Fast efficient evaluation of messages on automotive networks using look-up tables
Abstract:
A Support Vector Machine (SVM) classifier employs a kernel ƒ(xi,y)∝Πd=1DF(xi,d,yd) or ƒ(xi,y)∝σd=1DF(xi,d,yd) where D is the dimensionality of Support Vectors xi and vector y is a vector being classified, e.g. representing a message communicated on a Controller Area Network (CAN) bus of a vehicle. The SVM classifier comprises a microprocessor or microcontroller programmed to perform classification using the SVM classifier with at least some terms F(xi,d,yd) of the kernel ƒ(xi,y) determined using a look-up table. For those terms F(xi,d,yd) of the kernel ƒ(xi,y) determined using a look-up table, the dimension yd can only assume values defined by a finite set of values. In an embedded system, the microprocessor or microcontroller performs an embedded system function such as engine control, brake control, entertainment system operation, or so forth, or an embedded system function for a vehicle, ocean buoy, or satellite.
Information query
Patent Agency Ranking
0/0