Relative Measurement Error Research Articles

The show must go on: a reliability assessment platform for resistive random access memory crossbars.

Resistive random access memory (ReRAM) holds promise for building computing-in-memory (CIM) architectures to execute machine learning (ML) applications. However, existing ReRAM technology faces challenges such as cell and cycle variability, read-disturb and limited endurance, necessitating improvements in devices, algorithms and applications. Understanding the behaviour of ReRAM technologies is crucial for advancement. Existing platforms can either only characterize single cells and do not support CIM operations, or lack a comprehensive software stack for simple system integration. This article introduces NeuroBreakoutBoard (NBB), a versatile, integrable and portable instrumentation platform for ReRAM crossbars. The platform features a software stack enabling experiments via Python from a host PC. In a case study, we demonstrate the capabilities of NBB by conducting diverse experiments on TiN/Ti/HfO2/TiN cells. Our results show that NBB can characterize individual cells and perform CIM operations with a relative measurement error below 2%.This article is part of the theme issue 'Emerging technologies for future secure computing platforms'.

Development and application of a novel scintillation gel-based 3D dosimetry system for radiotherapy.

This study introduced a novel 3D dosimetry system for radiotherapy in order to address the limitations of traditional quality assurance methods in precision radiotherapy techniques. The research required the use of scintillation material, optical measurements, and a dose reconstruction algorithm. The scintillation material, which mimics human soft tissue characteristics, served as a both physical phantom and a radiation detector. The dose distribution inside the scintillator can be converted to light distributions, which were measured by optical cameras from different angles and manifested as pixel values. The proposed dose reconstruction algorithm, LASSO-TV, effectively reconstructed the dose distribution from pixel values, overcoming challenges such as limited projection directions and large-scale matrices. Various clinical plans were tested and validated, including a modified segment from the SBRT plan and IMRT clinical plan. The dosimetry system can execute full 3D dose determinations as a function of time with a spatial resolution of 1-2mm, enabling high-resolution measurements for dynamic dose distribution. Comparative analysis with mainstream device MapCHECK2 confirmed the accuracy of the system, with a relative measurement error of within 5%. Testing and validation results demonstrated the dosimetry system's promising potential for dynamic treatment quality assurance.

Aspherical Surface Wavefront Testing Based on Multi-Directional Orthogonal Lateral Shearing Interferometry.

To overcome the limitations of phase sampling points in testing aspherical surface wavefronts using traditional interferometers, we propose a high-spatial-resolution method based on multi-directional orthogonal lateral shearing interferometry. In this study, we provide a detailed description of the methodology, which includes the theoretical foundations and experimental setup, along with the results from simulations and experiments. By establishing a relational model between the multi-directional differential wavefront and differential Zernike polynomials, we demonstrate high-spatial-resolution wavefront reconstruction using multi-directional orthogonal lateral shearing interferometry. Theoretical calculations and simulations of aspherical surface wavefront testing are followed by experimental verification on an aspherical surface with a known asphericity. Comparing the measurement results with those from the LuphoScan profilometer, we achieve a relative measurement error with an RMS precision better than λ/100.

The Method of Measuring the Occupancy of the Radio Frequency Spectrum, Varying According to the Daily Cycle

Relevance. Management of the use of the radio frequency spectrum requires taking into account the actual occupancy of radio channels and frequency bands. However, the patterns of occupancy change underlying the International Telecommunication Union (ITU-R) Radiocommunication Sector documents do not fully meet the needs of practice. Some of the ITU-R recommendations are focused on estimating local occupancy for short time intervals; other recommendations involve measurements in stationary radio channels, although only a small part of real radio channels can be counted on to have the constant occupancy throughout the time axis. At the same time, the activities of many organizations, and therefore the resources they use, are subject to a daily cycle of activity, which allows us to recommend for consideration a model of occupancy change in accordance with the daily cycle. The aim of the work is to develop a methodology for collecting information and forming an estimate of the daily occupancy change for the analyzed radio channels. Methods used. The development of recommendations for the collection of data for occupancy monitoring is based on the practical approaches of radio monitoring services and classical methods of statistical analysis. Novelty. In this paper, a new model for changing the occupancy of radio frequency channels in accordance with the daily cycle is proposed for use, as well as a statistically justified method for measuring occupancy.Result. The proposed methodology allows us to obtain a practically oriented assessment of occupancy changes, which does not require replacing the fleet of equipment traditionally used by radio monitoring services; the necessary changes can be implemented with relatively simple software modification. The paper provides preliminary recommendations for ensuring the accuracy and reliability of measurements, based on a traditionally used mathematical apparatus, which involves limiting the relative measurement error. This entails the need for very long measurements for the most important low-occupied radio channels. So, it is necessary to look for alternative measurement accuracy requirements that would meet the real needs of radio monitoring services.Practical significance. The acceptance of the developed methodology will eliminate the currently existing contradiction between the theoretical provisions underlying the ITU recommendations and the practice of conducting radio-controlled measurements.

