Peak-constrained least-squares IIR digital filters using recursive quadratic programming

We presented a new weighted least-squares algorithm that can design peak-constrained least-squares IIR digital filters at the IEEE Asilomar Conference on Signals, Systems and Computers, p.349-52 (1994). In this paper we present a new recursive quadratic programming algorithm for designing peak-constrained least-squares IIR digital filters. The new algorithm is better than Deczky's (1972) well-known IIR filter design algorithm because it can simultaneously optimize the peak error and the total squared error.