A dual-frequency block maximum a posteriori probability demodulation method for continuous phase BPSK in mud

By employing a dual-frequency block maximum a posteriori (MAP) demodulation method based on continuous phase BPSK in mud, and utilizing transition frequency information and MAP decoding, the high bit error rate problem in existing technologies is solved, thereby reducing the bit error rate and improving communication performance.

CN123123087APending Publication Date: 2026-11-17CHINA PETROLEUM & CHEMICAL CORP +2
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611275467.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-08-21
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

Existing continuous phase BPSK signal demodulation techniques for mud fail to effectively utilize the redundant information in the transition frequency range and lack an effective joint decoding mechanism, resulting in limited bit error rate performance.

Method used

A dual-frequency block maximum a posteriori (MAP) demodulation method based on continuous phase BPSK in mud is adopted. Through information fusion and MAP decoding, the information of the transition signal frequency is utilized, combined with a recursive least squares decision feedback equalizer and a BCJR decoder to achieve fusion modeling and demodulation of data code and phase modulation code.

Benefits of technology

It effectively reduces the bit error rate of communication, improves the performance of the mud pressure phase shift keying communication system, and can be applied to existing systems without changing the hardware.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123123087A_ABST
    Figure CN123123087A_ABST
Patent Text Reader

Abstract

This invention discloses a dual-frequency block maximum a posteriori probability demodulation method for continuous phase BPSK in mud. This method utilizes the redundant information in the transition portion of the mud pulse phase shift keying signal, and simultaneously demodulates the data and transition portions through a dual-frequency block processing mechanism. The dual-frequency signals are separated by a bandpass filter, and a triple parallel RLS-DFE equalizer is used to extract the soft information of the data code after down-conversion at the data frequency, the phase-modulated code after down-conversion at the data frequency, and the phase-modulated code after down-conversion at the transition frequency, respectively. The data code and phase-modulated code are fused to form a convolutional coding structure, and an RI-BCJR decoding algorithm based on the maximum a posteriori probability criterion is used for information fusion. Joint processing of the received sequence through a state transition model significantly improves the symbol decision accuracy. This invention can reduce the bit error rate of mud pressure phase shift keying modulation.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of mud pulse telemetry, and in particular to a dual-frequency block maximum a posteriori probability demodulation method for continuous phase BPSK (phase shift keying modulation) of mud. Background Technology

[0002] Mud pulse transmission technology is a widely used wireless communication technology in logging while drilling. It utilizes mud as the transmission medium, transmitting information by generating pressure pulse signals within the mud. The transmitter is installed near the drill bit, using a pulse generator to create pressure changes in the mud. The receiver is installed on the surface, using a pressure sensor to detect the pressure pulse signals and acquire downhole information.

[0003] Mud pulse telemetry (MPT) systems include pulse modulation and continuous wave modulation, which use different signal generators. Pulse modulation is an earlier modulation method, divided into positive and negative pulses. Currently, to improve the communication rate of mud pulse telemetry systems, continuous wave modulation is being developed. Continuous wave modulation can be further divided into three types: on-off keying (OOK), frequency shift keying (FSK), and phase shift keying (PSK). Among them, the PSK signal is the most unique communication signal in mud pulse telemetry. Since the mud pressure signal is generated by changing the pressure through a mechanical device, the pressure cannot change suddenly. To create a phase change in the signal, a transition section needs to be inserted between the two signals. This transition section has a different frequency than the data section; typically, the carrier frequency of the transition section is half the frequency of the data section, thus changing the phase of the signal section and achieving PSK modulation of the mud pulse.

