Pure analog physical impairment feature extractor

CN122533645APending Publication Date: 2026-08-07陈立波
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
陈立波
Filing Date
2026-05-18
Publication Date
2026-08-07

AI Technical Summary

Technical Problem

1. 模数转换器和数字信号处理器的功耗较高,限制了光模块的集成度和散热能力

Benefits of technology

1. 功耗显著降低:本发明采用纯模拟电路实现特征提取,无需模数转换器和数字信号处理器,系统功耗可降低60%至70%。本发明的特征提取器作为光模块信号处理链路的一部分,可有效降低整体系统功耗。

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The application discloses an optical module pure analog physical damage feature extractor and belongs to the technical field of optical communication. In view of the problems of high power consumption, large delay and poor safety of the digital signal processing architecture in the prior art, the application is composed of a pure analog circuit and does not contain any programmable logic device, and the physical damage features of the electrical signals converted from optical signals are directly extracted through the pure analog circuit. All the feature extraction is completed by the pure analog circuit, and the analog-to-digital converter is not needed to participate in the extraction and encoding process of the physical damage features, and the digital signal processing and software algorithm are not needed to participate, and any adaptive equalization algorithm based on real-time signal quality feedback is not contained. The application can simplify the optical module signal processing flow, reduce the system power consumption and improve the real-time performance and safety of signal processing.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of optical communication technology, and more specifically to a device for extracting signal physical damage features of optical modules. Background Technology

[0002] With the rapid development of data centers and 5G mobile communication technology, the transmission rate of optical modules is constantly increasing, and the requirements for signal processing are also becoming more stringent. Currently, mainstream optical module signal processing solutions all adopt a digital signal processor architecture, which converts optical signals into digital signals through an analog-to-digital converter, then uses digital signal processing algorithms to extract the physical impairment characteristics of the signal, and finally adjusts the equalizer parameters based on the extracted characteristics.

[0003] The existing technology has the following shortcomings: 1. The high power consumption of analog-to-digital converters and digital signal processors limits the integration and heat dissipation capabilities of optical modules. 2. Digital signal processing introduces a certain delay, affecting the real-time performance of signal transmission. 3. The software-configurable digital signal processing architecture has security vulnerabilities and is susceptible to malicious attacks. 4. The adaptive equalization algorithm based on real-time signal quality feedback is complex, increasing the design difficulty and cost of the system. Summary of the Invention

[0004] The purpose of this invention is to overcome the above-mentioned shortcomings of the prior art and provide a pure analog physical damage feature extractor that can directly extract the physical damage features of optical signals through analog circuits without the need for analog-to-digital converters and digital signal processing, thereby reducing system power consumption and improving the real-time performance and security of signal processing.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows: A pure analog physical damage feature extractor for optical modules is provided, which is composed of pure analog circuits and does not contain any programmable logic devices. It directly extracts the physical damage features of the electrical signal after the optical signal is converted by pure analog circuits. All feature extraction is completed by pure analog circuits, without the need for analog-to-digital converters to participate in the extraction and encoding process of physical damage features, without the need for digital signal processing, and without the participation of software algorithms. It does not contain any adaptive equalization algorithm based on real-time signal quality feedback.

[0006] Furthermore, the pure analog circuit is a non-programmable custom analog circuit that uses at least one of an analog filter bank, an envelope detector, and a phase comparator to extract physical damage features.

[0007] Furthermore, the physical damage characteristics include at least one of the following: pulse broadening caused by dispersion, amplitude distortion caused by nonlinear effects, and energy dispersion caused by polarization mode dispersion.

[0008] Furthermore, it also includes an analog feature encoder, which directly receives the analog voltage signal output by the analog feature extractor and directly encodes the analog voltage into a digital address signal for output; the output terminal of the analog feature encoder is equipped with a hardware hysteresis circuit, and the hysteresis threshold is fixed once by a one-time non-volatile memory.

[0009] Furthermore, it also includes an address stability counter, which only allows the address of the subsequent circuit to be updated when the characteristic value changes beyond the hysteresis threshold and the address stability exceeds the number of clock cycles that the one-time non-volatile memory is fixed.

[0010] Furthermore, it also includes a temperature coefficient compensation circuit. The hardware looks up the temperature coefficient table stored in the one-time non-volatile memory based on the temperature sensor data on the chip, and automatically adjusts the reference voltage or bias current of the feature extractor according to the linear compensation formula.

[0011] Furthermore, it also includes an on-chip self-calibration circuit, where the hardware automatically generates a set of standard test signals with known characteristics, sends them to an analog feature extractor, compares the output feature value with the expected value stored in a one-time non-volatile memory, and detects whether there is a systematic offset; when an offset is detected, it automatically calculates a temporary compensation coefficient, superimposed on the stored parameters, and does not save it after power failure.

