Search

In the scenario of downstream Passive Optical Networks (PON), we analyze through simulations and experiments a 100 Gbps/λ link using digital signal processing (DSP) over up to 50 km single mode fiber (SMF) in C-band. In particular, we propose chromatic dispersion digital pre-compensation (CD-DPC) and quaternary pulse amplitude modulation (PAM-4) levels optimization at the transmitter side (TX), and simple non-linear compensation (NLC) in combination with adaptive equalization (AEQ) at the receiver side (RX). Regarding NLC, we compare two approaches: a quadratic polynomial function and a square-root-like function respectively. In this paper, we analyze in detail the performances of four proposed options, namely equispaced PAM-4 levels without NLC (baseline case), equispaced PAM-4 levels in combination with NLC, optimized PAM-4 levels without NLC, and optimized PAM-4 levels in combination with NLC. We demonstrate through simulations and experiments that optimized PAM-4 levels can only offer limited enhancement when NLC is applied, and up to 3.3 dB sensitivity gain can be obtained thanks to NLC at RX when setting the optimum parameters, with respect to the baseline case.

Passive Optical Networks (PON), able to operate at 50 Gbps per wavelength (λ), are under development and standardization, based on intensity-modulation (IM) and direct-detection (DD) systems. The next step in PON evolution will be driven by 5G/6G fronthauling capacity demands, and will require the development of 100 Gbps/λ (and beyond) systems, which poses big challenges if maintaining the DD-format. In this contribution, we analyze a 100 Gbps/λ PON architecture able to preserve the IM-DD approach at the Optical Network Unit (ONU), placing the complexity at the Optical Line Terminal (OLT), thanks to Digital Signal Processing (DSP). We experimentally demonstrate a 100 Gbps/λ transmission using this architecture in the downstream (DS) direction. Chromatic dispersion digital pre-compensation (CD-DPC) in combination with an IQ Mach-Zehnder Modulator (IQ-MZM) is used at the transmitter (TX). Keeping the ONU DSP as simple as possible, as compared with current DSP proposals for 50 Gbps/λ PON, is another main goal of this work. Adaptive equalization (AEQ) is used to correct for linear impairments, in addition to digital non-linear correction (NLC) at the receiver (RX). We compare two NLC approaches: a full Volterra Non-Linear Equalizer (VNLE) and a simpler NLC technique based on a square-root like function (SQRT). Operation over standard single-mode fiber (SMF) in C-band, achieving reaches from 0 km to 100 km and Optical Path Loss (OPL) values higher than 29 dB, are shown. The analyzed proposal is directly applicable to Terabit-capable wavelength division multiplexing (WDM)-PON, and can be extended to very high-speed Time Division Multiplexing (TDM)-PON and TWDM-PON, with some modifications discussed here. [ABSTRACT FROM AUTHOR]

Smooth polishing has a good improvement on the medium and high frequency errors of the workpiece, but the surface accuracy of the workpiece will be damaged subsequently. Therefore, a nonlinear compensation method is proposed. The realization mechanism, mathematical model and compensation effect of nonlinear compensated smooth polishing are illustrated separately and verified by experiments. In addition, the process parameters and motion trajectory are optimized by theoretical simulation. The experimental results demonstrate that the full-band error of the workpiece can be improved simultaneously by the smooth polishing method based on nonlinear compensation. The number of iterations for the figuring experiments and smooth polishing is greatly reduced by the nonlinear compensation model, which is significantly beneficial to improve the surface convergence speed and lower the manufacturing cost. • Mathematical model of nonlinear compensation is developed. • The full-band error of the workpiece is converged. • Toolpath is optimized. • Process parameters are optimized. [ABSTRACT FROM AUTHOR]

A new type of piezoelectric vibration acceleration sensor was studied, which sensitive element consists of cymbal transducer and a buffer chip. The nonlinear characteristic of voltage sensitivity of this vibration acceleration sensor was investigated. Using the general principles of nonlinear calibration method of BP neural network, established the mathematic model of nonlinear calibration of output voltage sensitivity for this piezoelectric vibration acceleration sensor, and then accomplished the computer simulation programming using MATLAB algorithm. The computer simulation results showed that the nonlinear error of output voltage sensitivity for this piezoelectric vibration acceleration sensor can be less than 0.1‰ after six-steps training using artificial neural network. The effect of the compensation is better than that of commonly piecewise linear interpolation method, and with the characteristic of simple structure and high accuracy. [ABSTRACT FROM AUTHOR]