CN-120238195-B - Method and system for realizing optical fiber interconnection transmission of high-speed serial transceiver based on FPGA

CN120238195BCN 120238195 BCN120238195 BCN 120238195BCN-120238195-B

Abstract

The invention relates to the technical field of optical fiber communication, in particular to a method and a system for realizing optical fiber interconnection transmission by a high-speed serial transceiver based on an FPGA, comprising the steps of constructing a link characteristic baseline and combining a multidimensional weighted quality index drive to dynamically adjust a pre-emphasis coefficient and a receiving balance parameter of the transceiver by monitoring bit error rate BER and eye pattern characteristic parameters of an optical fiber link in real time; extracting signal jitter characteristics, dynamically adjusting clock data recovery loop bandwidth based on the jitter characteristics and bit error rate, dynamically switching a forward error correction mode through hysteresis threshold logic, balancing time delay and error correction capability, and continuously updating baseline parameters through a closed loop feedback mechanism to realize real-time tuning of the quality of an optical fiber communication link.

Inventors

ZHONG YUELIANG
LIN HAO
Xia yuanyang

Assignees

昆山软龙格自动化技术有限公司

Dates

Publication Date: 20260508
Application Date: 20250414

Claims (2)

1. The method for realizing optical fiber interconnection transmission of the high-speed serial transceiver based on the FPGA is characterized by comprising the following steps of: S1, acquiring bit error rate BER and eye pattern characteristic parameters of a received signal in an optical fiber transmission process, wherein the eye pattern characteristic parameters comprise eye height EH and eye width EW, and establishing a link characteristic baseline by combining the BER and the eye pattern characteristic parameters, wherein the link characteristic baseline is Wherein Representing an initial bit error rate; Representing an initial eye height; The optical fiber transmission process comprises photoelectric conversion, modulation and demodulation processes, wherein the FPGA realizes modulation of optical signals through an embedded high-speed serial transceiver, the modulated optical signals are transmitted to an optical fiber link, and the optical signals are recovered at a receiving end through the demodulation process; s2, constructing a multi-dimensional weighted quality index MQM, wherein the MQM is obtained by integrating a normalized scoring function and a weight coefficient of BER, EH and EW, and the multi-dimensional weighted quality index is expressed as follows: ; Wherein, the Representing a multi-dimensional weighted quality index; Representing a bit error rate normalization scoring function; representing a eye height normalization scoring function; Representing an eye width normalization scoring function; 、 And Representing the weight coefficient; The method comprises the following steps of adjusting a transmission pre-emphasis coefficient and a receiving equalization parameter of a high-speed serial transceiver based on an improved particle swarm algorithm in an MQM driving FPGA, wherein the improved particle swarm algorithm comprises the following optimization processes: s2.1, initializing a pre-emphasis coefficient and a parameter range of a receiving equalization parameter, and randomly generating an initial particle group, wherein the position of each particle represents a group of parameter combinations; S2.2, calculating the MQM of each particle under the current parameters, estimating the gradient of the MQM to the parameters by adopting a finite difference method, and updating the current speed of the particle based on the gradient; S2.3, continuing iteration, judging convergence and outputting an optimal parameter combination if the MQM change rate is smaller than a threshold value; S3, extracting jitter characteristics of a received signal in an optical fiber link, wherein the jitter characteristics comprise jitter standard deviation and jitter peak value, clock data recovery CDR bandwidth is dynamically adjusted according to the jitter characteristics and current BER to balance jitter tracking capacity and noise suppression performance, recovery of the optical signal is accurately controlled by using a nonlinear mapping rule, the nonlinear mapping rule is realized through a lookup table, the input of the lookup table is a joint quantization interval of a jitter indication factor and a bit error rate, the joint quantization interval is output as a preset CDR bandwidth grade, the joint quantization interval is used for carrying out joint analysis on the jitter interval and the BER interval, the jitter interval comprises a low jitter interval, a medium jitter interval and a high jitter interval, the BER interval comprises a low error interval, a medium error interval and a high error interval, the jitter indication factor is obtained based on the jitter standard deviation and the jitter peak value, and the jitter indication factor is as follows: ; Wherein, the Representing a jitter indication factor; representing the jitter standard deviation; representing jitter peaks; And Representing the jitter indication factor influence coefficient; S4, filtering BER data, calculating trend slope, dynamically switching a Forward Error Correction (FEC) mode by combining BER, trend slope and preset hysteresis threshold logic, and optimizing time delay of signal transmission, wherein the trend slope is as follows: ; Wherein, the Representing a trend slope; representing the number of sampling points of a time window; representing a time stamp; Representation of A filtered BER value at a time; hysteresis threshold logic is defined as: when BER exceeds a threshold value and the trend slope is greater than 0, switching from a low error correction mode to a high error correction mode; switching from the high error correction mode to the low error correction mode when BER is less than the threshold and the trend slope is less than 0; s5, updating the baseline parameters in real time to form a closed loop, and continuously performing optical fiber link quality adjustment, wherein the updated baseline parameters are as follows ; Wherein the method comprises the steps of Representing the current bit error rate; Representing the current eye height; representing the current eye width.