[0004] Existing continuous-phase BPSK signal demodulation techniques for mud have significant shortcomings: traditional methods typically only process the data frequency portion, completely ignoring the redundant information carried by the transition frequency portion introduced to achieve smooth phase switching; at the same time, existing technologies process the data and transition information in isolation, lacking an effective joint decoding mechanism and failing to utilize their inherent correlation to achieve information complementarity; furthermore, conventional demodulation often employs non-optimal decision criteria and does not model based on the signal's inherent coding structure, resulting in limited bit error rate performance under complex mud channels. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention proposes a dual-frequency block maximum a posteriori probability demodulation method for continuous phase BPSK in mud. While performing conventional demodulation of the signal frequency, it fully utilizes the information of the transition signal frequency. Through information fusion and maximum a posteriori probability decoding, the information of the two frequency parts is made complementary, thereby reducing the bit error rate of communication.

[0006] The objective of this invention is achieved through the following technical solution: A dual-frequency block maximum a posteriori probability demodulation method for continuous phase BPSK in mud includes the following steps: S1: A mud pressure phase shift keying signal is generated at the bottom of the well via a pressure generator; S2: Acquire pressure change data on the ground using pressure sensors; S3: Use mud noise cancellation method to cancel noise in the signal; S4: Use synchronization methods in communication to obtain the starting position of the signal; S5: Use bandpass filters to obtain the data frequencies respectively. f d Signal and transition frequency f t The signal; S6: Down-convert signals of different frequencies, obtain baseband symbols through sampling, and use the baseband symbols as input to perform equalization on the symbols using a decision feedback equalizer based on recursive least squares, obtaining the constellation points of the data code and phase modulation code, containing three types of soft information: the data code with frequency... f d The soft information and phase modulation code after downconversion are based on frequency. f d The soft information and phase modulation code after downconversion are based on frequency. f t Soft information after downconversion; S7: The data code and phase modulation code are fused and modeled into a (3,1,1) convolutional code structure. The above three types of soft information are input into the BCJR decoder to calculate the transition probability and posterior probability of each state, and finally output the demodulation decision result.

[0007] Furthermore, the mud pressure phase-shift keying signal generated in S1 contains three carrier cycles within one symbol period: the first two carrier cycles are used to carry data, with a frequency of f d The last carrier cycle is used for symbol transition; when adjacent symbols are out of phase, this cycle uses a frequency of [frequency value missing]. f t = f d / 2 phase modulation waveform; when adjacent symbols have the same phase, the data waveform is repeated once.

[0008] Furthermore, in S3, an empirical waveform noise reduction algorithm is used to remove the structured noise caused by the mud pump, so that the remaining noise approximately follows an additive Gaussian white noise distribution.

[0009] Furthermore, in S4, a two-stage synchronization method using LFM coarse synchronization and m-sequence fine synchronization is employed to accurately determine the frame start position of the received signal.

[0010] Furthermore, S6 specifically includes: Frequency of use is f d The local carrier down-converts the BPSK signal, samples the data portion of the signal, and inputs the sampled data in blocks into the RLS-DFE equalizer to obtain the soft information of the data code. b data ; Frequency of use is f d The local carrier pair frequency is f d The received signal is down-converted, then low-pass filtered, and then the phase modulation portion of each cycle is sampled. After passing through an RLS-DFE equalizer, soft information is obtained. ; Frequency of use is f t The local carrier pair frequency is f t The received signal undergoes down-conversion and low-pass filtering. The phase modulation portion of each cycle is sampled, and then passed through an RLS-DFE equalizer to obtain soft information. ; The RLS-DFE equalizer uses a recursive least squares algorithm to adaptively update the coefficients of the feedforward and feedback filters, and combines a phase-locked loop for phase compensation to suppress inter-symbol interference and phase drift caused by multipath time-varying channels.

[0011] Furthermore, the BCJR decoder in S7 adopts the following state transition model: the current encoding state is determined by the transmitted bit in the previous moment, the state transition corresponds to the input bit 0 or 1, and each state transition corresponds to three observations, which are composed of the likelihood functions of three types of soft information under the Gaussian white noise assumption. Finally, the original bit sequence is determined by the maximum a posteriori probability criterion.

