CN110006413B

CN110006413B - Mobile terminal signal processing method of R-LATs system based on FPGA-ARM embedded system

Info

Publication number: CN110006413B
Application number: CN201910357258.XA
Authority: CN
Inventors: 贾康; 邵山; 刘志刚; 苏文军; 柯健镪; 孙庆龙; 洪军
Original assignee: Xian Jiaotong University
Current assignee: Xian Jiaotong University
Priority date: 2019-04-29
Filing date: 2019-04-29
Publication date: 2020-04-28
Anticipated expiration: 2039-04-29
Also published as: CN110006413A

Abstract

The invention provides a signal processing method of a mobile terminal of an R-LATs system based on an FPGA-ARM embedded system, which has the advantages of reasonable design, small signal error and high real-time and accuracy of signal processing and calculation of a photoelectric sensor. The method comprises the steps of identifying and extracting signals; subdividing each pulse sequence, and recording the rising edge time value and the pulse width of each pulse in real time; respectively tracing the planar pulse signals sent by a plurality of theodolites according to the recorded rising edge time value and the pulse width of each pulse through a plurality of threads arranged in parallel in the FPGA, and identifying and extracting the characteristic information of the pulse signals of different target laser theodolites in parallel; each thread only traces the source of the plane pulse signal sent by one theodolite; calculating and processing signals; the ARM receives the characteristic information identified and extracted by the FPGA, and calculates the time value when the optical plane scans to the center of the sensor; then, the scanning time value is put into different arrays for subsequent processing according to the corresponding mark signal.

Description

R-LATs system mobile terminal signal processing method based on FPGA-ARM embedded system

Technical Field

The invention belongs to the field of large-size space measurement, relates to large-scale rotating laser theodolite measurement network (R-LATs) networking measurement, and particularly relates to a signal processing method for a mobile terminal of an R-LATs system based on an FPGA-ARM embedded system.

Background

The rotating laser theodolite network (R-LATs) is an important method for measuring large-size space, can realize the measurement space expansion of any size through reasonable station arrangement, can keep the measurement precision within +/-0.2 mm, and is widely applied to the aerospace and military fields of aircraft manufacturing, ship manufacturing, large-scale antenna manufacturing and the like at present.

In the R-LATs work, each rotating laser theodolite emits two fan-shaped plane lights with an included angle, and rotates at a constant speed according to a specified speed, and simultaneously, a pulse light covering the whole space is emitted at the zero point of each rotating angle. Thus, the three light planes are swept across a photosensor P in space to obtain three time trigger signals. At this time, based on the three time signals and the rotating speed of the rotating laser theodolite, a unique ray L in the space can be determined, which passes through the laser surface emission center point O of the rotating laser theodolite and the photoelectric sensor P. Thus, when there are more than two rotating laser theodolites, their spatial straight lines L meet at a point in space, which is the spatial position point of the sensor P. Therefore, after the space relative position of the rotary laser theodolite is determined, the coordinate measurement of the photoelectric sensor in a large-size space can be realized.

In order to meet the requirement of large-space range measurement, tens of rotating laser theodolite devices are required to be placed in space, and a photoelectric sensor fixed on a measurement object receives laser plane signals emitted by the devices and performs signal feature identification and extraction. The rotating speeds of the rotating laser theodolite devices are different, so that the photoelectric sensor can correctly identify the theodolite device corresponding to each pulse, and the accurate calculation of the photoelectric sensor coordinate in the space by the R-LATs measuring network is realized.

The more the number of the rotary laser theodolites is, the more the number of the optical plane signals which need to be effectively identified by the photoelectric sensor is, which greatly increases the processing task of the photoelectric sensor, the complex and heavy task of identifying the optical plane signals by the photoelectric sensor causes low measurement efficiency of an R-LATs system, and the calculation efficiency and accuracy based on the photoelectric sensor are low; meanwhile, the probability of superposition of different transmitter plane signals at the sensor is increased, the optical aliasing phenomenon is serious, and the large error of the signals is directly increased. The method has great influence on real-time performance and accuracy of the R-LATs measurement network.

Disclosure of Invention

Aiming at the problems in the prior art, the invention provides the R-LATs system mobile terminal signal processing method based on the FPGA-ARM embedded system, which has the advantages of reasonable design, small signal error and high real-time and accuracy of photoelectric sensor signal processing and calculation.

