Method and system for real-time signal classification
First Claim
Patent Images
1. A method for real-time signal classification, the method to be executable on a low-power processor, the method comprising:
- providing a managed bandwidth signal from an accelerometer sensor;
establishing, by a processor, a time bound for executing steps in the method;
windowing, by the processor, within the time bound, the managed bandwidth signal to produce an acceleration vector based on an acceleration model;
performing, by the processor, within the time bound, feature computation using fixed point arithmetic on the vector to produce a model-compatible acceleration vector based on the acceleration model;
projecting, by the processor, within the time bound, the model-compatible acceleration vector having a first multi-dimensional factor space onto a second multi-dimensional factor space where the second multi-dimensional factor space is smaller than the first multi-dimensional factor space to produce a lower dimensional factor vector; and
evaluating, by the processor, within the time bound and using log representations, the lower dimensional factor vector against the acceleration model to produce a classification result.
1 Assignment
0 Petitions
Accused Products
Abstract
A method to achieve an accurate, extremely low power state classification implementation is disclosed. Embodiments include a sequence that matches the data flow from the sensor transducer, through analog filtering, to digital sampling, feature computation, and classification.
-
Citations
20 Claims
-
1. A method for real-time signal classification, the method to be executable on a low-power processor, the method comprising:
-
providing a managed bandwidth signal from an accelerometer sensor; establishing, by a processor, a time bound for executing steps in the method; windowing, by the processor, within the time bound, the managed bandwidth signal to produce an acceleration vector based on an acceleration model; performing, by the processor, within the time bound, feature computation using fixed point arithmetic on the vector to produce a model-compatible acceleration vector based on the acceleration model; projecting, by the processor, within the time bound, the model-compatible acceleration vector having a first multi-dimensional factor space onto a second multi-dimensional factor space where the second multi-dimensional factor space is smaller than the first multi-dimensional factor space to produce a lower dimensional factor vector; and evaluating, by the processor, within the time bound and using log representations, the lower dimensional factor vector against the acceleration model to produce a classification result. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9)
-
-
10. A system for real-time signal classification, the system having a lower power processor, the system comprising:
-
means for providing a managed bandwidth signal from an accelerometer sensor; means for establishing a time bound for execution of classification processing in the system; means for windowing, within the time bound, the managed bandwidth signal to produce an acceleration vector based on an acceleration model; means for performing, within the time bound, feature computation using fixed point arithmetic on the acceleration vector to produce a model-compatible acceleration vector based on the acceleration model; means for projecting, within the time bound, the model-compatible acceleration vector having a first multi-dimensional factor space onto a second multi-dimensional factor space where the second multi-dimensional factor space is smaller than the first multi-dimensional factor space to produce a lower dimensional factor vector; and means for evaluating, within the time bound and using log representations, the lower dimensional factor vector against the acceleration model to produce a classification result. - View Dependent Claims (11, 12, 13, 14, 15, 16, 17, 18, 19, 20)
-
Specification