FPGA (Field Programmable Gate Array) time sequence control algorithm for marine atom gravimeter
By using FPGA timing control algorithms and Bayesian inference cosine estimation methods, the problems of insufficient synchronization and anti-interference capabilities of marine atomic gravimeters in dynamic measurements were solved, achieving high-precision gravity measurements and adapting to the dynamic measurement needs in marine environments.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-01-08
- Publication Date
- 2026-04-07
AI Technical Summary
In existing technologies, marine atomic gravimeters suffer from insufficient timing control synchronization and anti-interference capabilities during dynamic measurements, resulting in inadequate gravity measurement accuracy. This is especially true when vibration interference is aggravated in the marine environment, making it difficult to achieve high-precision measurements.
By employing an FPGA timing control algorithm, combined with an optical subsystem and a Bayesian inference cosine estimation method, the FPGA control module enables synchronous output of multiple signals and data acquisition. In conjunction with the host computer processing, timing instructions are generated and data is calculated. The Bayesian inference cosine estimation method is used to extract the true phase information, thereby achieving high-precision gravity measurement.
It achieves high-precision gravity measurement with high timing synchronization accuracy and strong anti-interference capability, improving the measurement accuracy to 0.8 mGal, and is adapted to the expansion needs of marine atomic gravimeters from static laboratory measurements to dynamic field measurements.
Smart Images
Figure FT_1 
Figure FT_2 
Figure FT_3
Abstract
Description
Technical Field
[0001] This invention relates to the field of high-precision gravity matching navigation technology, and in particular to an FPGA timing control algorithm for marine atomic gravimeters. Background Technology
[0002] Gravimetry, as a key discipline that intersects with seismology, geology, fundamental physics, geodynamics, and other fields, holds irreplaceable strategic significance in areas such as the establishment of national surveying and mapping benchmarks, oil and gas resource exploration, military security, and disaster monitoring and early warning. The gravitational field can directly reflect the density distribution characteristics of Earth's materials. Precise gravity field information not only provides core data support for geophysical research and the detection of fundamental physical indicators, but also plays a crucial role in practical applications such as tidal model construction, seabed gravity field mapping, and passive navigation. Therefore, accurate measurement of gravity indicators is of great value for both the theoretical extension of basic disciplines and engineering practical applications.
[0003] As a novel high-precision inertial sensor for absolute gravity measurement, the marine atomic gravimeter utilizes a stable laser light source and an external magnetic field to construct a magneto-optical trap, achieving efficient cooling of alkali metal atoms. Through the beam splitting, reversal, and combining processes of Raman interferometry, combined with laser frequency scanning, it calculates gravity values, demonstrating significant performance advantages in the field of gravity measurement. With the development of cold atom interferometer technology, its application scenarios have gradually expanded from static laboratory measurements to dynamic field measurements, making miniaturized and intelligent marine atomic gravimeters a research hotspot. However, dynamic absolute gravity measurement places stringent requirements on the synchronization and time delay accuracy of timing control: on the one hand, the Raman interferometry process is extremely sensitive to synchronization and time delay, and precise Raman light sequence control is key to reducing measurement errors; on the other hand, interference factors such as wave vibrations in the marine environment can easily lead to loosening of equipment modules and signal fluctuations, further exacerbating the difficulty of timing control.
[0004] In existing technologies, while FPGA control alone can achieve multi-channel signal output, its limited fixed-point computing capabilities prevent it from handling complex gravity data calculations. Conversely, relying solely on host computer processing fails to meet the time delay accuracy requirements for synchronous output of multiple signals. Furthermore, traditional interferometric fringe calculation methods struggle to effectively extract true phase information under vibration interference, resulting in insufficient gravity measurement accuracy. Therefore, a timing control algorithm that balances synchronization performance, anti-interference capability, and computational accuracy is urgently needed to overcome the technical bottlenecks in dynamic measurements using marine atomic gravimeters. Summary of the Invention
[0005] The purpose of this invention is to address the shortcomings of existing technologies by proposing an FPGA timing control algorithm for marine atomic gravimeters.