The invention is realized by the following technical scheme:

the R-LATs system mobile terminal signal processing method based on the FPGA-ARM embedded system comprises the following steps,

identifying and extracting signals;

subdividing each pulse sequence, and recording the rising edge time value and the pulse width of each pulse in real time; respectively tracing the planar pulse signals sent by a plurality of theodolites according to the recorded rising edge time value and the pulse width of each pulse through a plurality of threads arranged in parallel in the FPGA, and identifying and extracting the characteristic information of the pulse signals of different target laser theodolites in parallel; each thread only traces the source of the plane pulse signal sent by one theodolite;

calculating and processing signals;

the ARM receives the characteristic information identified and extracted by the FPGA, and calculates the time value when the optical plane scans to the center of the sensor; then, the scanning time value is put into different arrays for subsequent processing according to the corresponding mark signal.

Preferably, when each pulse sequence is subdivided, the pulse sequence is counted by using a clock provided by a high-frequency crystal oscillator signal, a time value corresponding to a rising edge of each pulse signal and a count value corresponding to a pulse width of each pulse signal are used as useful characteristic information, and the time value corresponding to the rising edge is stored in an array DATA [ ] not less than 3 times the number of transmitters.

Preferably, each thread in the FPGA traces the source of the pulse signal corresponding to one theodolite according to the following steps;

step 1, identifying a dynamic stage;

step 1.1: comparing the rising edge time value of the current pulse with the time value recorded during the fine time to trace the source of the signal;

step 1.2: recording a rising edge moment value ris [ i ], a time interval true value ris [ i ] -ris [ j ] and a pulse width value of the pulse after tracing, and simultaneously generating a marking signal to indicate the serial number of the laser theodolite to which the current pulse belongs;

step 2, confirming whether the quasi-static measurement state is entered, if not, repeatedly executing the step 1; if yes, executing step 3;

step 3, identifying a quasi-static stage;

step 3.1: comparing the rising edge time value of the current pulse with the time value recorded during the fine time to trace the source of the signal;

step 3.2: recording a rising edge moment value ris [ i ], a time interval true value ris [ i ] -ris [ j ] and a pulse width value of the pulse after tracing, and taking the values as initial values of signals in a quasi-static measurement stage;

step 3.3: judging whether a pulse exists at the predicted moment or not by taking the initial value plus the rising edge time interval characteristic of the pulse signal to be identified in the quasi-static stage as the predicted moment when the next pulse to be traced appears, if so, recording the rising edge moment value of the pulse at the predicted moment and executing the step 3.4; if not, returning to the step 3.1;

step 3.4: judging whether light aliasing occurs at the prediction moment, if so, executing a step 3.5; if not, the pulse tracing corresponding to the predicted time is finished, and the step 3.6 is executed;

step 3.5: performing aliasing optical signal characteristic reconstruction by using the periodic characteristics;

step 3.6: and (3) recording the rising edge time value, the time interval true value and the pulse width value, simultaneously generating a mark signal to indicate the serial number of the laser theodolite to which the current pulse belongs, using the rising edge time value as an initial value for next judgment, and simultaneously returning to the step 3.3.

Further, in step 1.1, the tracing manner of the pulse signal is as follows: sequentially comparing a rising edge time value ris [ i ] of the pulse signal acquired in real time with a previously recorded time value: if a time value ris [ j ] exists, such that ris [ j ] + T-T-a < ris [ i ] < ris [ j ] + T + T + a, and a time value ris [ k ] exists, such that ris [ k ] + T-T-a < ris [ j ] < ris [ k ] + T + T + a, the tracing of the current pulse signal is considered to be completed; in the formula, T is the rotation period of the theodolite, T is the system error generated by the rotation of the theodolite, and a is the time interval change threshold caused by the movement of the object to be measured.

Further, in step 3.1, the tracing manner of the pulse signal is as follows: sequentially comparing a rising edge time value ris [ i ] of the pulse signal acquired in real time with a previously recorded time value: if a time value ris [ j ] exists, so that ris [ j ] + T-T < ris [ i ] < ris [ j ] + T + T, and a time value ris [ k ] exists, so that ris [ k ] + T-T < ris [ j ] < ris [ k ] + T + T, the tracing of the current pulse signal is considered to be completed; in the formula, T is the rotation period of the theodolite, and the value of T is the system error generated by the rotation of the theodolite.

Preferably, in step 3.4, the method for determining whether the signal is subjected to optical aliasing is as follows: recording the rising edge time value ris _ x of the time pulse, and calculating the time interval true value ris _ x-ris [ i ], and then judging: if ris _ x-ris [ i ] < T-T, judging that the light superposition occurs; in the formula, T is the rotation period of the theodolite, and the value of T is the system error generated by the rotation of the theodolite.