[0012] A dual-frequency block maximum a posteriori probability demodulation device for continuous phase BPSK in mud includes one or more processors for implementing the dual-frequency block maximum a posteriori probability demodulation method for continuous phase BPSK in mud.

[0013] An electronic device, comprising: One or more processors; A storage device for storing one or more programs that, when executed by the electronic device, enable the electronic device to implement a dual-frequency block maximum a posteriori probability demodulation method for mud continuous phase BPSK.

[0014] A computer-readable storage medium having a program stored thereon that, when executed by a processor, implements a dual-frequency block maximum a posteriori probability demodulation method for continuous phase BPSK in mud.

[0015] The beneficial effects of this invention are as follows: (1) The mud pressure phase shift keying dual-frequency block demodulation method of the present invention effectively utilizes the redundant phase modulation information in the signal and optimizes the decoding process.

[0016] (2) The maximum a posteriori probability decoding criterion of the mud pressure phase shift keying block of the present invention is the optimal decoding criterion, which can reduce the bit error rate of mud pressure phase shift keying modulation.

[0017] (3) The mud pressure phase shift keying block demodulation method of the present invention does not require any hardware changes and can be directly applied to the mud pressure phase shift keying communication system, thereby improving communication performance. Attached Figure Description

[0018] Figure 1 This is a flowchart of the dual-frequency block maximum a posteriori probability demodulation of mud continuous phase shift keying modulation according to an embodiment of the present invention.

[0019] Figure 2 This is a frame structure diagram of the mud continuous phase shift keying signal.

[0020] Figure 3 This is a schematic diagram of the signal connection status in a mud BPSK system where the phase shift intervals of consecutive symbols are the same or different.

[0021] Figure 4 This is a flowchart of the three-channel sampling equalization system of the mud BPSK system.

[0022] Figure 5 This is a schematic diagram of constellation point mapping for three symbols.

[0023] Figure 6 This is a schematic diagram of the state transition relationship of the passband signal corresponding to the mud continuous phase shift keying control signal.

[0024] Figure 7 This is a schematic diagram of the system transition between the encoding and decoding states. The part highlighted in red in the diagram indicates one of the decoding links.

[0025] Figure 8 This is a schematic diagram of the RI-BCJR algorithm flow.

[0026] Figure 9 This is a comparison chart of the bit error rate performance of the BPSK system of this invention with that of traditional methods. Detailed Implementation

[0027] The present invention will be described in detail below with reference to the accompanying drawings and preferred embodiments. The purpose and effects of the present invention will become clearer. It should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0028] Explanation of technical terms: BPSK, Binary Phase Shift Keying; LFM, Linear Frequency Modulation; DFE, decision feedback equalizer; RLS, Recursive Least Squares; RLS-DFE, decision-feedback equalization based on recursive least squares; MPT, mud pulse telemetry; ISI, Inter-symbol interference; PLL, phase-locked loop; RI-BCJR algorithm, Redundant Information BCJR algorithm; like Figure 1 As shown, the dual-frequency block maximum a posteriori probability demodulation method for continuous phase BPSK in mud according to an embodiment of the present invention specifically includes the following steps S1~S7: S1: A mud pressure phase shift keying signal is generated at the bottom of the well via a pressure generator.

[0029] Mud continuous waves are pressure waves generated by a mechanical pressure controller. Due to their mechanical characteristics, abrupt phase changes are not allowed; therefore, a transition waveform is necessary to transmit phase-shift keying signals. Introducing a transition waveform reduces the signal bit rate but also introduces information redundancy. Utilizing this information can reduce the signal error rate to some extent.

