform computations.
5. A method as in claim 1 , wherein the algorithm extracts a plurality of tone components using a plurality of comb filters.
6. A method for detecting a tone, comprising:
receiving an input signal;
filtering the input signal so that a first finite set of selected frequencies are not allowed to pass in order to produce a first tone component;
filtering the input signal so that a second finite set of selected frequencies are not allowed to pass in order to produce a second tone component; and
using the first and second tone components to detect the tone, wherein the step of using the first and second tone components to detect the tone comprises:
estimating a frequency component of the first tone component using a first polynomial function of the first tone component; and
estimating a frequency component of the second tone component using a first polynomial function of the second tone component.
7. A method as in claim 6 , wherein a first comb filter having a first set of coefficients is used to perform the step of filtering the input signal so that a first finite set of selected frequencies are not allowed to pass, and wherein a second comb filter having a second set of coefficients is used to perform the step of filtering the input signal so that a second finite set of selected frequencies are not allowed to pass.
8. A method as in claim 6 , wherein the first set of coefficients is adaptive and the second set of coefficients is adaptive.
9. A method as in claim 6 , wherein the step of using the first and second tone components to detect the tone further comprises:
estimating a magnitude component of the first tone component using a second polynomial function of the first tone component; and
estimating a magnitude component of the second tone component using a second polynomial function of the second tone component.
10. A method as in claim 9 , wherein the step of estimating the magnitude component of the first tone component comprises:
using a moving average to extract a magnitude of the first tone component.
11. A method as in claim 6 , wherein the step of estimating the frequency component of the first tone component comprises:
performing low-pass filtering using adaptive coefficients in order to reduce noise.
12. A method as in claim 6 , wherein the step of using the first and second tone components to detect the tone comprises:
deciding after each sample of the input signal whether the tone has been detected, wherein samples of the input signal need not be stored.
13. A method as in claim 6 , further comprising:
estimating power of the input signal using a polynomial function of the tone; and
if the power of the input signal is below a predetermined threshold, performing no further processing on the input signal until the power of the input signal is no longer below the predetermined threshold.
14. A method as in claim 6 , further comprising:
adjusting gain of the input signal.
15. A method as in claim 6 , wherein the input signal is dual-tone multiple frequency (DTMF) signal.
16. A method as in claim 6 , wherein an amount of time required to estimate a frequency component of the first tone component is independent of a frequency of the first tone component.
17. A method as in claim 6 , wherein the step of using the first and second tone components to detect the tone comprises:
using a first table for a first tone format; and
using a second table for a second tone format.
18. A method as in claim 6 , further comprising:
determining if the detected tone corresponds to a valid symbol of a tone format corresponding to the detected tone.