Preferably, in step 3.5, the optical aliasing signal reconstruction method is performed as follows: using the average value T _ a of the time interval truth values of the previous 3 moments as a predicted value of the time interval value of the moment, wherein ris [ j ] + T _ a is a reconstructed rising edge moment value; and taking the mean value of the pulse width values of the first 3 traced pulses as the pulse width value of the current pulse, and recording the mean value as the reconstructed pulse width true value.

Compared with the prior art, the invention has the following beneficial technical effects:

the invention relates to a signal processing method of a mobile terminal of an R-LATs system based on an FPGA-ARM embedded system, which is used for carrying out high-precision identification and tracing on photoelectric sensor signals in a parallel processing mode, and improves the real-time performance of system measurement and the expandability of the system.

Furthermore, the period time is used as an identification mark in the tracing process, and different period thresholds are adopted in the dynamic stage and the quasi-static stage, so that the measurement precision of the system is improved.

Furthermore, the identification and reconstruction of the aliasing light are carried out in the quasi-static stage at the same time, so that the accuracy of signal identification is ensured.

Drawings

FIG. 1 is a schematic diagram of the operation of a large-scale R-LATs measurement net.

Fig. 2 is a flow chart of a method for processing signals of a mobile terminal of the system according to the embodiment of the invention.

Detailed Description

The present invention will now be described in further detail with reference to specific examples, which are intended to be illustrative, but not limiting, of the invention.

The invention relates to a signal processing method for a mobile terminal of an R _ LATs system based on an FPGA-ARM embedded system, which mainly comprises two aspects of identification and extraction of signals at the FPGA terminal and calculation and processing of identified signals at the ARM terminal, and specifically comprises the following steps:

first, signal recognition and extraction.

The optical plane pulse sequence is subdivided by counting with a clock provided by a high frequency crystal oscillator signal. The time value corresponding to the rising edge of each pulse signal and the counting value corresponding to the pulse width of each pulse signal are useful characteristic information, and are stored in an array DATA which is not less than 3 times of the number of transmitters.

The characteristics of the pulse signals of the laser theodolite with different targets are identified and extracted in parallel by a method suitable for FPGA multi-core parallel. Here, each thread implicit in the FPGA only traces the source of the planar pulse signal emitted by one theodolite. The number of parallel threads in the specific operation is equal to the number of theodolites needing tracing.

According to different characteristics of the system facing to dynamic measurement tasks and quasi-static measurement tasks, the tracing method executed by each thread is divided into two parts of dynamic identification and quasi-static identification:

1.1 dynamic recognition Algorithm

In the dynamic measurement stage, the FPGA works as follows:

1) firstly, amplifying a confidence interval of each thread by n clock cycles according to the cycle time characteristics of the laser theodolite needing to trace the source and combining with the dynamic running speed to obtain time interval characteristics T1, T2, T3 … … and the like of a dynamic measurement stage;

2) each thread compares the rising edge time T of the current pulse signal with the characteristic information in the DATA array one by one, if the DATA array has the information which accords with the time interval characteristic of the dynamic measurement stage, the tracing of the current pulse signal is completed, and the time T is recorded;

3) generating a mark signal to indicate the information of the pulse signal source at the time T;

4) recording a real value T _ z of a pulse signal time interval after tracing in real time, then comparing the real value with a theodolite rotation period value T _ p, and if the difference value of the T _ z and the T _ p in 3 continuous periods is within the theodolite system error range, judging that the theodolite rotation period value T _ z enters a quasi-static operation stage;

1.2 quasi-static recognition Algorithm

After conversion to quasi-static measurements, the FPGA performs the following operations:

1) firstly, according to the periodic time characteristics of the laser theodolite needing tracing, each thread obtains time interval characteristics t1, t2, t3 … … and the like with a certain confidence interval in a quasi-static measurement stage by combining the systematic error generated by theodolite rotation;

2) each thread compares the rising edge time t of the current pulse signal with the characteristic information in the DATA array one by one, if the DATA array has the information which accords with the time interval characteristic of the static measurement stage, the tracing of the current pulse signal is completed, and the time t is recorded as the initial value of the signal of the quasi-static measurement stage;

3) adding t to the value of the time interval characteristic to obtain time t-next, judging whether the pulse signal at the time of the t-next is high level, if so, carrying out the next step, and if not, returning to the step 2);

