Parallel random number generation system and method based on hybrid integrated chaotic microcombs

Abstract

The application relates to the fields of optoelectronics and information security, and particularly relates to a parallel random number generation system and method based on a hybrid integrated chaotic micro comb, which comprises a hybrid integrated chaotic entropy source module, a channel selection module, an optoelectronic detection module and a post-processing unit; the hybrid integrated chaotic entropy source module comprises a hybrid integrated DFB laser chip and an on-chip silicon nitride micro ring resonator; the DFB laser chip directly provides pump light to the on-chip silicon nitride micro ring resonator through a coupling structure to output chaotic micro cavity optical comb; the output wavelength of the DFB laser chip is set by a fixed driving current, and the resonant wavelength of the on-chip silicon nitride micro ring resonator is set by a fixed temperature. The application self-starts to generate chaotic micro cavity optical comb without dynamic scanning of the pump frequency and external modulation, and realizes a chaotic entropy source generation scheme with high integration, low complexity and low power consumption.

Description

Technical Field

[0001] This invention relates to the fields of optoelectronics and information security technology, and in particular to a parallel random number generation system and method based on a hybrid integrated chaotic microcomb, which can automatically generate a chaotic microcomb and generate a parallel random bit sequence without the need for an erbium-doped fiber amplifier (EDFA) and without pump frequency scanning. Background Technology

[0002] Random numbers have wide and significant applications in information security, encrypted communication, key distribution, Monte Carlo simulation, reinforcement learning, and other fields. Physical random number generators utilize the inherent randomness of physical processes to generate truly random numbers, offering advantages over pseudo-random number generators in terms of unpredictability and non-repeatability. With the development of high-speed communication, parallel computing, and large-scale information security systems, existing random number generation technologies are placing higher demands on random number generation rate, number of channels, statistical quality, and system integration.

[0003] Existing random number generation schemes based on chaotic microcombs typically require an external tunable laser as the pump source, and the excitation and control of the chaotic microcomb are achieved through optical amplifiers, polarization controllers, microring resonators, and external modulation or dispersion broadening structures. In practical operation, some schemes require scanning the pump light frequency to induce a chaotic state after the pump light crosses a specific detuning interval; others require external phase modulation, wide dispersion spectrum, or complex feedback structures to extend the spectral bandwidth and increase the random number generation rate. While these methods can achieve high-speed physical random number generation, the systems generally suffer from long device links, high insertion loss, reliance on dynamic adjustment during startup, high parameter stability requirements, high engineering complexity, and low integration levels, making it difficult to achieve a low-cost, miniaturized, and stable self-starting random number generation system after power-on.

[0004] Furthermore, the effective RF spectrum bandwidth obtained after photodetection of a single chaotic comb tooth is typically limited by detector bandwidth, chaos intensity, sampling rate, and signal correlation. Increasing the sampling rate alone to improve the random bit generation rate may lead to strong correlations between adjacent sampling points, thus affecting the statistical quality of the random numbers. Conversely, using complex post-processing algorithms to reduce correlation may increase computational resource consumption and system latency. Therefore, current technologies still struggle to achieve a balance between high speed, multiple channels, high statistical quality, and low complexity.

[0005] Therefore, there is a need for a highly integrated, low-complexity, and stable chaotic entropy source generation scheme, as well as a random number extraction method that can reduce statistical bias, weaken sampling correlation, and retain high-entropy valid bits, in order to meet the application requirements of high-speed, high-quality, multi-channel physical random number generation. Summary of the Invention

[0006] The purpose of this invention is to provide a parallel random number generation system and method based on a hybrid integrated chaotic microcomb, in order to solve the problems that existing parallel random number generation schemes based on chaotic microcombs usually require scanning the amplified pump light frequency, or expanding the chaotic spectrum bandwidth through external modulation, dispersion broadening, etc., resulting in relatively complex system structures, low integration levels, and difficulty in achieving fixed-parameter self-starting generation of chaotic microcombs.

[0007] To achieve the above objectives, the present invention adopts the following technical solution:

[0008] A parallel random number generation system based on a hybrid integrated chaotic microcomb includes: a hybrid integrated chaotic entropy source module, a channel selection module, a photoelectric detection module, and a post-processing unit;

[0009] The hybrid integrated chaotic entropy source module includes a hybrid integrated packaged DFB laser chip and an on-chip silicon nitride microring resonator.