[0030] Phase-shift keying (PSK) waveforms are the most complex in mud waveforms. When the preceding and following symbols are inconsistent, meaning the phase needs to change, due to the inherent characteristic that mechanically generated pressure waves cannot change abruptly, a phase modulation component needs to be added. This is achieved by reducing the frequency to half of the normal frequency, resulting in a 180° phase reversal within one carrier cycle. In other words, the generated continuous-phase PSK signal in mud contains three carrier cycles within one symbol cycle: the first two carrier cycles carry data at a frequency of... f d The last carrier cycle is used for symbol transition; when adjacent symbols are out of phase, this cycle uses a frequency of [frequency value missing]. f t = f d / 2 phase modulation waveform; at the same time, in order to improve the demodulation effect, when adjacent symbols have the same phase, the data waveform needs to be repeated once.

[0031] The frame structure for data transmission is as follows Figure 2 As shown, it consists of LFM signals, intervals, m-sequences, and data packets.

[0032] S2: Obtain pressure change data on the ground using pressure sensors.

[0033] S3: Use mud noise cancellation method to cancel noise in the signal.

[0034] In this embodiment, an empirical waveform denoising algorithm is used to remove the structured noise caused by the mud pump, so that the remaining noise approximately follows an additive Gaussian white noise distribution.

[0035] S4: Use synchronization methods in communication to obtain the starting position of the signal.

[0036] In this embodiment, a two-stage synchronization method using LFM coarse synchronization and m-sequence fine synchronization is employed to accurately determine the frame start position of the received signal.

[0037] S5: Use bandpass filters to obtain the data frequencies respectively. f d Signal and transition frequency f t The signal.

[0038] Generally speaking f d =2 f t .like Figure 3 As shown, this embodiment provides a schematic diagram of a modulation process. This modulation method necessarily introduces a phase shift interval between two data intervals. Preferably, the length of the phase shift interval is configured to be half the length of the data interval.

[0039] In this embodiment, the carrier frequency within the data interval is defined as... f d When the symbols of the two data intervals are the same, the carrier frequency within that phase shift interval... f d This will continue for half a cycle of the previous data interval to achieve continuous phase connection, such as... Figure 3 As shown in case ①.

[0040] When the symbols of the two data intervals are different, a transition waveform segment is introduced within the phase shift interval, and the carrier frequency of this segment is... f t To achieve the transition between phases, such as Figure 3 As shown in case ②. It should be noted that this transition waveform segment is not a complete sine wave cycle, but rather half a sine wave cycle.

[0041] The mud phase-shift keying signal contains two main frequency components, one of which is a frequency of f d The data component includes two parts: one is data that appears only when the sign changes. f t Components. In the previous demodulation, the frequency was... f d The data components are demodulated. But in reality, f t The frequency component signal can be viewed as a phase-modulated code; when it appears, the sign changes; when it doesn't appear, the sign remains the same. Assuming that after eliminating pump noise, the statistical law of the noise conforms to the Gaussian white noise law, that is, the noise of the two frequency components follows a Gaussian distribution and is independent of each other, then by utilizing... f t Signals with frequency components can reduce the demodulation bit error rate.

[0042] S6: Down-convert signals of different frequencies, obtain baseband symbols through sampling, and use the baseband symbols as input to perform equalization on the symbols using a decision-feedback equalization based on recursive least squares (RLS-DFE) to obtain the constellation points of the data code and phase modulation code, containing three types of soft information: the data code with frequency... f d The soft information and phase modulation code after downconversion are based on frequency. f d The soft information and phase modulation code after downconversion are based on frequency. f t Soft information after downconversion.

[0043] like Figure 4 As shown, S6 specifically includes: Frequency of use is f d The local carrier down-converts the BPSK signal, samples the data portion of the signal, and inputs the sampled data in blocks into the RLS-DFE equalizer to obtain the soft information of the data code. b data .

[0044] Frequency of use is f d The local carrier pair frequency is f d The received signal is down-converted, then low-pass filtered, and then the phase modulation portion of each cycle is sampled. After passing through an RLS-DFE equalizer, soft information is obtained. ; Frequency of use is f t The local carrier pair frequency is f t The received signal undergoes down-conversion and low-pass filtering. The phase modulation portion of each cycle is sampled, and then passed through an RLS-DFE equalizer to obtain soft information. .