2. The high-speed serial transceiver based on the FPGA realizes the optical fiber interconnection transmission system, and is characterized in that the system comprises: The link characteristic base line construction module acquires bit error rate BER and eye pattern characteristic parameters of a received signal in the optical fiber transmission process, wherein the eye pattern characteristic parameters comprise eye height EH and eye width EW, and establishes a link characteristic base line by combining the BER and the eye pattern characteristic parameters, and the link characteristic base line is Wherein Representing an initial bit error rate; Representing an initial eye height; The optical fiber transmission process comprises photoelectric conversion, modulation and demodulation processes, wherein the FPGA realizes modulation of optical signals through an embedded high-speed serial transceiver, the modulated optical signals are transmitted to an optical fiber link, and the optical signals are recovered at a receiving end through the demodulation process; The equalization optimization module is used for constructing a multi-dimensional weighted quality index MQM, wherein the MQM is obtained by integrating a normalized scoring function and a weight coefficient of BER, EH and EW, and the multi-dimensional weighted quality index is expressed as follows: ; Wherein, the Representing a multi-dimensional weighted quality index; Representing a bit error rate normalization scoring function; representing a eye height normalization scoring function; Representing an eye width normalization scoring function; 、 And Representing the weight coefficient; The method comprises the following steps of adjusting a transmission pre-emphasis coefficient and a receiving equalization parameter of a high-speed serial transceiver based on an improved particle swarm algorithm in an MQM driving FPGA, wherein the improved particle swarm algorithm comprises the following optimization processes: s2.1, initializing a pre-emphasis coefficient and a parameter range of a receiving equalization parameter, and randomly generating an initial particle group, wherein the position of each particle represents a group of parameter combinations; S2.2, calculating the MQM of each particle under the current parameters, estimating the gradient of the MQM to the parameters by adopting a finite difference method, and updating the current speed of the particle based on the gradient; S2.3, continuing iteration, judging convergence and outputting an optimal parameter combination if the MQM change rate is smaller than a threshold value; The optical signal control module extracts jitter characteristics of a received signal in an optical fiber link, wherein the jitter characteristics comprise jitter standard deviation and jitter peak value, clock data are dynamically adjusted according to the jitter characteristics and current BER to recover CDR loop bandwidth so as to balance jitter tracking capacity and noise suppression performance, recovery of the optical signal is accurately controlled by utilizing a nonlinear mapping rule, the nonlinear mapping rule is realized through a lookup table, the input of the lookup table is a joint quantization interval of a jitter indication factor and a bit error rate, the input of the lookup table is a preset CDR bandwidth level, the joint quantization interval is used for carrying out joint analysis on the jitter interval and the BER interval, the jitter interval comprises a low jitter interval, a middle jitter interval and a high jitter interval, the BER interval comprises a low error interval, a middle error interval and a high error interval, the jitter indication factor is obtained based on the jitter standard deviation and the jitter peak value, and the jitter indication factor is as follows: ; Wherein, the Representing a jitter indication factor; representing the jitter standard deviation; representing jitter peaks; And Representing the jitter indication factor influence coefficient; The signal transmission delay optimization module is used for carrying out filtering processing on BER data and calculating a trend slope, and dynamically switching an FEC mode by combining BER, the trend slope and preset hysteresis threshold logic to optimize the delay of signal transmission, wherein the trend slope is as follows: ; Wherein, the Representing a trend slope; representing the number of sampling points of a time window; representing a time stamp; Representation of A filtered BER value at a time; hysteresis threshold logic is defined as: when BER exceeds a threshold value and the trend slope is greater than 0, switching from a low error correction mode to a high error correction mode; switching from the high error correction mode to the low error correction mode when BER is less than the threshold and the trend slope is less than 0; The feedback module updates the baseline parameters in real time to form a closed loop and continuously adjusts the quality of the optical fiber link, wherein the updated baseline parameters are as follows ; Wherein the method comprises the steps of Representing the current bit error rate; Representing the current eye height; representing the current eye width.