[0006] To achieve the above objectives, the present invention adopts the following technical solution: The FPGA timing control algorithm for a marine atomic gravimeter includes the following steps: S1: Construct an optical subsystem, which includes an MOT module, an interference module, and a detection module. The DI / DO module of the FPGA control module outputs a TTL signal to control the AOM switch, and the AI / AO module outputs an analog voltage to control the magnetic field detuning, cooling optical power, and polarization gradient cooling. S2: The host computer data processing module generates timing control instructions, packages them, and sends them to the FPGA control module. The FPGA control module sets the duration of each time period through the Time FIFO module to achieve accurate time output. It controls the AOM switch and Raman light switching through the DI / DO module. The AI / AO module continuously outputs voltage values according to the time setting to control frequency detuning and cooling optical power. S3: Set the number of acquisition points through the DMA FIFO module of the FPGA control module, receive the trigger signal to synchronously start AI signal acquisition, acquire the voltage signal of the detection module, and calculate the population by solving the amplitude of the two detection voltage signals; S4: The Bayesian inference cosine estimation method for interferometric fringe calculation is used to process the collected population data, calculate the gravity value, and display it dynamically.
[0007] As a further improvement of the present invention, the MOT module in step S1 is fixedly connected to the light tube system and outputs a red detuned and shaped light spot through AOM frequency modulation, the diameter of which is 16mm.
[0008] As a further improvement of the present invention, the output terminal of the DI / DO module in step S1 is connected in series with a voltage follower, the time delay of the FPGA control module is less than 1μs, and the long-term stable voltage fluctuation is less than 0.1V.
[0009] As a further improvement of the present invention, the time delay of the polarization gradient cooling stage in step S1 is maintained at 750ns, the Π pulse time of the speed selection stage is set to 16μs, and the corresponding AOM switch is controlled by the output digital TTL signal driven by the DI / DO module.
[0010] As a further improvement of the present invention, in step S1, the interference stage is driven by the digital module of the FPGA control module to output TTL signal through the data interface, so as to precisely control the Raman beam width and time interval, and realize the beam splitting, reversing and combining process of the atomic package.
[0011] As a further improvement of the present invention, in step S1, the detection stage controls the detection light to be switched on and off twice through the DI / DO module, the FPGA synchronously collects the voltage signal and transmits it to the host computer, and completes the calculation of the atomic group distribution probability in conjunction with the fluorescence acquisition module.
[0012] As a further improvement of the present invention, the Bayesian inference cosine estimation interference fringe solution method in step S4 includes the following sub-steps: S41: Input the raw "absolute time - transition probability" data, perform outlier removal and normalization, and extract the data dimensions; S42: Constructing the likelihood observation model: in Let be the population at time i, C be the fringe contrast, ω be the fringe angular frequency, B be the fringe DC bias, ϕ be the fringe initial phase, and εi be the Gaussian noise. σ is the noise standard deviation; S43: Set the prior distribution, construct the design matrix G, information matrix S and projection vector L, perform joint sampling of parameters through MCMC sampling or nested sampling, calculate the posterior mean or MAP estimate and stripe contrast, and obtain the accurate gravity value by taking the 95% confidence interval of the posterior sample.
[0013] As a further improvement of the present invention, the prior distribution satisfies in This is a linear parameter vector related to the stripe contrast C and the DC bias B. It follows a uniform distribution with upper and lower limits. We adopt Jeffreys prior, i.e., no-information reference prior, in the form of:
[0014] As a further improvement of the present invention, the interference fringe solution method based on Bayesian inference cosine estimation improves the external coincidence accuracy of gravity measurement to 0.8 mGal, and the fringe contrast is...
[0015] The beneficial effects of this invention are: High timing synchronization accuracy and strong stability: Through the collaborative work of the DI / DO, AI / AO modules and TimeFIFO module of the FPGA control module, precise timing control of TTL signals and analog voltage output is achieved, with a time delay of less than 1μs and a long-term stable voltage fluctuation of less than 0.1V. Moreover, the timing parameters of key stages such as polarization gradient cooling and speed selection are precisely controllable, effectively meeting the stringent requirements of Raman interferometry for synchronization and time delay accuracy, laying the foundation for the accuracy of gravity measurement.