[0045] After denoising, synchronization, down-conversion, and low-pass filtering, the signal received by the MPT system can be defined as: .

[0046] in, x ( n ) represents the launch symbol. h ( t ) represents the generalized channel impulse response, which includes mud channel characteristics and transceiver filter response. T The symbol period consists of the data interval and the phase shift interval. Indicates the delay amount. This is due to carrier phase distortion. Since the input signal is assumed to be a denoised mud pulse signal, free from structured pump noise, therefore... v ( t It only exhibits as pure additive noise.

[0047] Then, by analyzing the received signal... r ( t Sampling is performed to obtain r ( n Its description is as follows: .

[0048] in, h ( n,l ) indicates the firstl indivual( l=0,…,L- 1) The path at time n The gain, v( n The noise is additive white Gaussian noise. It should be noted that, for simplified calculations, the phase rotation parameter is already included in the channel gain. To mitigate inter-symbol interference (ISI), this invention employs a decision feedback equalizer (DFE):

[0049]

[0050]

[0051] in and These represent the coefficients of the feedforward and feedback filters, respectively. Indicates observation symbols, The decision symbol is used. The calculated estimated value is... Compared with the true value x ( n estimation error between ) e ( n This can be represented as: , In training mode, x ( n It is generated from a known training sequence; while in balanced mode, x ( n Then by Replacement.

[0052] Phase estimate The calculation can be performed using a phase-locked loop (PLL) as follows:

[0053] In this formula, This represents the output of the equivalent phase detector. Coefficient. and It is a proportional constant used to adjust the gain. Typically, there is a relationship between them. The final expression is:

[0054] in Input parameters Used to update the Kalman gain vector, which is compared with... w ( n), Calculated together w ( n+ 1) The RLS algorithm used in the equations enables the phase-compensated RLS-DFE to effectively suppress ISI caused by multipath time-varying mud channels and phase drift caused by insufficient motor control accuracy. The symbols after equalization processing... This constitutes soft information b, which is used for subsequent decoding. Through the triple-parallel RLS-DFE algorithm, three types of soft information are ultimately obtained: b data , and .

[0055] S7: The data code and phase modulation code are fused and modeled into a (3,1,1) convolutional code structure. The above three types of soft information are input into the BCJR decoder to calculate the transition probability and posterior probability of each state, and finally output the demodulation decision result.

[0056] In mud BPSK signal modulation, information s data It can be viewed as an inherent self-encoding system. Let the transmitted symbol be... s data After modulation and mapping, the symbol generates a message in the data interval and two redundant messages corresponding to different frequencies in the phase shift interval. and Send sequence s Depend on s data , and The composition can be interpreted as the output of a (3,1,1) encoder:

[0057] Three symbols s data , and The constellation point mapping relationship is as follows Figure 5 As shown, the state transition relationship of the corresponding passband signal is as follows: Figure 6 As shown. The state transition diagram of the compilation and decoding states is as follows. Figure 7 As shown. S 0 and S 1 represents the current state of the encoding register, and the input-output relationship is represented as 0 / 0,1,0. Assuming the red link in the encoding process is shown, from time... t -1 to time t Register status And input The resulting encoded output is , and .

[0058] The modulator will each a i Mapped to a symbol in the symbol table s i ,in Corresponding bit mode s i . ,and Represents the complex field. To illustrate the algorithm to be derived, we use... Figure 8 The three symbolic constellation diagrams are listed. The sequence is obtained through complex number mapping: .

