Method and apparatus for automated identification of signal characteristics
A method of assessing a signal to identify particular signal characteristics comprises application of machine learning to multi-dimensional histograms derived from multi-tap sampling of the signal. The signal is sampled from at least two tap points to retrieve a sample set, and the at least two tap points are adapted to retrieve distinct samples from the signal, such as time spaced samples or spectrally distinct samples. Multiple sample sets are retrieved from the signal over time. The at least two dimensional histogram is built from the joint probability distribution of the plurality of sample sets. A machine learning algorithm then processes the multi-dimensional histogram, and is trained to predict a value of at least one characteristic of the signal.
1. A method of assessing a signal, the method comprising:
repetitively sampling the signal from at least two tap points to retrieve a single successive sample set at each sampling, the at least two tap points being separated by a fixed time delay, the sampling being asynchronous relative to the signal, each sample set having a single distinct value of the sampled signal from each and every tap point;
retrieving a plurality of such sample sets over time that define a probability density;
building an at least two dimensional histogram of the probability density of the plurality of sample sets; and
passing the at least two dimensional histogram to a machine learning algorithm trained to predict a value of at least one characteristic of the signal.
2. The method of claim 1 wherein the machine learning algorithm comprises a supervised learning technique.
3. The method of claim 2 wherein the machine learning algorithm comprises regression.
4. The method of claim 2 wherein the machine learning algorithm comprises ridge regression.
5. The method of claim 2 wherein the machine learning algorithm comprises support vector regression.
6. The method of claim 1 wherein the machine learning algorithm utilizes a linear kernel.
7. The method of claim 1 wherein the machine learning algorithm utilizes a non-linear kernel.
8. The method of claim 7 wherein the machine learning algorithm utilizes a non-linear kernel comprising a polynomial of degree n.
9. The method of claim 1 wherein the machine learning algorithm is trained on training sets created from simulations.
10. The method of claim 1 wherein the machine learning algorithm is trained on training sets created from experimental measurements.
11. The method of claim 1 wherein the two dimensional histogram is built by determining a number of sample points occurring in each of a plurality of sub-areas or sub-volumes of the joint probability density function.
12. A device for assessing a signal, the device comprising:
a multi-tap sampler for repetitively sampling the signal from at least two tap points to retrieve a single successive sample set at each sampling, the at least two tap points being separated by a fixed time delay, the sampling being asynchronous relative to the signal, each sample set having a single distinct value of the sampled signal from each and every tap point, and for retrieving a plurality of sample sets over time that define a probability density;
a processor for building an at least two dimensional histogram of the probability density of the plurality of sample sets; and
a machine learning module trained to predict a value of at least one characteristic of the signal from processing the at least two dimensional histogram.