Search

Recently, PAM-4 transmitters have been realized at 112Gb/s or even faster to satisfy the continuously growing demands for wireline communications [1]–[5]. Although PAM-4 signaling performs two-fold bandwidth efficiency compared with the NRZ counterpart, the NRZ signal still has the advantage in low-loss scenarios like ultra-short-reach (USR) interconnection for optical engines due to its high output swing and better SNR. This paper presents a 100Gb/s NRZ transmitter fabricated in 40nm CMOS by incorporating a 7b DAC with an 8-tap FFE implemented in DSP. The DAC-based architecture demonstrates high resolution for FFE coefficients, achieving an output eye diagram with 73mV eye height and 760fs rms jitter under 7.1dB loss at 50GHz while consuming 619mW of power from 1.1V/1.2V/1.5V supplies.

Hydraulic active suspension systems have nonlinearity and time-varying dynamic behaviour. It is difficult to establish an accurate model for traditional controllers design. This unknown nonlinear dynamic system is approximated by an second order linear sub-model plus a nonlinear fuzzy sub-model in this study. An on-line recursive least square with forgetting factor scheme is employed to establish this linear sub-model. Then the pole assignment adaptive control strategy is used to control this simplified linear time-varying system. A fuzzy control loop is introduced into this adaptive control structure to detect the unknown nonlinear sub-model for compensating the system modelling error and improving the control performance. The experimental results show that this control strategy has effectively reduced the position and acceleration oscillation amplitudes of vehicle body due to the road surface disturbances. It can be employed to improve the vehicle riding comfort.