[0059] Utilizing soft information b obtained from triple parallel RLS-DFE to estimate data a data And obtain the data through demapping. s data The form of 'b' is as follows:

[0060] Sequences a and b satisfy the following relation:

[0061] in, n data , and These are three different types of Gaussian white noise. Assumption: Follows a mean of 0 and a variance of Gaussian distribution; Follows a mean of 0 and a variance of The Gaussian distribution; and n data Follows a mean of 0 and a variance of The Gaussian distribution.

[0062] In order to accurately extract the sequence at the receiving end Sequence a is derived using the maximum a posteriori probability criterion. At time t... t Given the current state and the next state When the input is 0, the set of state transitions is denoted as... S The set of state transitions with input 0 and input 1 is denoted as S 1. Therefore, , The key is to calculate this probability in the following way: .

[0063] To calculate the joint probability The soft information b needs to be split into three parts: the first part is t The data received before the time, the second part is t The last part of the information received at any time is t The information received after a certain time. These three parts can be collectively referred to as... Therefore, the probability is calculated by multiplying these three components together: , in, , , .

[0064] The BCJR decoder in S7 adopts the following state transition model: the current encoding state is determined by the transmitted bit in the previous moment, the state transition corresponds to the input bit 0 or 1, and each state transition corresponds to three observations, which are composed of the likelihood functions of three types of soft information under the assumption of Gaussian white noise. Finally, the original bit sequence is determined by the maximum a posteriori probability criterion.

[0065] The processing procedure of the RI-BCJR algorithm (Reduced-Interval BCJR) is as follows: Figure 8 As shown, the detailed processing procedure is as follows: First calculate It indicates that upon receiving soft information b t From the state s p To the next state s q The transition probability.

[0066]

[0067] in It represents a moment. t status s p At the moment t+1 status s q The probability of transformation. At the same time, That is, in the state s p Transition to state s q Received soft information b t The probability of . Among them, Represents the input bits. Because... , and They each follow three different Gaussian distributions, therefore Can be converted to:

[0068] Then calculate according to the recursive formula. a t+1 ( q ):

[0069] and by β t+1 ( q Recursive calculation β t ( p ) .

[0070] Based on this, the original signal sequence can be estimated. a data Then get s data .

[0071] Next, the RI-BCJR algorithm with a triple parallel equalizer is evaluated through numerical simulation and field experiments, and its performance is compared with that of the traditional hard decision algorithm.

[0072] The following parameters were selected in the simulation: f d =12 Hz, f t =6 Hz, symbol rate R B =4Hz, sampling frequency f s = 1000 The frequency was Hz, each data packet contained 544 symbols, and the Monte Carlo simulation was performed 10,000 times. The channel delay was set to 750 sampling periods, a value based on experimentally estimated channel length, and the simulation was conducted under Gaussian white noise conditions. Figure 9 The final simulation results show that, when Eb / N0 is 10dB, the RI-BCJR algorithm using a three-way parallel equalizer improves performance by 1dB compared to the hard-decision equalization algorithm.

[0073] It will be understood by those skilled in the art that the above descriptions are merely preferred examples of the invention and are not intended to limit the invention. Although the invention has been described in detail with reference to the foregoing examples, those skilled in the art can still modify the technical solutions described in the foregoing examples or make equivalent substitutions for some of the technical features. All modifications and equivalent substitutions made within the spirit and principles of the invention should be included within the scope of protection of the invention.

Claims

1. A dual-frequency block maximum a posteriori probability demodulation method for continuous phase BPSK in mud, characterized in that, Includes the following steps: S1: A mud pressure phase shift keying signal is generated at the bottom of the well via a pressure generator; S2: Acquire pressure change data on the ground using pressure sensors; S3: Use mud noise cancellation method to cancel noise in the signal; S4: Use synchronization methods in communication to obtain the starting position of the signal; S5: Use bandpass filters to obtain the data frequencies respectively. f d Signal and transition frequency f t The signal; S6: Down-convert signals of different frequencies, obtain baseband symbols through sampling, and use the baseband symbols as input to perform equalization on the symbols using a decision feedback equalizer based on recursive least squares, obtaining the constellation points of the data code and phase modulation code, containing three types of soft information: the data code with frequency... f d The soft information and phase modulation code after downconversion are based on frequency. f d The soft information and phase modulation code after downconversion are based on frequency. f t Soft information after downconversion; S7: The data code and phase modulation code are fused and modeled into a (3,1,1) convolutional code structure. The above three types of soft information are input into the BCJR decoder to calculate the transition probability and posterior probability of each state, and finally output the demodulation decision result.