[0016] Balancing Synchronization Performance and Computational Adaptability: Combining the hardware control advantages of FPGA with the instruction issuance function of host computer, FPGA is responsible for real-time tasks such as multi-channel signal synchronous output and data acquisition, while host computer is responsible for timing instruction generation and dynamic data display. This solves the technical pain points of insufficient fixed-point computing power of FPGA alone and insufficient latency accuracy when relying solely on host computer, thus achieving a balance between synchronization performance and computational adaptability.
[0017] Outstanding anti-interference capability and improved measurement accuracy: The interference fringe solution method using Bayesian inference cosine estimation, through outlier removal, normalization preprocessing, and parameter estimation methods such as MCMC sampling or nested sampling, can still effectively extract true phase information under the interference of ocean wave vibration, improving the external coincidence accuracy of gravity measurement to 0.8 mGal, which significantly optimizes the problem of insufficient measurement accuracy of traditional solution methods in interference scenarios.
[0018] Functional integration to meet dynamic measurement needs: The algorithm covers the entire process of optical subsystem control, timing command execution, signal acquisition, data calculation and gravity value output, realizing integrated control of MOT cooling, Raman interferometry, signal detection and data processing. It is adapted to the expansion needs of marine atomic gravimeters from static laboratory measurements to dynamic field measurements, and provides technical support for the application of miniaturized and intelligent marine atomic gravimeters.
[0019] This invention ensures long-term system stability while effectively improving the timing synchronization performance of each module of the marine atomic gravimeter, enabling high-precision absolute gravity measurement. Attached Figure Description
[0020] Figure 1 This is a schematic diagram of the FPGA timing control flow of the FPGA timing control algorithm for a marine atomic gravimeter proposed in this invention. Figure 2 This is a schematic diagram of the DIO delay test of the FPGA timing control algorithm for marine atomic gravimeter proposed in this invention; Figure 3 This is a schematic diagram of the AO delay in the FPGA timing control algorithm for a marine atomic gravimeter proposed in this invention; Figure 4 This is a schematic diagram illustrating the synchronization test of the FPGA timing control algorithm for a marine atomic gravimeter proposed in this invention. Figure 5 This is a complete flowchart of the FPGA timing control algorithm for a marine atomic gravimeter proposed in this invention; Figure 6 This is a flowchart of the interference fringe fitting process for Bayesian cosine estimation in the FPGA timing control algorithm for marine atomic gravimeter proposed in this invention. Figure 7This is a schematic diagram of the timing control chassis for the FPGA timing control algorithm for a marine atomic gravimeter proposed in this invention; Figure 8 This is a schematic diagram of AOM control for the FPGA timing control algorithm for marine atomic gravimeter proposed in this invention; Figure 9 This is a schematic diagram of the gravity acquisition module solving the FPGA timing control algorithm for marine atomic gravimeter proposed in this invention; Figure 10 This is a schematic diagram illustrating the data acquisition and population calculation of the FPGA timing control algorithm for marine atomic gravimeter proposed in this invention. Figure 11 This is a schematic diagram of the same-direction resonance peak scanning of the FPGA timing control algorithm for marine atomic gravimeter proposed in this invention.
[0021] Figure 12 This is a schematic diagram of Raman co-directional interferometric fringe fitting for the FPGA timing control algorithm for a marine atomic gravimeter proposed in this invention. Figure 13 The image of the MOT trapped atomic group is provided by the FPGA timing control algorithm for the marine atomic gravimeter proposed in this invention. Detailed Implementation
[0022] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.
[0023] I. System Setup Preparation (I) Hardware Selection and Setup FPGA control module: An FPGA chassis with multi-channel AI / AO and DI / DO interfaces is selected, and a TimeFIFO module, a DMA FIFO module and a high-precision synchronous timer are configured to ensure the accuracy of timing control; a voltage follower, model OPA277, is connected in series at the output of the DI / DO module to improve signal driving capability and control stability.
[0024] Optical subsystem setup: MOT Module: It adopts a fixed structure of light tube system. The light tube is made of high-strength aluminum alloy to prevent the module from loosening due to wave vibration. The optical path system is built and frequency-modulated by AOM crystal (model Gooch & Housego 4001-143) to output red detuned and shaped light spot. The light spot diameter is calibrated to 16mm by optical lens group.
[0025] Interference module: Construct a Raman interference optical path, configure two narrow linewidth lasers (linewidth ≤ 1kHz) as Raman light sources, and adjust the optical path direction and polarization state through optical mirrors and polarizers to ensure the stability of Raman light beam splitting, reversal and combining processes.