4) and comparing the current pulse with the previous traced pulse rising edge moment to obtain a time interval true value. Then, comparing the time interval truth value with the cycle information, and judging whether the pulse at the moment generates optical aliasing (when the sensor receives more than two theodolite pulse signals at the same time, the original rising edge signal is covered, namely the optical aliasing is generated): if not, updating T, generating a mark signal, indicating the source of the pulse signal at the time T, and returning to the step 3); if so, carrying out the next step;

5) separating and extracting the current aliasing light by using the traced pulse signal characteristics, recording a predicted value as a true value, updating t, and returning to the step 3);

secondly, signal calculation and processing.

The ARM receives data from the FPGA, and specifically comprises a rising edge time value t of a traced signal corresponding to each transmitter of each thread, a pulse width count value a and a mark signal p. Then, taking t + a/2 as the real time when each light plane scans to the center of the sensor, and putting the real time of each light plane pulse signal of each transmitter into different arrays for subsequent calculation processing according to the corresponding mark signal p.

Take the measurement scenario shown in fig. 1 as an example.

1. Parallel recognition and extraction of signals

And timing by a 150MHz clock provided by an FPGA external crystal oscillator, subdividing the pulse sequence, and recording a time value date _ t corresponding to the rising edge time of each pulse signal and a count value date _ w corresponding to the pulse width of each pulse. Then, the rising edge time value is sequentially put into an array DATA [ ] not less than 3 times of the number of transmitters, and DATA in the array is updated in real time to ensure that the array does not overflow. The array is used for identification of a subsequent light plane trigger pulse signal.

The rotating speeds of the laser theodolites are different, the periods of the corresponding light plane trigger pulses are different, and the corresponding pulse rising edge time intervals are also different, so that the light plane signals can be traced. In the process of tracing the source of the pulse signals, in order to improve the efficiency, a multi-thread parallel processing mode is designed in the FPGA platform to execute the algorithm, wherein each thread only traces the source of the pulse signals corresponding to a certain theodolite, so that the identification and extraction of the pulse signals of a plurality of target laser theodolites are completed, the efficiency is obviously improved, and the hardware operation complexity is reduced.

R-LARs are widely applied to real-time coordinate measurement tasks of moving objects, and targets such as assembly butt joint, fixed-point stop and the like are completed. In practical application, the method is generally divided into two stages, namely a dynamic operation stage and a quasi-static operation stage. The dynamic operation stage has higher requirement on the real-time performance of measurement. And the quasi-static phase puts higher requirements on the measurement accuracy.

Specifically, the following idea only explains the tracing algorithm implementation of any transmitter. Due to the adoption of parallel processing, other transmitter tracing methods are consistent with the method.

1.1 dynamic phase identification method

The characteristic size of the rising edge time interval of the pulse signal to be identified is T +/-T +/-a, wherein T is a corresponding period value of the laser theodolite to be traced at a set rotating speed; t is the system error caused by the hardware structure; a is the time interval change threshold caused by the movement of the object to be measured;

in the dynamic measurement process, the rising edge time value ris [ i ] of the pulse signal acquired in real time is sequentially compared with the time values recorded in the array DATA [ ]: if a time value ris [ j ] exists, such that ris [ j ] + T-a < ris [ i ] < ris [ j ] + T + a, and a time value ris [ k ] exists, such that ris [ k ] + T-a < ris [ j ] < ris [ k ] + T + a, the tracing of the current pulse signal is considered to be completed. At this time, the rising edge time value ris [ i ], the time interval true value ris [ i ] -ris [ j ] and the pulse width value of the pulse are recorded, and a marking signal is generated to indicate the serial number of the laser theodolite to which the current pulse belongs.

1.2 quasi-static phase recognition Algorithm

The measurement accuracy requirement in the quasi-static stage measurement stage is higher, and an optimization method scheme matched with the requirement needs to be adopted at the moment as follows:

the characteristic size of the rising edge time interval of the pulse signal to be identified is T +/-T, wherein T is a corresponding period value of the laser theodolite to be traced at a set rotating speed; and t is the system error caused by the hardware structure. Because the object to be measured is in a quasi-static measurement state, the time interval fluctuation caused by the speed is negligible.

