OPTICAL/ELECTRICAL MODULE, COMMUNICATION METHOD, AND RELATED DEVICE

    公开(公告)号:US20240154699A1

    公开(公告)日:2024-05-09

    申请号:US18416436

    申请日:2024-01-18

    CPC classification number: H04B10/2507 H04B10/40 H04B2210/006

    Abstract: An optical/electrical module includes an analog equalizer and an optical/electrical conversion unit. The analog equalizer is connected to the optical/electrical conversion unit. The analog equalizer is configured to: perform first analog signal equalization processing on a first electrical signal to obtain a second electrical signal, and send the second electrical signal to the optical/electrical conversion unit, or is configured to: receive a third electrical signal from the optical/electrical conversion unit, and perform second analog signal equalization processing on the third electrical signal to obtain a fourth electrical signal. The optical/electrical conversion unit is configured to: receive the second electrical signal from the analog equalizer and convert the second electrical signal into a first optical signal. According to embodiments of this application, equalization processing may be performed on an electrical signal in an analog domain.

    Transmitter, receiver, and signal processing method

    公开(公告)号:US10594538B2

    公开(公告)日:2020-03-17

    申请号:US16045725

    申请日:2018-07-25

    Abstract: A transmitter, a receiver, and a signal processing method are provided. The transmitter includes a constellation mapper, a signal conversion module, a digital signal processor, and a digital-to-analog converter. The constellation mapper is configured to determine a mapping relationship between a bit stream and a constellation point in a polar coordinate system, and generate a constellation symbol data flow according to the mapping relationship. The signal conversion module is configured to convert the constellation symbol data flow into an amplitude signal and a phase signal, where the amplitude signal is a 2-level analog signal, and the phase signal is an 8-level digital signal. The digital signal processor is configured to perform digital signal processing on the phase signal, to generate a multi-level digital signal. The digital-to-analog converter is configured to convert the multi-level digital signal into a multi-level analog signal.

    Neural Network Model Processing Method and Apparatus

    公开(公告)号:US20220121936A1

    公开(公告)日:2022-04-21

    申请号:US17434563

    申请日:2019-02-27

    Abstract: A neural network model processing method includes obtaining a first low-bit neural network model through training, where the model includes a first operation layer and a second operation layer. Each operation layer includes at least one operation. Values/a value of a parameter and/or data used for the operation are/is represented by using N bits, and N is a positive integer less than 8. The neural network model processing method further includes compressing the model to obtain a second low-bit neural network model, where the compressed model includes a third operation layer. The third operation layer is equivalent to the first operation layer and the second operation layer, and an operation layer other than the third operation layer in the at least one operation layer is the same as an operation layer other than the first operation layer and the second operation layer in the at least two operation layers.

Patent Agency Ranking