[0012] Furthermore, it also includes a purely hardware-based automatic gain control circuit at the receiving front end, which monitors the amplitude of the input signal in real time using an analog envelope detector, compares it with a target amplitude threshold stored in a one-time non-volatile memory, and controls the gain of the variable gain amplifier through a purely hardware feedback loop.

[0013] Furthermore, it also includes a pure analog signal loss detection circuit and a DC offset compensation circuit, the compensation value of which is permanently stored in a one-time non-volatile memory; the signal loss detection circuit directly monitors the voltage corresponding to the input optical power by an analog comparator and compares it with the fixed alarm threshold; the DC offset compensation circuit automatically detects the offset under the condition of no light input and stores it in the fixed register, and automatically deducts it during normal operation.

[0014] Furthermore, it also includes a multi-channel consistency calibration controller, which measures the gain difference, skew, and wavelength shift of each channel using a known training sequence before leaving the factory, and then stores the compensation coefficients in a one-time non-volatile memory; the hardware automatically loads the compensation coefficients during normal operation. Beneficial effects

[0015] Compared with the prior art, the present invention has the following beneficial effects: 1. Significantly Reduced Power Consumption: This invention uses pure analog circuitry for feature extraction, eliminating the need for analog-to-digital converters and digital signal processors, thus reducing system power consumption by 60% to 70%. The feature extractor of this invention, as part of the optical module signal processing link, effectively reduces overall system power consumption.

[0016] 2. Significantly reduced latency: The signal processing latency of purely analog circuits is in the nanosecond range, which is more than an order of magnitude smaller than that of digital signal processing schemes. 3. Significantly enhanced security: This invention does not contain any programmable logic or software code, fundamentally eliminating the risk of software attacks. 4. Simple and reliable structure: Pure analog circuits have a simple structure, high integration, and good reliability, and can adapt to various harsh working environments. 5. Lower cost: Eliminating the need for high-performance analog-to-digital converters and digital signal processors effectively reduces the manufacturing cost of optical modules. Detailed Implementation

[0017] The present invention will be further described in detail below with reference to specific embodiments.

[0018] This embodiment provides a purely simulated physical damage feature extractor for a 100G optical module, comprising the following components: 1. Receiving front-end circuit: including a photodetector, transimpedance amplifier, and variable gain amplifier, used to convert optical signals into electrical signals and perform preliminary amplification. 2. Analog Feature Extraction Circuit: Composed of an analog filter bank, an envelope detector array, and a phase comparator array. The analog filter bank includes low-pass, band-pass, and high-pass filters, used to separate signals with different frequency components. - Envelope detector arrays are used to detect amplitude changes in signals and extract amplitude distortion features caused by nonlinear effects. - A phase comparator array is used to detect phase changes in a signal and extract pulse broadening due to dispersion and energy dispersion characteristics due to polarization mode dispersion. 3. Analog Feature Encoder: Converts the analog voltage signal output from the analog feature extraction circuit into an 8-bit digital address signal. 4. Hardware hysteresis circuit: installed at the output of the analog feature encoder, with a hysteresis threshold of 1 / 4 of the voltage difference between adjacent addresses. 5. Address Stabilization Counter: Set to allow subsequent circuits to update their addresses only after the address has stabilized for more than 10 clock cycles. 6. Temperature Coefficient Compensation Circuit: Based on the output of the on-chip temperature sensor, the corresponding temperature compensation coefficient is read from the one-time non-volatile memory, and the bias current of the analog feature extraction circuit is adjusted accordingly. 7. On-chip self-calibration loop: Generates standard test signals periodically upon system power-on and during operation to calibrate the analog feature extraction circuit. 8. Automatic Gain Control Circuit: Composed of an analog envelope detector and a pure hardware feedback loop, it controls the gain of the variable gain amplifier to keep the output signal amplitude within a suitable range. 9. Signal loss detection circuit: Composed of an analog comparator, it generates an alarm signal when the input optical power is lower than a preset threshold. 10. DC Offset Compensation Circuit: Detects the DC offset of each channel under conditions of no light input and automatically subtracts it during normal operation. 11. Multi-channel consistency calibration controller: The gain, skew, and wavelength offset of each channel are calibrated before shipment, and the compensation coefficients are stored in a one-time non-volatile memory. The working process of this embodiment is as follows: 1. The photodetector converts the received optical signal into an electrical signal, which is then amplified by a transimpedance amplifier and a variable gain amplifier before being sent to the analog feature extraction circuit. 2. The analog feature extraction circuit directly extracts the dispersion broadening, nonlinear distortion amplitude, and polarization mode dispersion energy dispersion characteristics of the signal through an analog filter bank, envelope detector array, and phase comparator array. 3. The analog feature encoder converts the extracted analog feature values ​​into digital address signals for output. 4. Hardware hysteresis circuits and address stabilization counters process the address signal to prevent address jitter. 5. The temperature coefficient compensation circuit adjusts the parameters of the analog feature extraction circuit according to the current temperature to compensate for the impact of temperature changes on circuit performance. 6. The on-chip self-calibration loop periodically calibrates the analog feature extraction circuit to compensate for the effects of process variations and device aging. 7. The automatic gain control circuit adjusts the gain of the variable gain amplifier according to the input signal amplitude to ensure signal amplitude stability. 8. The signal loss detection circuit monitors the input optical power in real time and generates an alarm signal when the optical power is too low. 9. The DC offset compensation circuit automatically subtracts the DC offset of each channel, improving signal quality. 10. The multi-channel consistency calibration controller applies the factory-fixed compensation coefficients to ensure consistent performance across all channels. The pure analog physical damage feature extractor in this embodiment does not require an analog-to-digital converter or digital signal processing. It can directly extract the physical damage features of optical signals and has the advantages of low power consumption, low latency, high security, simple and reliable structure. It is suitable for various high-speed optical modules.