Based on the size fluctuation stability degree of a time interval truth value in a certain time period, whether the system enters a quasi-static measurement state stage or not is determined, and the specific method comprises the following steps: and recording a true value T _ z of a pulse signal time interval after tracing in real time, comparing the true value with a theodolite rotation period value T _ p, and judging that the theodolite enters a quasi-static operation stage if the difference value of the T _ z and the T _ p in 3 periods is within the error range of a theodolite system. After entering a quasi-static measurement state is confirmed, a first rising edge moment value after the state change is recorded and used as an initial value of quasi-static stage tracing, and then if the conditions such as signal loss occur, the initial value of the rising edge needs to be obtained again.

The initial value acquisition method comprises the following steps:

comparing the rising edge time ris [ i ] of the current pulse signal with the characteristic information in the array DATA one by one: if a time value ris [ j ] exists, such that ris [ j ] + T-T < ris [ i ] < ris [ j ] + T + T, and a time value ris [ k ] exists, such that ris [ k ] + T-T < ris [ j ] < ris [ k ] + T + T, the tracing of the current pulse signal is considered to be completed. And recording the rising edge time ris [ i ], the time interval true value ris [ i ] -ris [ j ] and the pulse width value, and taking the values as initial values of signals in the quasi-static measurement stage.

And adding T + T to the initial rising edge time value to obtain a predicted value of the next pulse time. At this prediction time, that is, at ris [ i ] + T, it is determined whether or not the pulse signal is at a high level. This time is discussed in two cases:

a. if it is low level

The predicted time is blocked or has other conditions, so that the pulse is lost; at this time, the tracing initial value needs to be identified and extracted again.

b. If it is high level

At this time, the pulse is not lost, but optical aliasing may occur, which causes some optical plane characteristic signals to be lost or distorted, and further identification and extraction are required. The method utilizes the time interval characteristics to judge whether the light aliasing occurs.

First, recording a rising edge time value ris _ x of the predicted time pulse, and calculating a time interval true value ris _ x-ris [ i ], at this time, judging: if ris _ x-ris [ i ] < T-T, it is determined that optical superposition occurs. At this time, aliasing optical signal characteristic reconstruction is required: using the average value T _ a of the time interval truth values of the previous 3 moments as a predicted value of the time interval value of the moment, wherein ris [ j ] + T _ a is a reconstructed rising edge moment value; taking the mean value of the pulse width values of the first 3 traced pulses as the pulse width value of the current pulse, and taking the mean value as the reconstructed pulse width true value to record; if no optical aliasing occurs, the tracing of the corresponding pulse at the moment is finished, a rising edge moment value, a time interval true value and a pulse width value are recorded, and the rising edge moment value is used as an initial value for next judgment.

2. Signal calculation and processing

And sending the rising edge time value of the traced pulse signal to an ARM for further data processing. The ARM receives characteristic information from the FPGA, wherein the characteristic information comprises a rising edge time value, a pulse width value and a mark signal corresponding to the time value. And the ARM end calculates the time value when the light plane scans the center of the sensor according to the rising edge time value and the pulse width value. Then, the scanning time values are put into different arrays according to the corresponding flag signals.

Aiming at different application occasions and specific configuration modes of the R-LATs network, the ARM end can be organized to realize different functions: if the calculation can be carried out according to the calculation packaging algorithm, and the calculation result is transmitted to the handheld end; or the obtained pulse characteristic information is packaged and transmitted to an upper computer end for calculation, so that the processing of large-scale R-LATs measurement network mobile end signals is realized.

Claims

1. based on the R-LATs system mobile terminal signal processing method of FPGA-ARM embedded system, it is characterized in that, comprises,

Signal identification and extraction;

Subdivide each pulse sequence, and record the rising edge time value and pulse width of each pulse in real time; through multiple threads set in parallel in the FPGA, according to the recorded rising edge time value and pulse width of each pulse, respectively Trace the source of the plane pulse signals sent by multiple theodolites, and identify and extract the characteristic information of different target laser theodolite pulse signals in parallel; each thread only traces the source of the plane pulse signal sent by one theodolite; through parallel processing , the cycle time is used as the identification mark, and different cycle thresholds are used in the dynamic stage and the quasi-static stage;

When subdividing each pulse sequence, the clock provided by the high-frequency crystal oscillator signal is used to count, and the time value corresponding to the rising edge of each pulse signal and the count value corresponding to the pulse width of each pulse signal are used as useful feature information. The time value corresponding to the rising edge is stored in the array DATA[] which is not less than 3 times the number of transmitters;

Calculation and processing of signals;

The feature information identified and extracted by the FPGA is received by the ARM, and the time value when the light plane is scanned to the center of the sensor is calculated; then, according to the corresponding sign signal, the scan time value is put into different arrays for subsequent processing.