[0026] Detection module: A photomultiplier tube (model H10721-01) is selected as the fluorescence detection device, which, together with the fluorescence acquisition module, achieves high-sensitivity acquisition of the fluorescence signal of the atomic group; a dual-channel detection optical switch is configured, which is controlled by the DI / DO module.
[0027] Host computer and display module: The host computer is an industrial control computer with LabVIEW 2021 software development environment installed to develop functional modules such as timing control instruction generation, data acquisition and processing, and gravity value calculation; the display module is a high-definition touch screen for dynamically displaying information such as gravity value, stripe contrast, and timing parameters.
[0028] (II) Software Configuration The host computer software is developed based on LabVIEW and includes a timing control instruction editing module, a data acquisition and storage module, a population calculation module, a Bayesian inference cosine estimation calculation module, and a data visualization module. It supports instruction package distribution, real-time data reception, and dynamic calculation of gravity values.
[0029] FPGA firmware: Written in Verilog HDL, it implements functions such as Time FIFO timing scheduling, DMA FIFO data buffering, and I / O port synchronization control, and is embedded in the FPGA chip to ensure fast response and accurate execution of timing instructions.
[0030] II. Step-by-step execution process of the algorithm (I) Step S1: Start-up and control of the optical subsystem MOT Module Startup: The host computer issues a MOT startup command. The FPGA control module outputs a TTL signal through the DIO0 interface of the DI / DO module to control the magnetic field switch to close, and outputs a TTL signal through the DIO1 interface to control the return pump optical switch to close. The AI / AO module outputs an analog voltage (initial value set to 2.5V) through the AO0 interface to control the magnetic field detuning, and outputs an analog voltage (initial value set to 3.0V) through the AO1 interface to control the cooling optical power. The AOM control chassis drives the AOM crystal response to achieve atomic cluster trapping in the magneto-optical trap system. After the MOT module starts up, it runs stably for 30 seconds to ensure the atomic cluster trapping effect.
[0031] Polarization gradient cooling control: The host computer issues polarization gradient cooling commands, and the FPGA control module outputs a fixed analog voltage (set to 1.8V) through the AI / AO module. Combined with the FPGA's built-in clock synchronization mechanism, the cooling timing is precisely matched with the MOT area timing, the time delay is strictly maintained at 750ns, and the cooling duration is set to 500μs to achieve deep cooling of the atomic cluster.
[0032] Velocity selection control: The FPGA control module outputs a 16μs Π pulse TTL signal through the DI / DO module to control the Raman optical switch to turn on, and then activates the pump state optical selection, with the selection duration set to 200μs. During the velocity selection and state preparation stages, digital TTL signals are output through the DIO2 and DIO3 interfaces of the DI / DO module to control the on / off state of the corresponding AOM switches, thereby completing the atomic velocity screening.
[0033] Interference phase control: The host computer presets the Raman beam width (set to 8μs) and time interval (set to 2T, where T is the free fall time of atoms, set to 10ms according to measurement requirements). The digital module of the FPGA control module drives the data interface to output TTL signals. The FPGA precisely controls the on / off duration and interval of the Raman beam through the pulse width, realizing the beam splitting, reversal, and beam combining process of the atom packet. The duration of the interference phase is 2T+20μs.
[0034] Detection phase control: The FPGA control module outputs control signals through the DIO4 interface of the DI / DO module to control the detection light to be switched on and off twice, with the interval between the two switching on and off being set to 5μs; at the same time, the DMA FIFO module is triggered to start AI signal acquisition, which acquires the voltage signal output by the photomultiplier tube, with the sampling rate set to 1MSa / s and the acquisition duration set to 10μs. After the acquisition is completed, the data is transmitted to the host computer.
[0035] (II) Step S2: Issuance and execution of timing control instructions The host computer sets the duration parameters (such as MOT trapping time, cooling time, interference time, detection time, etc.) and signal output parameters (such as the output voltage range of the AO module and the TTL signal level of the DI module) for each stage in the timing control instruction editing module according to the measurement requirements, and generates a timing control instruction set.