[0010] The DFB laser chip provides pump light directly to the on-chip silicon nitride microring resonator through an end-face coupling structure, and outputs a chaotic microcavity optical comb; the output wavelength of the DFB laser chip is set by a fixed driving current, and the resonant wavelength of the on-chip silicon nitride microring resonator is set by a fixed temperature.

[0011] The channel selection module is used to select multiple relatively pump-asymmetric comb teeth from the chaotic microcavity optical comb as parallel random number generation channels.

[0012] The photoelectric detection module is used to convert the time-domain chaotic optical signals of each channel into corresponding electrical signals.

[0013] The post-processing unit is used to sample, quantize, and digitally post-process the electrical signal to generate a random bit sequence.

[0014] Furthermore, the DFB laser chip is optically coupled to the silicon nitride chip input bus waveguide through a coupling structure; the input bus waveguide then couples the pump light into the microring through the coupling region between it and the silicon nitride microring resonant cavity.

[0015] Furthermore, the channel selection module is a wavelength selection switch or an adjustable filter, and its output is directly connected to the input of the photoelectric detection module.

[0016] Furthermore, the photoelectric detection module includes at least one photoelectric detector, and the sampling and quantization unit in the post-processing unit includes a high-speed analog-to-digital converter or an oscilloscope to sample the analog electrical signal at a preset sampling rate and convert each sampling point into a corresponding digital amplitude value.

[0017] The parallel random number generation method based on the above system includes the following steps:

[0018] S1. The DFB laser chip and the on-chip micro-ring resonator are integrated and packaged together. The temperature of the on-chip micro-ring resonator is set by the temperature controller and the driving current of the DFB laser chip is set by the driving current control unit. The laser wavelength output by the DFB laser chip and the resonant wavelength of the on-chip micro-ring resonator form a preset non-zero detuning amount and fall into the chaotic working window.

[0019] S2. After the system is powered on and thermally stabilized, it automatically generates a chaotic microcavity optical comb under fixed drive current and fixed temperature conditions.

[0020] S3. Select multiple relatively pump-asymmetric comb teeth from the chaotic microcavity optical comb as independent parallel random number generation channels;

[0021] S4. The chaotic optical signals of each channel are converted into analog electrical signals by a photodetector, and the analog electrical signals are input into the sampling and quantization unit for digitization.

[0022] S5. Post-process the digital amplitude values ​​of each channel to generate a random bit sequence.

[0023] Furthermore, the post-processing in S5 specifically includes: performing fourth-order differential processing on the digital amplitude value obtained after analog-to-digital conversion, normalizing the amplitude based on the maximum voltage amplitude of the differential sequence, dividing the normalized amplitude interval into 256 quantization intervals for eight-bit binary encoding, reversing the high and low bits of the encoded bit stream, and dividing the reversed bit stream into two segments and performing segmented delay XOR operations.

[0024] The beneficial effects of this invention are as follows:

[0025] (1) The system of the present invention achieves the fixed-parameter self-starting generation of chaotic optical combs by hybrid integration of DFB laser chip and silicon nitride micro-ring resonator, and by establishing a preset non-zero detuning window with fixed driving current and fixed temperature. It does not require scanning pump light frequency, external modulation or setting EDFA in pump link, which greatly simplifies the system structure and improves the degree of integration. Based on the large-scale parallelization capability of chaotic micro-comb, a single entropy source can provide hundreds of potential chaotic comb tooth channels at the same time. In one embodiment, about 340 independent optical comb teeth can be generated in the chaotic spectrum range of more than 100 nanometers, providing a channel basis for large-scale parallel random number generation. The chaotic signals between each comb tooth channel have low cross-correlation, ensuring the independence of the random number sequence generated by each channel.