Description

Method and system for realizing optical fiber interconnection transmission of high-speed serial transceiver based on FPGA Technical Field The invention relates to the technical field of optical fiber communication, in particular to a method and a system for realizing optical fiber interconnection transmission of a high-speed serial transceiver based on an FPGA. Background With the rapid development of optical fiber communication, the optical fiber interconnection technology becomes a core transmission scheme due to the high bandwidth, low time delay and anti-interference capability. However, in high-speed optical fiber transmission, signals are affected by factors such as optical fiber dispersion, temperature fluctuation, mechanical vibration and the like, so that problems such as rising of Bit Error Rate (BER), closing of an eye diagram, increasing of jitter and the like are caused, and the quality of an optical fiber communication link is easily limited due to the fact that fixed parameter balance is adopted in the traditional solution. Therefore, a method and a system for realizing optical fiber interconnection transmission by a high-speed serial transceiver based on an FPGA are provided. Disclosure of Invention The invention aims to provide a method and a system for realizing optical fiber interconnection transmission by a high-speed serial transceiver based on an FPGA, wherein the method and the system dynamically adjust the pre-emphasis coefficient and the receiving equalization parameter of the transceiver by monitoring the bit error rate BER and the eye pattern characteristic parameter of an optical fiber link in real time, constructing a link characteristic base line and combining a multidimensional weighting quality index to drive an improved particle swarm algorithm; extracting signal jitter characteristics, dynamically adjusting clock data recovery loop bandwidth based on the jitter characteristics and bit error rate, dynamically switching a forward error correction mode through hysteresis threshold logic, balancing time delay and error correction capability, and continuously updating baseline parameters through a closed loop feedback mechanism to realize real-time tuning of the quality of an optical fiber communication link. In order to achieve the above purpose, the present invention provides the following technical solutions: the method for realizing optical fiber interconnection transmission of the high-speed serial transceiver based on the FPGA comprises the following steps: s1, acquiring bit error rate BER and eye pattern characteristic parameters of a received signal in an optical fiber transmission process, wherein the eye pattern characteristic parameters comprise eye height EH and eye width EW; S2, constructing a multi-dimensional weighted quality index MQM, wherein the MQM is obtained by integrating normalized scoring functions and weight coefficients of BER, EH and EW; S3, extracting jitter characteristics of a received signal in an optical fiber link, wherein the jitter characteristics comprise jitter standard deviation and jitter peak value, dynamically adjusting clock data according to the jitter characteristics and current BER to recover CDR loop bandwidth so as to balance jitter tracking capacity and noise suppression performance, and accurately controlling recovery of the optical signal by utilizing a nonlinear mapping rule; s4, filtering the BER data, calculating a trend slope, dynamically switching the FEC mode by combining the BER, the trend slope and a preset hysteresis threshold logic, and optimizing the time delay of signal transmission; S5, updating the baseline parameters in real time to form a closed loop, and continuously performing optical fiber link quality adjustment. Preferably, the optical fiber transmission process includes photoelectric conversion, modulation and demodulation processes, the FPGA realizes modulation of an optical signal through an embedded high-speed serial transceiver, and transmits the modulated optical signal to an optical fiber link, and the optical signal is recovered at a receiving end through the demodulation process. Preferably, the modified particle swarm algorithm in S2 comprises the following steps: s2.1, initializing a pre-emphasis coefficient and a parameter range of a receiving equalization parameter, and randomly generating an initial particle group, wherein the position of each particle represents a group of parameter combinations; S2.2, calculating the MQM of each particle under the current parameters, estimating the gradient of the MQM to the parameters by adopting a finite difference method, and updating the current speed of the particle based on the gradient; s2.3, continuing iteration, and if the MQM change rate is smaller than the threshold value, judging convergence and outputting the optimal parameter combination. Preferably, the nonlinear mapping rule in S3 is implemented by a lookup table, where the input of the lookup