FPGA

The quantization approach used in this paper has an added advantage when dealng with current FPGA hardware. FPGAs, like early processors, are much more efficient on integer operations, than floating point operations. In addition they provide much more precision flexiblity than traditional processors. Floating point arrithmtic can be replaced by fixed-point integer math. Early work suggests that FFT portion of migration can be sped up by a factor of 3-10 given current hardware.