[0026] (2) The method of the present invention effectively reduces the statistical bias and adjacent sampling correlation of the original chaotic amplitude sequence through the post-processing process of fourth-order difference, amplitude normalization, 256-level amplitude interval encoding, bit stream inversion, segmented delay XOR and high-entropy effective bit extraction; in the embodiment with a sampling rate of 100GSa / s and retaining 5 high-entropy bits, high-speed random bit generation of 500Gbps per channel and 8Tbps total rate of 16 channels can be achieved. Attached Figure Description

[0027] Figure 1 This is a schematic diagram of the hybrid integrated chaotic entropy source module provided by the present invention;

[0028] Figure 2 This is a schematic diagram of the chaotic time signal of one of the channels provided by the present invention;

[0029] Figure 3 This is a schematic diagram of the autocorrelation function of the chaotic time signal provided by the present invention;

[0030] Figure 4 This is a schematic diagram of the cross-correlation function of chaotic signals between channels provided by the present invention;

[0031] Figure 5 This is a schematic diagram of the random number generation principle provided by the present invention;

[0032] Figure 6 This is a schematic diagram summarizing various randomness indicators based on the NIST SP800-22 test provided by the present invention. Detailed Implementation

[0033] The technical solution of the present invention will be further described below with reference to the accompanying drawings and specific embodiments.

[0034] This invention provides a self-starting parallel random number generation system based on a hybrid integrated chaotic microcomb, comprising: a hybrid integrated chaotic entropy source module, a channel selection module, a photoelectric detection module, and a post-processing unit; the hybrid integrated chaotic entropy source module includes a hybrid integrated packaged DFB laser chip and an on-chip silicon nitride microring resonator, the on-chip silicon nitride microring resonator being fabricated from CMOS process-compatible silicon nitride material; wherein, the DFB laser chip directly provides pump light to the on-chip silicon nitride microring resonator through a hybrid integrated low-loss coupling structure, without an EDFA or optical power amplifier between them; the output wavelength of the DFB laser chip is set by a fixed drive current, and the resonant wavelength of the on-chip silicon nitride microring resonator is set by a fixed temperature. A preset non-zero detuning value is established between the DFB laser output wavelength and the microring resonant wavelength by fixing the driving current and temperature. When the preset non-zero detuning value falls within the chaotic working window, the DFB pump light enters the high-quality factor microring resonant cavity via low-loss hybrid integrated coupling. Relying on the cavity power enhancement to excite the Kerr nonlinear effect, a chaotic microcavity optical comb is automatically generated without the need for EDFA, dynamic scanning of the pump frequency, or external modulation. Multiple comb teeth in the chaotic microcavity optical comb can serve as parallel channels, each carrying a time-domain chaotic optical signal. The channel selection module is connected to the hybrid integrated chaotic entropy source module and is used to select multiple comb teeth that are relatively asymmetric to the pump from the chaotic microcavity optical comb as parallel random number generation channels.

[0035] like Figure 1 As shown, the module consists of a DFB laser chip, a hybrid integrated low-loss coupling structure, a silicon nitride microring resonator, a drive current control unit, and a temperature control unit. No EDFA or pump amplifier is placed between the DFB laser chip and the silicon nitride microring resonator. A preset non-zero detuning is established by fixing current and temperature parameters, enabling the self-starting generation of the chaotic optical comb.

[0036] The hybrid integrated low-loss coupling structure specifically includes mode switching, end-face coupling, edge coupling, and reverse tapered waveguide, etc.

[0037] The DFB laser chip is optically coupled to the silicon nitride chip's input bus waveguide via a hybrid integrated low-loss coupling structure. The input bus waveguide then couples the pump light into the silicon nitride microring resonator through a coupling region between the waveguide and the microring. No EDFA or pump amplifier is used in the pump link between the DFB and the microring.

[0038] The DFB laser chip and the on-chip silicon nitride microring resonator are packaged and disposed in the same thermal control platform or standard package using a hybrid integration method.

[0039] This invention establishes a preset non-zero detuning between the DFB laser output wavelength and the microring resonant wavelength through a fixed driving current and temperature matching relationship. This preset non-zero detuning differs from the method of temporarily crossing the resonant interval by continuously scanning the pump laser frequency during startup; in this invention, after the system is powered on and thermally stabilized, the fixed operating point falls within the chaotic operating window. The pump light achieves intracavity enhancement within the high-quality factor microring, exciting Kerr nonlinear processes such as four-wave mixing and modulation instability, thereby forming a chaotic microcavity optical comb. The system requires no external frequency scanning, additional modulation, or EDFA amplification, thus reducing system complexity and improving integration.

