Touch sensor gesture recognition for operation of mobile devices

Abstract

Touch sensor gesture recognition for operation of mobile devices. An embodiment of a mobile device may include a touch sensor for the detection of gestures, the touch sensor including sensor elements to generate touch sensor data for a detected gesture, and a processor to process the touch sensor data produced by the sensor elements of the touch sensor. In some embodiments the processor is to process the touch sensor data using a hybrid touch sensor algorithm, the hybrid touch sensor algorithm including a plurality of touch sensor algorithms, the plurality of touch sensor algorithms including a first algorithm and a second algorithm, where the processor dynamically changes between the plurality of algorithms depending on of the nature of the received touch sensor data. In some embodiments the processor utilizes a support vector machine with a radial basis function kernel in the interpretation of detected gestures. In some embodiments, the processor is to determine an alignment between a first input sequence of sensor data and a second input sequence of sensor data using dynamic time warping.

What is claimed is: 
     
         1 . A mobile device comprising:
 a touch sensor for the detection of gestures, the touch sensor including a plurality of sensor elements; and   a processor, the processor to process sensor data from the touch sensor to interpret the gestures detected by the touch sensor;   wherein the processor utilizes a support vector machine with a radial basis function kernel in the interpretation of detected gestures.   
     
     
         2 . The mobile device of  claim 1 , wherein support vectors for the radial basis function are determined in a training phase for the support vector machine. 
     
     
         3 . The mobile device of  claim 2 , wherein the training phase includes providing training data for use in the determination of the support vectors. 
     
     
         4 . The mobile device of  claim 1 , wherein an output of the support vector machine for sensor data representing a detected gesture includes gesture best matches for the sensor data. 
     
     
         5 . The mobile device of  claim 4 , further comprising a post filter to filter the gesture best matches to produce gesture output data. 
     
     
         6 . The mobile device of  claim 5 , further comprising a feedback filter receiving the gesture output and providing feedback for generation of the radial basis function. 
     
     
         7 . A method comprising:
 detecting a gesture at a touch sensor of a mobile device, the touch sensor including a plurality of sensor elements; and   processing the sensor data from the touch sensor to interpret the gesture;   wherein the processing includes providing the sensor data to a support vector machine with a radial basis function kernel.   
     
     
         8 . The method of  claim 7 , further comprising conducting a training phase to determine support vectors for the radial basis function. 
     
     
         9 . The method of  claim 8 , wherein the training phase includes providing training data for use in the determination of the support vectors. 
     
     
         10 . The method of  claim 7 , further comprising producing an output of the support vector machine including gesture best matches for the sensor data. 
     
     
         11 . The method of  claim 10 , further comprising filtering the gesture best matches to produce gesture output data. 
     
     
         12 . The method of  claim 11 , further comprising feeding back filtered gesture output data for generation of the radial basis function. 
     
     
         13 . A non-transitory computer-readable storage medium having stored thereon data representing sequences of instructions that, when executed by a processor, cause the processor to perform operations comprising:
 detecting a gesture at a touch sensor of a mobile device, the touch sensor including a plurality of sensor elements; and   processing the sensor data from the touch sensor to interpret the gesture;   wherein the processing includes providing the sensor data to a support vector machine with a radial basis function kernel.   
     
     
         14 . The medium of  claim 13 , further comprising instructions that, when executed by the processor, cause the processor to perform operations comprising:
 conducting a training phase to determine support vectors for the radial basis function.   
     
     
         15 . The medium of  claim 14 , wherein the training phase includes providing training data for use in the determination of the support vectors. 
     
     
         16 . The medium of  claim 13 , further comprising instructions that, when executed by the processor, cause the processor to perform operations comprising:
 producing an output of the support vector machine including gesture best matches for the sensor data.   
     
     
         17 . The medium of  claim 16 , further comprising instructions that, when executed by the processor, cause the processor to perform operations comprising:
 filtering the gesture best matches to produce gesture output data.   
     
     
         18 . The medium of  claim 17 , further comprising instructions that, when executed by the processor, cause the processor to perform operations comprising:
 feeding back filtered gesture output data for generation of the radial basis function.