[0036] The host computer packages the instruction set into standard data frames (frame header is 0xAA, frame tail is 0x55, data length is 64 bytes) and sends them to the FPGA control module via the Ethernet interface, with the transmission baud rate set to 100Mbps.
[0037] After receiving the data frame, the FPGA control module splits the data through the frame parsing module, extracts the control instructions and parameters of each interface, and stores them in the internal register. The required duration of each time period is set through the Time FIFO module, and the accurate time output is achieved with the help of the FPGA time counting module (counting frequency of 100MHz), driving the DI / DO module and AI / AO module to perform corresponding operations according to the preset timing sequence.
[0038] (III) Step S3: Data collection and population calculation The host computer data acquisition module receives voltage signal data transmitted by the FPGA control module and stores it on the local hard drive in .csv format, which contains information such as acquisition timestamp, channel number, and voltage value.
[0039] The array processing module preprocesses the acquired voltage signal, uses the moving average method (with the window size set to 10 data points) to remove noise, and then extracts the voltage signal amplitudes V1 and V2 from the two detections.
[0040] The population calculation module calculates the atomic population (p = \frac{V1}{V1+V2}\) according to the formula \(p = \frac{V1}{V1+V2}\), where V1 is the voltage amplitude of the first detection and V2 is the voltage amplitude of the second detection. After the calculation is completed, an "absolute time-population" data sequence is generated.
[0041] (iv) Step S4: Interference fringe solution and gravity value output based on Bayesian inference cosine estimation Data preprocessing: Input the "absolute time-population" data sequence into the Bayesian inference cosine estimation solution module. First, the 3σ criterion is used to remove outliers, and then normalization is performed (the population range is mapped to [0,1]). The data dimension N (set to 1000 data points) is extracted.
[0042] Model building: Constructing a likelihood observation model in observation vector The population at time i, where C is the contrast ratio. B is the fringe angular frequency, and B is the fringe DC bias. The noise is Gaussian, and N is the total number of data points. The value at time i contains a certain mapping relationship of chirping. The initial phase of the stripes, The noise standard deviation and hyperparameters are normalized and estimated.
[0043] Prior distribution setting: Set the prior distribution in Let C and B be linear parameter vectors. It follows a uniform distribution in the range [100π, 200π]. Using Jeffreys' no-information reference prior The prior covariance matrix Set as a diagonal matrix with all diagonal elements being 1, and use the prior mean vector. Set to [0.5, 0.5].
[0044] Matrix Construction and Sampling: Construct a design matrix with dimensions of... Information Matrix and projection vector\ The MCMC sampling algorithm (emcee library) is used to sample the parameters. Joint sampling was performed, with the number of sampling steps set to 10,000. The first 5,000 steps were for the combustion period, and the last 5,000 steps were used for parameter estimation.
[0045] Gravity value calculation and output: Calculate the posterior mean or MAP estimate based on the sampling results, according to the formula. Calculate the fringe contrast using the 95% confidence interval of the posterior sample; based on the fringe angular frequency... The mapping relationship between gravity and gravitational acceleration g (g = \frac{\omega\lambda}{4\pi}\) (λ is the Raman wavelength, known to be 780nm) is used to calculate the gravity value; the gravity value, stripe contrast and 95% confidence interval are dynamically displayed through the display module, and the measurement results are stored in the database, supporting historical data query and export.
[0046] III. Performance Testing and Verification (a) Delay Test The DIO delay and AO delay of the FPGA control module were measured using an oscilloscope (RIGOL MSO5204). The test results show that the DIO delay is less than [value missing]. The AO delay is approximately 625ns, which meets the accuracy requirements of timing control. In long-term stability testing, the voltage fluctuation is less than 0.1V after 24 hours of continuous operation, indicating good system stability.
[0047] (ii) Synchronization test The output signals of the DI / DO module and AI / AO module were simultaneously acquired by an oscilloscope, and the phase difference of the signals was observed. The test results showed that the input and output synchronization of each module was good, and the phase difference was less than 10ns, ensuring the synchronous execution of key links such as Raman interferometry and detection.
[0048] (III) Accuracy Testing Gravity measurement experiments were conducted on a marine simulated vibration platform (vibration frequency 0.1-10Hz, amplitude 0.1-1mm). Compared with the measurement results of a standard gravimeter (accuracy ±0.1mGal), the gravity measurement extrinsic accuracy of the algorithm of this invention reached 0.8mGal, meeting the requirements for high-precision dynamic absolute gravity measurement.
