Exponential signal synthesis in digital pulse processing

Digital pulse processing allows the synthesis of exponential signals that can be used in pulse shaping and baseline restoration. A recursive algorithm for the synthesis of high-pass filters is presented and discussed in view of its application as a baseline restorer. The high-pass filter can be arranged in a gated baseline restorer configuration similar to widely used analog implementations. Two techniques to synthesize time-invariant, finite impulse response (FIR) cusp shapers are presented. The first technique synthesizes a true cusp shape in the discrete-time domain. This algorithm may be sensitive to round-off errors and may require a large amount of computational resources. The second method for synthesis of cusp shapes is suitable for implementation using integer arithmetic, particularly in hardware. This algorithm uses linear interpolation to synthesize close approximations of true cusp shapes. The algorithm does not introduce round-off errors and has been tested in hardware.