Utilizing artificial intelligence to determine bone mineral density using spectral CT

BackgroundDual-energy computed tomography (DECT) has demonstrated the feasibility of using HAP-water to respond to BMD changes without requiring dedicated software or calibration. Artificial intelligence (AI) has been utilized for diagnosising osteoporosis in routine CT scans but has rarely been used in DECT. This study investigated the diagnostic performance of an AI system for osteoporosis screening using DECT images with reference quantitative CT (QCT). MethodsThis prospective study included 120 patients who underwent DECT and QCT scans from August to December 2023. Two convolutional neural networks, 3D RetinaNet and U-Net, were employed for automated vertebral body segmentation. The accuracy of the bone mineral density (BMD) measurement was assessed with relative measurement error (RME%). Linear regression and Bland–Altman analyses were performed to compare the BMD values between the AI and manual systems with those of the QCT. The diagnostic performance of the AI and manual systems for osteoporosis and low BMD was evaluated using receiver operating characteristic curve analysis. ResultsThe overall mean RME% for the AI and manual systems were − 15.93 ± 12.05 % and − 25.47 ± 14.83 %, respectively. BMD measurements using the AI system achieved greater agreement with the QCT results than those using the manual system (R2 = 0.973, 0.948, p < 0.001; mean errors, 23.27, 35.71 mg/cm3; 95 % LoA, −9.72 to 56.26, −11.45 to 82.87 mg/cm3). The areas under the curve for the AI and manual systems were 0.979 and 0.933 for detecting osteoporosis and 0.980 and 0.991 for low BMD. ConclusionThis AI system could achieve relatively high accuracy for automated BMD measurement on DECT scans, providing great potential for the follow-up of BMD in osteoporosis screening.

Measuring up: Ensuring Intra- and Interobserver Reliability in Stretched Penile Length with the SPLINT Technique.

A discrepancy between the true and measured value of stretched penile length (SPL) may be a result of errors that can either be systematic or random. Hence, it becomes important to focus on the quality of measurements to prevent any iatrogenic harm to the patients. The objective of this study was to assess the magnitude of intra- and interobserver variations in the measurement of SPL with the SPLINT technique. SPL was measured prospectively in a cohort of 449 boys aged 0-14 years including 68 infants (substratified into Group I: >4 years, Group II: 4-8 years, and Group III: >8 years) with the SPLINT technique by expert (E: E1 and E2) and trainee (T: T1 and T2) surgeons after completing a three-tiered training module. Intra- and interobserver variability was assessed through descriptive statistics, intraclass correlation (ICC), relative technical error of measurement (rTEM), and reliability or R (%). Intraobserver variability: the mean difference between the two readings (E1 and E2) is 0.08 cm (95% confidence interval [CI]: 0.073-0.087), ICC was 0.998 (95% CI: 0.997-0.998), and intraobserver variability ≤0.1 cm in 85% of the participants (n = 370 of 433). The rTEM and reliability (%) were 1.82% and 98.1% (Group I), 1.65% and 98.9% (Group II), and 1.09% and 99.7% (Group III), respectively. The intraobserver variability was observed to be inversely proportional to the age of the participants (correlation coefficient = -0.56). Interobserver variability was calculated separately for expert versus trainee and trainee versus trainee (T-vs-T) measurements. For expert versus expert, ICC, rTEM, and reliability (%) were 0.984, 2.4%, and 96.8% (Group 1), 0.992, 2.07%, and 98.3% (Group 2), and 0.997, 1.38%, and 99.05% (Group 3), respectively. A similar pattern of variability was observed for T-vs-T measurements. The reliability (%) of the SPL by experts is consistently more than that of trainees across all age groups; however, the difference ameliorates with the age of participant. The study has validated the SPLINT technique by demonstrating a high level of intra- and interobserver reliability. The adequacy of the training modules for SPL measurements described in this study has also been established. Evidence that the SPL can be used as an objective marker of penile dimensions is herewith furnished.