[0049] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.
Claims
1. An FPGA timing control algorithm for a marine atomic gravimeter, characterized in that, Includes the following steps: S1: Construct an optical subsystem, which includes an MOT module, an interference module, and a detection module. The DI / DO module of the FPGA control module outputs a TTL signal to control the AOM switch, and the AI / AO module outputs an analog voltage to control the magnetic field detuning, cooling optical power, and polarization gradient cooling. S2: The host computer data processing module generates timing control instructions, packages them, and sends them to the FPGA control module. The FPGA control module sets the duration of each time period through the Time FIFO module to achieve accurate time output. It controls the AOM switch and Raman light switching through the DI / DO module. The AI / AO module continuously outputs voltage values according to the time setting to control frequency detuning and cooling optical power. S3: Set the number of acquisition points through the DMA FIFO module of the FPGA control module, receive the trigger signal to synchronously start AI signal acquisition, acquire the voltage signal of the detection module, and calculate the population by solving the amplitude of the two detection voltage signals; S4: The Bayesian inference cosine estimation method for interferometric fringe calculation is used to process the collected population data, calculate the gravity value, and display it dynamically.
2. The FPGA timing control algorithm for a marine atomic gravimeter according to claim 1, characterized in that, The MOT module mentioned in step S1 is fixedly connected to the light tube system and outputs a red detuned and shaped light spot through AOM frequency modulation. The diameter of the light spot is 16mm.
3. The FPGA timing control algorithm for a marine atomic gravimeter according to claim 1, characterized in that, In step S1, a voltage follower is connected in series at the output of the DI / DO module, the time delay of the FPGA control module is less than 1μs, and the long-term stable voltage fluctuation is less than 0.1V.
4. The FPGA timing control algorithm for a marine atomic gravimeter according to claim 1, characterized in that, In step S1, the time delay of the polarization gradient cooling stage is maintained at 750 ns, and the Π pulse time of the speed selection stage is set to 16 μs. The corresponding AOM switch is controlled by the output digital TTL signal driven by the DI / DO module.
5. The FPGA timing control algorithm for a marine atomic gravimeter according to claim 1, characterized in that, In step S1, during the interference stage, the TTL signal is output through the digital module drive data interface of the FPGA control module to precisely control the Raman beam width and time interval, thereby realizing the beam splitting, reversal, and beam combining process of the atomic package.
6. The FPGA timing control algorithm for a marine atomic gravimeter according to claim 1, characterized in that, In step S1, during the detection phase, the DI / DO module controls the double switching of the detection light, and the FPGA synchronously acquires the voltage signal and transmits it to the host computer, which works in conjunction with the fluorescence acquisition module to complete the calculation of the atomic group distribution probability.
7. The FPGA timing control algorithm for a marine atomic gravimeter according to claim 1, characterized in that, The Bayesian inference cosine estimation interference fringe solution method described in step S4 includes the following sub-steps: S41: Input the raw "absolute time - transition probability" data, perform outlier removal and normalization, and extract the data dimensions; S42: Constructing the likelihood observation model: in For the first The time-space population, C is the fringe contrast, ω is the fringe angular frequency, B is the fringe DC bias, ϕ is the fringe initial phase, and εi is the Gaussian noise. σ is the noise standard deviation; S43: Set the prior distribution, construct the design matrix G, information matrix S and projection vector L, perform joint sampling of parameters through MCMC sampling or nested sampling, calculate the posterior mean or MAP estimate and stripe contrast, and obtain the accurate gravity value by taking the 95% confidence interval of the posterior sample.
8. The FPGA timing control algorithm for a marine atomic gravimeter according to claim 7, characterized in that, The prior distribution satisfies in To contrast with stripes and DC bias The relevant linear parameter vector, It follows a uniform distribution with upper and lower limits. We adopt Jeffreys prior, i.e., no-information reference prior, in the form of:
9. The FPGA timing control algorithm for a marine atomic gravimeter according to claim 1, characterized in that, The interference fringe solution method based on Bayesian inference cosine estimation improves the external coincidence accuracy of gravity measurements to 0.8 mGal, and the fringe contrast...