2. The dual-frequency block maximum a posteriori probability demodulation method for continuous phase BPSK in mud according to claim 1, characterized in that, The mud pressure phase-shift keying signal generated in S1 contains three carrier cycles within one symbol period: the first two carrier cycles are used to carry data, with a frequency of [frequency missing]. f d The last carrier cycle is used for symbol transition; when adjacent symbols are out of phase, this cycle uses a frequency of [frequency value missing]. f t = f d / 2 phase modulation waveform; when adjacent symbols have the same phase, the data waveform is repeated once.

3. The dual-frequency block maximum a posteriori probability demodulation method for continuous phase BPSK in mud according to claim 1, characterized in that, In S3, an empirical waveform noise reduction algorithm is used to remove the structured noise caused by the mud pump, so that the remaining noise approximately follows an additive Gaussian white noise distribution.

4. The dual-frequency block maximum a posteriori probability demodulation method for continuous phase BPSK in mud according to claim 1, characterized in that, The S4 uses a two-stage synchronization method of LFM coarse synchronization and m-sequence fine synchronization to accurately determine the frame start position of the received signal.

5. The dual-frequency block maximum a posteriori probability demodulation method for continuous phase BPSK in mud according to claim 1, characterized in that, S6 specifically includes: Frequency of use is f d The local carrier down-converts the BPSK signal, samples the data portion of the signal, and inputs the sampled data in blocks into the RLS-DFE equalizer to obtain the soft information of the data code. b data ; Frequency of use is f d The local carrier pair frequency is f d The received signal is down-converted, then low-pass filtered, and then the phase modulation portion of each cycle is sampled. After passing through an RLS-DFE equalizer, soft information is obtained. ; Frequency of use is f t The local carrier pair frequency is f t The received signal undergoes down-conversion and low-pass filtering. The phase modulation portion of each cycle is sampled, and then passed through an RLS-DFE equalizer to obtain soft information. ; The RLS-DFE equalizer uses a recursive least squares algorithm to adaptively update the coefficients of the feedforward and feedback filters, and combines a phase-locked loop for phase compensation to suppress inter-symbol interference and phase drift caused by multipath time-varying channels.

6. The dual-frequency block maximum a posteriori probability demodulation method for continuous phase BPSK in mud according to claim 1, characterized in that, The BCJR decoder in S7 adopts the following state transition model: the current encoding state is determined by the transmitted bit at the previous moment, the state transition corresponds to the input bit 0 or 1, and each state transition corresponds to three observations, which are composed of the likelihood functions of three types of soft information under the assumption of Gaussian white noise. Finally, the original bit sequence is determined by the maximum a posteriori probability criterion.

7. A dual-frequency block maximum a posteriori probability demodulation device for continuous phase BPSK in mud, characterized in that, It includes one or more processors for implementing the dual-frequency block maximum a posteriori probability demodulation method for continuous phase BPSK in mud as described in any one of claims 1 to 6.

8. An electronic device, characterized in that, include: One or more processors; A storage device for storing one or more programs, which, when executed by the electronic device, cause the electronic device to implement the dual-frequency block maximum a posteriori probability demodulation method for mud continuous phase BPSK as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, It stores a program that, when executed by a processor, implements the dual-frequency block maximum a posteriori probability demodulation method for continuous phase BPSK in mud as described in any one of claims 1 to 6.