Evaluation of the effect of orthodontic treatment on the reliability of facial recognition by using three-dimensional model superimposition technique

3D-3D registration of facial models, has great advantages in personal individual identification in forensic medicine. However, orthodontic treatment has brought changes in facial soft and hard tissues, which has a potential effect in personal identification. The aim of the study was to explore whether orthodontic treatment affects 3D-3D facial recognition. A total of 68 patients aged between 18 and 38 were selected (30 subjects with tooth extraction, 38 subjects with non-tooth extraction) and a control group consisting of 30 volunteers without orthodontic treatment was selected from a database of 3dMD facial models. 3dMD facial model acquisition was performed twice for each subject. For both extraction and non-extraction groups, T0 was acquired before treatment, and T1 was acquired at the end of treatment. The time span of 2 times 3dMD facial model acquisition in the control group was about 2 years. 3dMD facial models were then registered onto other 3dMD models belonging to the same and different individuals according to the minimum point-to-point distance, getting a cohort of matches and mismatches. Root mean square (RMS) value of the minimum point-to-point distance between two models was then calculated. The intra- and inter-observer repeatability coefficients were 0.986 (P<0.001) and 0.982 (P<0.001). The absolute technical error of measurement (TEM) value was 0.05 mm and 0.04 mm, and relative technical error of measurement (rTEM) value was 4.91 % and 3.60 %, respectively. Possible significant differences between groups were assessed through independent samples t-test or Mann–Whitney U test (p<0.001). The average RMS value was 1.03±0.33 mm in matches of orthodontic group, 0.67±0.22 mm in matches of control group, and 2.66±0.51 mm in mismatches. An RMS value range of 1.70–1.77 mm could distinguish matches from mismatches in 100 % of cases in the present study. This study showed that orthodontic treatment would not affect 3D-3D facial recognition in adults for the purpose of individual identification in forensic medicine.

Imaging device to measure the reflective and transmissive part of isotropic BSSRDF.

The bidirectional scattering-surface reflectance distribution function (BSSRDF) is a fundamental scattering quantity that characterizes the appearance of translucent materials. The few existing BSSRDF measurement facilities that ensure traceable results have shown the high complexity of such measurement, while only accounting for light scattered in the reflection hemisphere. We have developed an imaging device for measuring both the reflective and the transmissive part of the spectral BSSRDF. The complete analysis of the uncertainties associated with the facility allows relative measurement errors of around 10%. The measurements performed on two translucent samples and primary comparisons with metrological measurements are provided and discussed.

Separated spacecraft attitude algorithm research based on star sensor multi-information fusion

Faced with the problem of poor anti-interference ability of star map recognition technology, in order to improve its anti-interference ability and achieve higher attitude measurement accuracy, this article proposes a star map recognition method based on binary search trees. Star sensors and gyroscopes are combined to solve the problem of the poor real-time performance of star sensors, optimizing the Extended Kalman Filtering (EKF) algorithm. This improves attitude output accuracy and formulates corresponding attitude measurement schemes considering the situation where the star sensor is not visible. The results showed that the improved star pattern recognition algorithm had better adaptability and robustness than the rasterization algorithm, with a high recognition success rate of 95.6% when the star sensor field of view was 3°*3°, while the rasterization algorithm cannot recognize small fields of view. The recognition success rate of the algorithm was 98.6% when the standard deviation of magnitude noise was 2.0 pixels, which was 15.4% higher than the rasterization algorithm. Unlike the EKF algorithm, the optimized EKF algorithm had a higher attitude output accuracy, with an average error of about 0.000 211°; compared to before optimization, it has increased by 93.43%. In the attitude measurement scheme, the relative attitude angle measurement error was small, and the maximum attitude measurement error in the rolling angle direction was 1.8 × 10−3°. This paper adopts a method to achieve high-precision satellite attitude measurement with good adaptability, which can be applied in complex space missions.