[0040] In one specific implementation, the DFB laser chip and the silicon nitride microring resonator are hybridized and integrated, then packaged in a standard butterfly-shaped package. The system regulates the temperature of the chip and the microring through a temperature controller and sets the DFB drive current to achieve stable chaotic microcomb output within a preset non-zero detuning window; the typical operating point is set to a microring temperature of 28°C and a drive current of 215mA. It should be noted that 215mA and 28°C are operating point parameters in a specific embodiment. The focus of this invention is on establishing a preset non-zero detuning amount through the coordinated use of drive current and temperature to achieve self-starting with fixed parameters, and is not limited to the specific values ​​mentioned above. After the system is powered on and thermally stabilized, it can self-start into a chaotic state without frequency scanning, external modulation, or EDFA, and can maintain stable output continuously without further adjustment of the drive current and temperature.

[0041] In one embodiment, the chaotic microcomb forms approximately 340 spectral comb teeth in a wavelength range exceeding 100 nanometers, providing abundant candidate channels for multi-channel parallel random number generation. A wavelength selection switch selects 16 relatively pumped asymmetric comb tooth channels, and a photodetector converts the optical signals of each channel into analog electrical signals and inputs them into a sampling quantization unit. Combined with 100 GSa / s sampling, 8-bit analog-to-digital conversion, fourth-order differential, amplitude normalization, 256-level amplitude interval encoding, bitstream inversion, segmented delay XOR, and 5-bit high-entropy bit extraction, parallel random bit output with a total rate of 8 Tbps across 16 channels can be achieved.

[0042] The chaotic microcomb comprises multiple spectral comb teeth, at least some of which can function as independent parallel entropy source channels. The channel selection module can employ devices such as wavelength-selective switches, tunable filters, or arrayed waveguide gratings to select multiple comb teeth of a specified wavelength from the chaotic microcomb. To reduce inter-channel correlation, comb teeth at asymmetric positions relative to the pump wavelength are preferably selected as random number generation channels. The selected comb teeth are converted into corresponding chaotic electrical signals after photoelectric detection and then sent to subsequent sampling and processing units.

[0043] In one embodiment, a chaotic optical comb output amplifier can be set after the hybrid integrated self-starting entropy source module to amplify the generated chaotic microcavity optical comb or selected comb tooth signal, so as to improve the signal-to-noise ratio of subsequent channel selection and photoelectric detection; the amplifier is not located between the DFB laser chip and the silicon nitride microring resonant cavity, and does not participate in the generation process of the chaotic microcavity optical comb.

[0044] like Figure 2 The diagram shows a chaotic time signal from one of the channels provided by this invention. The time-domain signal of a single channel exhibits rapid fluctuations, irregular amplitude changes, and noise-like characteristics, indicating that the channel is in a chaotic state. This type of chaotic time-domain signal originates from nonlinear effects and mode interactions within the microcavity; its amplitude fluctuations are difficult to predict and can serve as an entropy source for generating physical random numbers. By selecting multiple comb channels, multiple independent chaotic time sequences can be obtained simultaneously from the same chaotic microcomb entropy source.

[0045] like Figure 3 As shown, the autocorrelation function of the chaotic time series only exhibits a sharp main peak near zero delay, which then rapidly decays to near zero, without any obvious periodic side peaks. This result indicates that the chaotic time series lacks a stable periodic structure and possesses strong randomness and low temporal correlation. The closer the autocorrelation characteristics are to the ideal impulse pattern, the more abundant the non-periodic entropy source components in the signal that can be used for random number generation.

[0046] like Figure 4 As shown, the cross-correlation coefficients between different comb channels fluctuate slightly around zero, with no significant correlation peaks, indicating that the chaotic signals in each channel have good independence. Low cross-correlation is an important prerequisite for parallel random number generation, ensuring that the random bit sequences output by different channels are independent and avoiding a decrease in random number quality due to correlation between channels.

[0047] This invention also provides a self-starting parallel random number generation method based on a hybrid integrated chaotic microcomb, comprising the following steps:

[0048] S1. The DFB laser chip and the on-chip micro-ring resonator are integrated and packaged together. The temperature of the on-chip micro-ring resonator is set by the temperature controller and the driving current of the DFB laser chip is set by the driving current control unit. The laser wavelength output by the DFB laser chip and the resonant wavelength of the on-chip micro-ring resonator form a preset non-zero detuning amount and fall into the chaotic working window.

