A method for fast feature extraction in threshold scans

We present a fast, analytical method to calculate the threshold and noise parameters from a threshold scan. This is usually done by fitting a response function to the data which is computationally very intensive. The runtime can be minimized by a hardware implementation, e.g. using an FPGA, which in turn requires to minimize the mathematical complexity of the algorithm in order to fit into the available resources on the FPGA. The systematic errors of the method are analyzed and reasonable choices of the parameters for use in practice are given.