Claims

1. A pure simulation physical damage feature extractor for optical modules, characterized in that, It is composed of pure analog circuits and does not contain any programmable logic devices; it directly extracts the physical damage characteristics of the electrical signal after the optical signal is converted through pure analog circuits. All feature extraction is performed by pure analog circuits, without the need for analog-to-digital converters in the extraction and encoding of physical damage features, digital signal processing, or software algorithms; it does not include any adaptive equalization algorithms based on real-time signal quality feedback.

2. The feature extractor according to claim 1, characterized in that, The pure analog circuit is a non-programmable custom analog circuit that uses at least one of an analog filter bank, an envelope detector, and a phase comparator to extract physical damage features.

3. The feature extractor according to claim 1, characterized in that, The physical damage characteristics include at least one of the following: pulse broadening due to dispersion, amplitude distortion due to nonlinear effects, and energy dispersion due to polarization mode dispersion.

4. The feature extractor according to claim 1, characterized in that, It also includes an analog feature encoder, which directly receives the analog voltage signal output by the analog feature extractor and directly encodes the analog voltage into a digital address signal for output; the output terminal of the analog feature encoder is equipped with a hardware hysteresis circuit, and the hysteresis threshold is fixed once by a one-time non-volatile memory.

5. The feature extractor according to claim 4, characterized in that, It also includes an address stability counter, which allows the address of the subsequent circuit to be updated only when the characteristic value changes beyond the hysteresis threshold and the address stability exceeds the number of clock cycles that the one-time non-volatile memory is fixed.

6. The feature extractor according to claim 1, characterized in that, It also includes a temperature coefficient compensation circuit. The hardware looks up the temperature coefficient table stored in the one-time non-volatile memory based on the temperature sensor data on the chip, and automatically adjusts the reference voltage or bias current of the feature extractor according to the linear compensation formula.

7. The feature extractor according to claim 1, characterized in that, It also includes an on-chip self-calibration circuit, where the hardware automatically generates a set of standard test signals with known characteristics, sends them to an analog feature extractor, compares the output feature value with the expected value stored in a one-time non-volatile memory, and detects whether there is a systematic offset. When an offset is detected, a temporary compensation coefficient is automatically calculated and superimposed on the stored parameters. This coefficient is not saved after power failure.

8. The feature extractor according to claim 1, characterized in that, It also includes a pure hardware automatic gain control circuit at the receiving front end, which monitors the amplitude of the input signal in real time by an analog envelope detector, compares it with the target amplitude threshold fixed in a one-time non-volatile memory, and controls the gain of the variable gain amplifier through a pure hardware feedback loop.

9. The feature extractor according to claim 1, characterized in that, It also includes a pure analog signal loss detection circuit and a DC offset compensation circuit, the compensation value of which is permanently stored in a one-time non-volatile memory; the signal loss detection circuit directly monitors the voltage corresponding to the input optical power by an analog comparator and compares it with the fixed alarm threshold; the DC offset compensation circuit automatically detects the offset under the condition of no light input and stores it in the fixed register, and automatically deducts it during normal operation.

10. The feature extractor according to any one of claims 1 to 9, characterized in that, It also includes a multi-channel consistency calibration controller, which measures the gain difference, skewness, and wavelength shift of each channel using a known training sequence before leaving the factory, and then stores the compensation coefficients in a one-time non-volatile memory; the hardware automatically loads the compensation coefficients during normal operation.