Research on SLAM Localization Algorithm for Orchard Dynamic Vision Based on YOLOD-SLAM2

With the development of agriculture, the complexity and dynamism of orchard environments pose challenges to the perception and positioning of inter-row environments for agricultural vehicles. This paper proposes a method for extracting navigation lines and measuring pedestrian obstacles. The improved YOLOv5 algorithm is used to detect tree trunks between left and right rows in orchards. The experimental results show that the average angle deviation of the extracted navigation lines was less than 5 degrees, verifying its accuracy. Due to the variable posture of pedestrians and ineffective camera depth, a distance measurement algorithm based on a four-zone depth comparison is proposed for pedestrian obstacle distance measurement. Experimental results showed that within a range of 6 m, the average relative error of distance measurement did not exceed 1%, and within a range of 9 m, the maximum relative error was 2.03%. The average distance measurement time was 30 ms, which could accurately and quickly achieve pedestrian distance measurement in orchard environments. On the publicly available TUM RGB-D dynamic dataset, YOLOD-SLAM2 significantly reduced the RMSE index of absolute trajectory error compared to the ORB-SLAM2 algorithm, which was less than 0.05 m/s. In actual orchard environments, YOLOD-SLAM2 had a higher degree of agreement between the estimated trajectory and the true trajectory when the vehicle was traveling in straight and circular directions. The RMSE index of the absolute trajectory error was less than 0.03 m/s, and the average tracking time was 47 ms, indicating that the YOLOD-SLAM2 algorithm proposed in this paper could meet the accuracy and real-time requirements of agricultural vehicle positioning in orchard environments.

An ultrafast optical sensor for simultaneous detection of NH3 temperature and concentration based on ultraviolet absorption spectral redshift combined with spectral reconstruction

As a renewable and carbon-free fuel with high energy density, ammonia (NH3) is a promising alternative to fossil fuels. Real-time detection of its temperature and concentration is significant in improving NH3 combustion efficiency. In this paper, we report an optical sensor, which enables simultaneous measurement of temperature and concentration of NH3. The key idea is to acquire the temperature of NH3 through absorption spectral redshift, and determine its concentration by building three-dimensional field map combined with spectral reconstruction. Firstly, we explain the generation mechanism of temperature-induced redshift in ultraviolet differential absorption spectrum based on energy level distribution theory. The relationship between spectral redshift and temperature is then established using spectral autocorrelation algorithm. Secondly, a spectral reconstruction method based on absorbed intensity mapping wavelength is proposed to ensure real-time detection of NH3. Finally, a three-dimensional field map between temperature, reconstruction slope, and concentration is constructed to achieve the detection. Results indicated that the mean relative error in temperature measurement was 1.04 % and the mean absolute error in concentration detection was 0.26 mg/m3 in air. Additionally, the NH3 sensor enabled long-term detection with a detection limit of 15.50 μg/m2, demonstrating its potential to broaden the application of NH3 as a fuel.

Implementation of a low-cost comprehensive pavement inspection system

Assessing the condition of roads is crucial to the maintenance and rehabilitation process as a country's progress is closely linked to its transport infrastructure. Therefore, it is essential to have well-maintained roads and to be able to control and monitor them properly. Technological advancements have transformed the way pavement inspections are carried out. This study presents an innovative approach that combines stereo cameras and a GPS module for efficient and accurate data collection. This integration of low-cost technologies provides a detailed three-dimensional view of pavements, complemented by accurate geospatial information. The experimental results showed that the 3D images of pavement damage had a relative volume measurement error of 0.80 %. Unlike traditional systems such as LIDAR and ground-penetrating radar, which involve more expensive technologies, the proposed method offers a cost-effective solution. This methodology not only simplifies the inspection process but also improves the planning and execution of road maintenance and repair activities. Its low cost makes it a viable option for various projects and applications in road infrastructure.

Design and characterization of a novel turbidity sensor based on quadrature demodulation

Abstract Turbidity is regarded as a comprehensive indicator in water quality monitoring, and the turbidity sensor deployed in the water supply network can record the dynamic changes of water quality in time. However, the weak photoelectric signal from the photodetector contains a quantity of noise. In order to improve signal-to-noise ratio (SNR), a novel on-line turbidity sensor based on quadrature demodulation principle has been proposed in this paper. A near-infrared light-emitting diode (LED) with a wavelength of 860 nm was selected as a stable monochromatic light source, and a photodiode (PD) with an angle of 90° to the incident light from the LED was selected as the photodetector. Using signal modulation and demodulation technology, the weak photoelectric signal extraction, conversion, amplification and output of the turbidity sensor were realized through the effective integration. A corresponding test apparatus of the turbidity sensor was established and experimental results showed that within a 0~5 NTU measurement range, the turbidity sensor had good linearity and stability, the relative measurement error was within ±1% and the limit of detection could reach as low as 0.0049 NTU. The developed turbidity sensor has good detection performance and can meet the needs of low turbidity detection of drinking water.&amp;#xD;