[0049] S2. After the system is powered on and thermally stabilized, under fixed drive current and fixed temperature conditions, it generates a chaotic microcavity optical comb without dynamic scanning of pump frequency, external modulation and optical power amplifier.

[0050] S3. Select multiple relatively pump-asymmetric comb teeth from the chaotic microcavity optical comb as independent parallel random number generation channels;

[0051] S4. The chaotic optical signals of each channel are converted into analog electrical signals by a photodetector, and the analog electrical signals are input into the sampling and quantization unit for digitization.

[0052] S5. Post-process the digital amplitude values ​​of each channel to generate a random bit sequence.

[0053] like Figure 5 As shown, the random number generation chain of the system is as follows: chaotic optical comb entropy source output → channel selection → photoelectric detection conversion → analog-to-digital sampling and quantization → digital post-processing → random bit stream output. After obtaining the chaotic electrical signals of each channel, the analog electrical signals output by the photodetector are input to the sampling and quantization unit, which can be an oscilloscope or a high-speed analog-to-digital converter circuit. An 8-bit analog-to-digital converter discretely samples the analog electrical signals at a preset sampling rate and converts the amplitude of each sampling point into a digital amplitude value. In one embodiment, the sampling rate is 100 GSa / s, and each sampling point is represented by an 8-bit analog-to-digital converter, so the original amplitude data corresponds to 256 distinguishable amplitude levels. Since the amplitude probability distribution of the original chaotic time signal may be skewed, directly extracting the bit sequence may not meet the requirements of randomness detection. Therefore, it is necessary to post-process the quantized digital amplitude values ​​to reduce statistical bias and improve the quality of the output random bits.

[0054] The post-processing first includes fourth-order difference processing. Specifically, the digital amplitude value sequence output by the oscilloscope is subjected to fourth-order difference to obtain a difference amplitude sequence. Fourth-order difference can reduce the impact of the original amplitude distribution imbalance and low-frequency trend on the quality of random bits, and reduce the correlation between adjacent sampling points.

[0055] The post-processing also includes amplitude normalization, 8-bit binary encoding, bitstream inversion, and segmented delay XOR processing. Specifically, the amplitude of each sampling point is normalized according to the maximum voltage amplitude of the fourth-order differential sequence, ensuring it falls within a unified normalized amplitude range. Using an 8-bit analog-to-digital converter, the normalized amplitude range is divided into 256 quantization intervals, and an 8-bit binary number is generated corresponding to the interval where the normalized amplitude of each sampling point falls. Then, the high and low bit order of the 8-bit binary numbers corresponding to each sampling point is reversed to form a new bitstream. This new bitstream is then divided into two segments, with one segment used as a delayed bitstream and XORed with the other segment. This processing further breaks down the fixed patterns in amplitude encoding, reduces statistical bias, and enhances the randomness of the bit sequence. In one embodiment, the acquired time series can be divided into two segments, with the latter segment having a fixed time delay relative to the former, and the corresponding bitstreams of the two segments are then XORed.

[0056] The post-processing also includes the extraction of high-entropy significant bits. Specifically, the entropy value of the segmented delayed XOR bit sequence is evaluated, and a preset number of high-entropy significant bits are retained while low-entropy bits are discarded. In one embodiment, 5 high-entropy significant bits can be retained from the 8-bit data corresponding to each sampling point. This method can improve the single-channel random bit output rate while ensuring the quality of randomness; when the sampling rate is 100 GSa / s and 5 high-entropy bits are retained for each sampling point, the single-channel random bit output rate can reach 500 Gbps.

[0057] After the above processing, each comb channel can output one high-speed random bit sequence. Since the chaotic time signals of different comb channels have low autocorrelation and low cross-correlation characteristics, multiple channels can work in parallel to generate multiple random bits simultaneously. In one embodiment, if 16 parallel channels are used, the sampling rate is 100 GSa / s, and each sampling point retains 5 valid high-entropy bits, the total output rate is 16 × 100 GSa / s × 5 bits = 8 Tbps.

