Huawei Technologies Co., Limited

Jianhong Ke

Chuandong Li

Zhiping Jiang

Dajiang Jin

Xuefeng Tang

Minggang Si

Xiaodong Luo

Zilong He

Hongliang Li

Pre-compensation for linear and non-linear distortion in data symbols comprises generating a set of data symbols to be transmitted and converting the set into a set of frequency-domain samples, which are then pre-processed. Linear distortion compensation is provided, prior to compensating for non-linear distortions, to the pre-processed set of frequency-domain samples. The linear distortion compensated set of frequency-domain samples is converted into a linear distortion compensated set of time domain samples. Non-linear distortion compensation is provided to the linear distortion compensated set of time domain samples. A Pattern Dependent Lookup Table (PDLUT) is generated by receiving consecutive sequences of samples included in a set of time domain samples, quantizing the consecutive sequences of samples into integer-valued consecutive sequences of samples, computing error correction values corresponding to the integer-valued consecutive sequences of samples, and storing the indexes of integer-valued consecutive sequences of samples and the corresponding error correction values in the PDLUT.

Method and system for pre-compensating non-linear distortion in data samples

A digital signal processor (DSP) for a receiver and a method for processing signals in a receiver are provided. The DSP comprises a processor configured to: receive a digital signal at a symbol rate in a frequency domain; and compensate an impairment of the digital signal in the frequency domain.

Method and apparatus for low-complexity symbol-rate receiver digital signal processing

The disclosed systems, and methods are directed to compressing a Pattern Dependent Look-up Table (PDLUT) including distortion correction values corresponding to data symbol sequences, including: i) symmetrically rearranging the data symbol sequences and the associated distortion correction values; ii) folding and amplitude reversing a second half of the rearranged data symbol sequences and the associated distortion correction values; iii) determining approximate distortion correction values for a first half of the rearranged data symbol sequences based on the distortion correction values associated with the first half of the rearranged data symbol sequences and the folded and amplitude reversed distortion correction values associated with the second half of the data symbol sequences; and iv) updating the PDLUT based on the approximate distortion values corresponding to the first half of the rearranged data symbol sequences.

Method and system for compressing and employing pattern dependent look-up table

The disclosed systems, structures, and methods are directed to a single-stage frequency-domain equalization (FDEQ) structure implemented on a processor, comprising a data preprocessing unit configured to concatenate received data samples in time-domain digital signals, transform the concatenated data samples in the time-domain digital signals to frequency-domain digital signals, and an adaptive equalizer comprising 2&#xd7;2 multiple-input multiple output (MIMO) configured to compensate for non-time-varying fixed impairments and time-varying adaptive impairments in the frequency-domain digital signals.

System and method for single-stage frequency-domain equalization

Methods and systems for performing signal pre-correction of a transmission signal comprising a sequence of symbols using a Pattern Dependent Look-up Table (PDLUT) containing one or more distortion correction values. Upon accessing a distortion correction value from the PDLUT for a symbol of the sequence of symbols, the accessed distortion correction value is quantized into one or more quantized values according to one or more quantizations, thereby reducing the bit-width of the distortion correction value. The transmission signal and the distortion correction value with reduced bit-width may undergo linear correction compensation, such as through a finite impulse response (FIR) filter, independent of one another, where the one or more quantized correction values with reduced bit-width reduce a number of calculation steps performed during linear correction compensation resulting in power savings. The linearly compensated quantized values and the linearly compensated signal are combined to form a pre-corrected signal.

Method and apparatus of low-complexity pattern dependent lookup table pre-compensation

Methods and devices are described for frequency domain equalization with low complexity and loop delay. A transmitter inserts pilot symbols into a data signal at intervals of every n data bits. These pilot symbols are used by a receiver-side frequency-domain equalizer to calculate error levels and equalize the received data signal to effect impairment compensation such as SOP tracking. QPSK or BPSK symbols may be used for the pilot symbols, simplifying error calculation into an addition operation instead of the conventional multiplication operation required by conventional FDEQs. Equalizers are described that may operate in a pilot-assisted mode, a conventional decision-directed mode, or both.

Method and apparatus for frequency domain equalization with low complexity and loop delay

Aspects of the present application provide methods and devices for compensating crosstalk in in the digital domain, the crosstalk occurring in the analog domain, for an optical coherent transceiver on a photonic integrated circuit (PIC).

Method and apparatus for transmit/receive radio frequency crosstalk compensation in a photonic integrated circuit

A system monitors optical performance of an optical link within an optical network. The system includes an optical transmitter having an expanded-spectrum pilot-tone modulator for modulating an expanded-spectrum pilot tone onto a high-speed data signal to generate an expanded-spectrum optical signal and an optical receiver for receiving the expanded-spectrum optical signal and for detecting and decoding the expanded-spectrum pilot tone to enable monitoring of the optical performance of the optical link.

Optical channel monitoring using expanded-spectrum pilot tone

Embodiments can provide spurs removal in a pilot-tone spread signal. For achieving this, at least one peak in the pilot-tone spread signal may be found. A predetermined small range of the spectra power around the at least one peak may be removed. In some situations, the removal of the spurs in the pilot-tone spread signal may result in inadvertent removal of a normal part of the pilot-tone spread signal. For addressing this, a power ratio between the spectrum of the pilot-tone spread signal before the removal and after the removal can be calculated. For accounting for the power loss due to the spurs removal, this power ratio can be applied to the pilot-tone spread signal after the removal to obtain a corrected pilot-tone spread signal.

Spur removal in pilot-tone OPM

During optical performance monitoring in low SNR conditions, the detection of pilot data may be more difficult because the detector may mistake noise for the pilot data signal. Systems and methods are disclosed herein that try to address this problem. In one embodiment, a pilot tone detector processes the received signal to determine a maximum correlation peak, and then performs tracking of the correlation peak over time. Unlike the pilot data signal, noise is typically more transient in nature. Therefore, if a correlation peak does not actually correspond to the pilot data signal, but instead corresponds to noise, then the correlation peak typically disappears over time when tracked. A search for a new correlation peak may then be performed. When a correlation peak is determined that actually corresponds to the pilot data signal, then the correlation peak typically remains when tracked.

System and method for pilot data detection using correlation peak tracking

An aspect of the disclosure provides methods and systems for encoding a data bit stream onto a pilot tone signal. Another aspect of the disclosure provides method and systems for pilot tone detection. In both, a coded pilot tone signal is encoded using a code sequence mfor each bit value of 1 (b) and a code sequence mfor each bit value of 0 (b) of a data bit stream including pilot tone data bit values of 1 (b) and bit values of 0 (b), with each code sequence having multiple coding bits in the duration of each bit. Pilot tone detection can further include decoding each code sequence of the coded pilot tone signal using a plurality of successive overlapping measurement windows. In some embodiments each measurement window is of the same duration, being of the duration of each code sequence, and detecting each code sequence comprises selecting one of the plurality of measurement windows to represent a complete code sequence.

Growing community of inventors

Stittsville, Canada

Jianhong Ke