Increasing the accuracy of electrical energy accounting at reduced load

The object of research is a three-phase electricity metering unit, which includes a digital meter and measuring current transformers. The reduction of non-technological energy losses is restrained due to the insufficient accuracy of the accounting of electric energy in distribution power networks under a reduced load current of the metering unit. The possibility of representing the dependence of the relative error of electricity measurement on current values by a fuzzy function at reduced load has been confirmed. The boundaries of such a function are approximated with sufficient accuracy by the sum of two exponents, which is explained by its significant nonlinearity in the range of reduced current. The proposed EMRL software allows to estimate the real consumption and the most possible level of underaccounting based on the array of electricity meter readings. The accuracy of estimating by the EMRL the amount of electricity consumed with a probability of 0.7 can be estimated with a relative error not exceeding 2 %. The probability of psychophysical assessments of the accuracy of EMRL «very good» and «good» is at least 0.833. The trend of a significant decrease in the relative value of underaccounting with an increase in the level of electricity consumption was revealed. With a daily consumption of up to 10 kW·h, the amount of underaccounting can reach 18 %, and with a consumption of more than 20 kW·h, it does not exceed 6 %. The adequacy of the results of estimating the amount of consumed electricity at reduced load using the EMRL was confirmed by experimental data at a significance level of 0.05. The software capabilities allow to increase the accuracy of the accounting of electrical energy in distribution networks with a reduced load current of the metering unit. The program can be used as part of automated systems of commercial electricity metering or advanced metering infrastructure to determine the most possible underaccounting due to the operation of metering units at a reduced load

Nighttime measurement errors in the runway visual range and optimization of runway lighting systems

Considering that existing theories consider runway lights as markers, neglecting their roles as lighting devices that increase sky background luminance and thus affect runway visual range (RVR) measurements, the actual values of sky background luminance observed by pilots during take-off are calculated and compared with the instrumental RVR measurements in this paper. The RVR measurement errors due to the installation positions of these devices are subsequently analysed. Furthermore, the impacts of different runway light types and positions on sky background luminance are explored, leading to the proposal of an optimized runway lighting scheme. The results show that the representation of sky background luminance sensor data in RVR measurement systems for nighttime airport runways is limited. A deviation of over 90% exists between the sky background luminance of the runway entrance and the RVR measurement equipment location, resulting in a relative error in RVR measurements of over 18%. The centreline and entrance lights contribute significantly to the total sky background luminance. Lights closer to the observer contribute more luminance. In the second proposed scheme, modifying the lighting range to 600 m yields the highest RVR improvement, with an optimization magnitude of over 5% across all visibility ranges.

Design of control system for seawater total alkalinity in-situ analyzer

Abstract To address the inadequacies of traditional seawater total alkalinity detection methods, which fail to meet the demands for continuous on-site monitoring at sea, a seawater total alkalinity analyzer based on single-point titration and spectrophotometry has been developed. This paper focuses on the design of the analyzer’s system architecture, hardware, and control systems. The instrument adopts a modular design concept and utilizes microfluidic technology for detection, with the STM32 serving as the control core. Through the design of the control system for the detection instrument, accurate in-situ automatic measurement of seawater total alkalinity is achieved. Laboratory testing and analysis demonstrate that the instrument’s measurement relative error is less than 1%, satisfying the accuracy requirements for seawater total alkalinity measurement. This research provides critical monitoring data support for a timely understanding of marine ecological environments and the marine carbon cycle.