2. the R-LATs system mobile terminal signal processing method based on FPGA-ARM embedded system according to claim 1, is characterized in that, each thread in FPGA is traced to the pulse signal corresponding to a theodolite according to the following steps;

Step 1, dynamic stage identification;

Step 1.1: Compare the time value of the rising edge of the current pulse with the time value recorded during subdivision to trace the source of the signal;

Step 1.2: Record the rising edge time value ris[i], the time interval true value ris[i]-ris[j] and the pulse width value of the pulse after the source tracing, and generate a flag signal to indicate the serial number of the laser theodolite to which the current pulse belongs;

Step 2, confirm whether to enter the quasi-static measurement state, if not, repeat step 1; if so, execute step 3;

Step 3, quasi-static stage identification;

Step 3.1: Compare the time value of the rising edge of the current pulse with the time value recorded during subdivision to trace the source of the signal;

Step 3.2: Record the rising edge time value ris[i], the time interval true value ris[i]-ris[j] and the pulse width value of the pulse after the source tracing, and use it as the initial value of the signal in the quasi-static measurement stage;

Step 3.3: By adding the initial value and the time interval characteristic of the rising edge of the pulse signal to be identified in the quasi-static stage, as the predicted time of the next pulse to be traced, determine whether there is a pulse at the predicted time, and if so, record the rising edge time of the pulse at the predicted time value and go to step 3.4; if not, return to step 3.1;

Step 3.4: Determine whether optical aliasing occurs at the predicted time, and if so, go to Step 3.5; if not, then the pulse source tracing corresponding to the predicted time is completed, and go to Step 3.6;

Step 3.5: Use periodic features to reconstruct aliased optical signal features;

Step 3.6: Record the rising edge time value, the true value of the time interval and the pulse width value, and at the same time generate a flag signal to indicate the serial number of the laser theodolite to which the current pulse belongs, and use the rising edge time value as the initial value of the next judgment, and return to step 3.3 at the same time.

3. the R-LATs system mobile terminal signal processing method based on FPGA-ARM embedded system according to claim 2, is characterized in that, in step 1.1, pulse signal traceability mode is as follows: by the pulse signal rising edge moment of real-time acquisition The value ris[i] is compared with the previously recorded time value in turn: if there is a time value ris[j] such that ris[j]+T-t-a<ris[i]<ris[j]+T+t+a, there are A time value ris[k] makes ris[k]+T-t-a<ris[j]<ris[k]+T+t+a, that is, it is considered that the current pulse signal traceability is completed; in the formula, T is the theodolite rotation period, t value Through the systematic error generated by the theodolite rotation, the value of a is the threshold value of the time interval change caused by the movement of the object to be measured.

4. the R-LATs system mobile terminal signal processing method based on FPGA-ARM embedded system according to claim 2, is characterized in that, in step 3.1, pulse signal traceability mode is as follows: by the pulse signal rising edge moment of real-time acquisition The value ris[i] is compared with the previously recorded time value in turn: if there is a time value ris[j] such that ris[j]+T-t<ris[i]<ris[j]+T+t, there is also a time The value ris[k] is such that ris[k]+T-t<ris[j]<ris[k]+T+t, that is, it is considered that the current pulse signal traceability is completed; in the formula, T is the theodolite rotation period, and the t value is generated by the theodolite rotation system error.

5. the R-LATs system mobile terminal signal processing method based on FPGA-ARM embedded system according to claim 1, is characterized in that, in step 3.4, the mode of judging whether optical aliasing occurs in the signal is as follows: record this moment The rising edge time value of the pulse is ris_x, and the true value of the time interval ris_x-ris[i] is calculated. At this time, the judgment is made: if ris_x-ris[i]<T-t, it is judged that the light overlap occurs; in the formula, T is the theodolite Rotation period, t value System error generated by the theodolite rotation.

6. the R-LATs system mobile terminal signal processing method based on FPGA-ARM embedded system according to claim 1, is characterized in that, in step 3.5, carry out optical aliasing signal reconstruction mode as follows: utilize the first 3 moments of The average value T_a of the true value of the time interval is used as the predicted value of the time interval value at this moment, and ris[j]+T_a is the value of the rising edge after reconstruction; the average value of the pulse width values of the first three traced pulses is used as the current pulse value. The pulse width value is recorded as the reconstructed true value of the pulse width.