[0058] To further verify the quality of the random numbers, standard randomness test suites such as NIST SP800-22 can be used to statistically test the generated bit sequences. Test metrics may include frequency test, intra-block frequency test, run test, longest run test, discrete Fourier transform test, non-overlapping template test, approximate entropy test, random walk test, random walk variant test, sequence test, and linear complexity test. In one embodiment, 1000 1Mbit bitstream samples were used for testing at a significance level of 0.01. The original sequence, without post-processing, failed the NIST SP800-22 randomness test. After fourth-order differencing, amplitude normalization, 256-level amplitude interval encoding, bitstream inversion, segmented delay XOR, and high-entropy bit extraction, the p-values ​​for all tests were greater than 0.0001, and the sample pass rate was within the range of 0.99 ± 0.0094392, indicating that the generated random bit sequence has good statistical randomness.

[0059] Figure 6 This paper presents a summary of the results for various metrics based on the NIST SP800-22 test, including p-values ​​and pass rates for frequency count test, intra-block frequency test, runs test, longest runs test, discrete Fourier transform test, non-overlapping template test, approximate entropy test, random walk test, random walk variant test, sequence test, and linear complexity test. Figure 6 As can be seen, the randomness criterion of the output bit sequence of a randomly selected channel is met in all the above indicators, which further verifies the effectiveness of the present invention in achieving self-starting parallel random number generation without EDFA or frequency scanning.

[0060] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A parallel random number generation system based on a hybrid integrated chaotic microcomb, characterized in that, include: The system integrates a chaotic entropy source module, a channel selection module, a photoelectric detection module, and a post-processing unit. The hybrid integrated chaotic entropy source module includes a hybrid integrated packaged DFB laser chip and an on-chip silicon nitride microring resonator; the DFB laser chip directly provides pump light to the on-chip silicon nitride microring resonator through a coupling structure, and outputs a chaotic microcavity optical comb; the output wavelength of the DFB laser chip is set by a fixed driving current, and the resonant wavelength of the on-chip silicon nitride microring resonator is set by a fixed temperature. The channel selection module is used to select multiple relatively pump-asymmetric comb teeth from the chaotic microcavity optical comb as parallel random number generation channels. The photoelectric detection module is used to convert the time-domain chaotic optical signals of each channel into corresponding electrical signals. The post-processing unit is used to sample, quantize, and digitally post-process the electrical signal to generate a random bit sequence.

2. The system according to claim 1, characterized in that, The DFB laser chip is optically coupled to the silicon nitride chip input bus waveguide through a coupling structure; the input bus waveguide then couples the pump light into the microring through the coupling region between the input bus waveguide and the silicon nitride microring resonant cavity.

3. The system according to claim 1, characterized in that, The channel selection module is a wavelength selection switch or an adjustable filter, and its output is directly connected to the input of the photoelectric detection module.

4. The system according to claim 1, characterized in that, The photoelectric detection module includes at least one photoelectric detector, and the sampling and quantization unit in the post-processing unit includes a high-speed analog-to-digital converter or an oscilloscope to sample the analog electrical signal at a preset sampling rate and convert each sampling point into a corresponding digital amplitude value.

5. A parallel random number generation method based on the system of any one of claims 1 to 4, characterized in that, Includes the following steps: S1. The DFB laser chip and the on-chip micro-ring resonator are integrated and packaged together. The temperature of the on-chip micro-ring resonator is set by the temperature controller and the driving current of the DFB laser chip is set by the driving current control unit. The laser wavelength output by the DFB laser chip and the resonant wavelength of the on-chip micro-ring resonator form a preset non-zero detuning amount and fall into the chaotic working window. S2. After the system is powered on and thermally stabilized, it automatically generates a chaotic microcavity optical comb under fixed drive current and fixed temperature conditions. S3. Select multiple relatively pump-asymmetric comb teeth from the chaotic microcavity optical comb as independent parallel random number generation channels; S4. The chaotic optical signals of each channel are converted into analog electrical signals by a photodetector, and the analog electrical signals are input into the sampling and quantization unit for digitization. S5. Post-process the digital amplitude values ​​of each channel to generate a random bit sequence.

6. The parallel random number generation method according to claim 5, characterized in that, The post-processing in S5 specifically includes: performing fourth-order differential processing on the digital amplitude value obtained after analog-to-digital conversion, normalizing the amplitude based on the maximum voltage amplitude of the differential sequence, dividing the normalized amplitude interval into 256 quantization intervals for eight-bit binary encoding, reversing the high and low bits of the encoded bit stream, dividing the reversed bit stream into two segments and performing segmented delay XOR operations.

Images