High-accuracy phase-equalization filter design using iterative linear-programming technique
Noboru Ito · 2014
In digital communications, non-linear phase channels are equalized by using a phase-equalization filter such that the whole phase responses of the communication channels are linear and the signal waveforms are not distorted after the transmission. This paper presents an iterative linear-programming (LP) technique for designing an all-pass (AP) phase-equalization filter such that its phase response fits a given ideal phase response in the minimax error sense. The AP phase-equalization filter design is formulated as an iterative LP problem and the optimal coefficients of the digital AP system are found through iteratively solving the LP problems. As compared with the existing LP design and the second-order-cone programming (SOCP)-based design, the proposed iterative LP design method achieves much more accurate fitting. We will use a design example to demonstrate the performance improvement.