Parameter Optimization of Electromagnetic Sensing and Driving Scheme of a Compact Falling-body Viscometer

In order to achieve automatic fast reset of the falling-body (FB) viscometer, and reduce the volume of the device and the amount of sample used, this paper proposes to use a single coil to reset the FB and use another single coil to measure the FB position. In this paper, the characterization ability of the sensing coil impedance to the FB position and its influencing factors, and the reset ability of the driving coil to the FB and its enhancement factors, are studied via electromagnetic finite element simulations and experiments. There is a linear zone between the FB position and the sensing coil impedance, with the slope being largest. The lower limit of the FB motion should be designed in this linear zone to accurately determine the moment when the FB reaches the lower limit point. Increase in the FB height, and in the number of turns of the sensing coil, and decrease in the wire diameter are beneficial to the FB positioning. There is a maximum force point when the FB approaches the driving coil, and the FB’s motion range needs to cover this point for reliable reset. Using iron plugs allows the FB to obtain greater electromagnetic attraction and ensure its successful reset. Experimental results show that the device requires only 1 mL of sample to measure liquid viscosity. For 9.5-1265 mPa⋅s dimethyl silicone oil, the average absolute value of the relative measurement error is 0.22% and the maximum value is 4.3%.

Optimizing guided wave propagation for sensitive axial stress measurement in steel pipes

Steel pipe structures are commonly used in the industrial field. Stress-induced structural failures can have a significant impact on equipment safety. Therefore, effective stress monitoring is a crucial area of research. Constrained by the multi-frequency and multi-modal characteristics of ultrasonic guided waves, the limitations of traditional stress measurement methods based on these waves lie in the lack of a systematic analysis of the impact of multi-modal fused signals on stress measurement. To explore the optimal guided wave stress measurement strategy, a mathematical model for the propagation of longitudinal guided waves in prestressed steel pipes, based on the principles of acoustoelasticity, is proposed. Using this model, the sensitivity of stress to each sub-mode of the longitudinal guided waves is analyzed, leading to the identification of the optimal guided wave mode (L(0,2) mode) and frequency range. In order to avoid the excitation of other low-sensitivity modes, further analysis of the optimal parameters for the piezoelectric array is conducted. Simulation results indicate that the designed transducer can effectively excite the target mode with high purity. The stress sensitivity of the target mode was experimentally determined to be −2.5×10−5MPa−1, which closely aligns with the theoretical results. Comparative analysis of the experiments emphasizes the influence of modal control on measurement outcomes. By selecting and controlling the appropriate mode, the maximum relative error in stress measurement is observed to be 5%.

Photon‐Counting 3D Velocimetry Empowered by OAM‐Based Multi‐Point Doppler Effect

AbstractVelocimetry of a motion target within 3‐D space is highly desirable in numerous applicable realms, ranging from explosion and shock wave physics, aerospace engineering to astronomical surveys. However, it is challenging to achieve synchronous, real‐time, and photon‐counting 3‐D velocimetry in modern frameworks as they either require separate multi‐directional detections, and cumbersome calculation processes or are confined to achieve in situ measurements. Here, a new conceptual paradigm is proposed to circumvent these constraints using orbital‐angular‐momentum (OAM)‐driven multi‐point Doppler effect at the photon‐counting level. This scheme, emanating from a single‐direction launch of an on‐demand engineered sequence OAM light mode onto a motion surface, enables simultaneous and independent detections of time‐varying Doppler photon‐count events from three orthogonal echo light paths. Concretely, at the range of motion velocity of 0.25–0.5 ms−1, the relative measurement errors of this proof‐of‐principle prototype are below 1.5%, thus achieving high‐accuracy 3‐D velocimetry at the photon‐counting level for the first time. The exploration of the OAM‐photon‐counting 3‐D velocimetry techniques provides unprecedented advantages in potential applications of synchronous, real‐time, high‐efficiency, and long‐range quantum lidar.

Design of Digital Lock-in Amplifier based on FPGA

Abstract In order to achieve high-precision measurement of strong noise background signal, a digital lock-in amplifier based on field programmable gate array (FPGA) is designed. The core of the lock-in amplifier consists of a signal processing module, an acquisition and quantisation module and a digital demodulation module. The signal processing module completes the amplification and gain adjustment of the input signal and modulates the signal to be measured into a specific frequency band, which is digitally quantised by the analog to digital converter (ADC) after anti-aliasing filtering and transmitted to the FPGA, and the signal to be measured is extracted by phase-sensitive detection through the mutual-correlation digital demodulation method, thus realising the detection of the weak signal. The performance of the lock-in amplifier is evaluated with a signal generator as the input source, and the results show that the signal-to-noise improvement ratio is 35dB and the relative measurement error of voltage is less than 0.03% after the digital lock-in amplifier is processed, which accurately restores the direct current (DC) signals in the noisy background, and provides an effective technical way for the accurate measurement of weak signals.