Digital Finite-Impulse-Response Filtering of Transmitted Data Signals for Nyquist Channels

The spectral shaping of multilevel data signals is of fundamental importance in advanced data communication systems, where it is often used to achieve bandwidth efficient Nyquist (i.e., zero intersymbol interference) channels. For transmitted M-state signals, the traditional filtering methods usually engender lengthy development intervals and costly manufacturing methods. 

In this talk, a memory based digital approach is described that not only mitigates the aforementioned limitations, but intrinsically codes high-speed data into M-state symbols while ensuring highly reproducible baud normalized impulse shaping.