A dynamic code ring phase discrimination method

Abstract

A dynamic code loop phase discrimination method comprises: S1, obtaining a segment of received spread spectrum data required by a current epoch, and performing carrier frequency and phase compensation on the received spread spectrum data to obtain current epoch loop input data; S2, a control unit selects different code loop phase discrimination modes according to a scene or a loop state; S3, a local code generation unit generates multiple local spread spectrum codes; S4, a correlation unit completes multiple correlations of data and local spread spectrum codes, and a phase discrimination unit obtains a current epoch code loop phase discrimination value based on a correlation result; S5, a dynamic filtering unit dynamically adjusts filtering coefficients based on multiple correlation integrals of the correlation unit, dynamically filters the code loop phase discrimination value, and outputs a code phase; S6, a feedback unit converts the code phase into an offset sampling point and feeds back the offset sampling point to control selection of next epoch received spread spectrum data. The method can accurately perform code loop phase discrimination and accelerate code loop convergence in a multipath and non-multipath scene, improve reliability of signal tracking, and improve system reception performance.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of spread spectrum communication technology, in particular to a dynamic code ring phase discrimination method. BACKGROUND

[0002] Spread spectrum communication, short for spread spectrum communication, spreads the original signal to be transmitted to a wide frequency band using a specific pseudo-random code (spreading sequence) and transmits it in the channel. The receiving end recovers the original data by correlating with the corresponding spreading sequence. There are mainly direct sequence spread spectrum, time hopping spread spectrum, frequency hopping spread spectrum, etc. In order to correctly demodulate and recover the original data, the direct spread spectrum receiver should first synchronize the signal. The synchronization process of the spread spectrum system receiver is generally divided into two steps: acquisition and tracking. Acquisition is a coarse synchronization process, and the purpose is to obtain a rough carrier frequency and phase, but the acquisition accuracy is not enough to achieve correct demodulation of data, and more accurate carrier frequency and phase need to be obtained by tracking and pseudo code phase. After successful acquisition, the state is passed to tracking, at this time the carrier Doppler error is within the tracking discriminator range, and the pseudo code phase difference is within 1 / 2 chip. Tracking takes this state as the initial condition to perform more accurate synchronization.

[0003] The tracking loop in the direct sequence spread spectrum receiver is usually composed of a code ring and a carrier ring, which are closely related and support each other to complete various baseband digital signal processing tasks. The carrier ring includes a frequency lock loop and a phase lock loop, which is responsible for stripping the locally reproduced carrier. The code ring is responsible for the consistency of the local code and the received code phase to achieve pseudo code stripping. The input data is correlated with the local code, and the output signal is correlated with the code phase difference. The output is input to the discriminator, and the discriminator output is low-pass filtered to filter out loop noise. The filtered code phase value forms a feedback used to adjust the local code or received data, which is the code ring.

[0004] When the signal is a single main path transmission signal, the phase discrimination accuracy is mainly affected by noise. When in a multipath scenario, due to different transmission paths, direct and reflected, scattered signals exist relative phase and delay, and amplitude characteristics exist differences, the received loop signal is distorted, the pseudo code autocorrelation function curve is distorted, and the zero crossing point is shifted, which affects the code phase discrimination accuracy. Therefore, it is required that the design of the code ring discriminator should as far as possible reduce the phase discrimination error caused by multipath, and the code ring filter should quickly realize code phase synchronization and make the code ring converge to ensure the stability of the code ring operation, thereby improving the system reception performance. SUMMARY

[0005] With the wide application of direct spread spectrum receiver, the synchronization of received data is improved, the acquisition completes the coarse synchronization process, and the coarse carrier frequency and phase are obtained, but the acquisition accuracy is not enough to complete the data demodulation, generally the carrier loop and code loop work together to complete more accurate carrier frequency and phase synchronization and pseudo code phase synchronization, and the phase discriminator can accurately and stably discriminate in various scenes, and quickly converge the loop to promote code phase synchronization, and further promote carrier synchronization. In order to accurately discriminate the code loop in the multipath and non-multipath scenes and speed up the code phase synchronization, the application provides a dynamic code loop phase discrimination method, which is used in the code loop and carrier loop working together, and finally realizes the accurate synchronization of carrier frequency and phase, and code phase, correctly recovers the modulation information contained in the spread spectrum code, improves the reliability of signal tracking, reduces the bit error rate, and improves the system receiving performance.

[0006] In order to achieve the above purpose, the technical scheme of the application is as follows:

[0007] A dynamic code loop phase discrimination method is used in a direct sequence spread spectrum receiver, comprising the following steps:

[0008] S1, a current i, i is in [1, Nbits] epoch segment with a length of SF is obtained, which is offset from the original base address by SampAdj i-1 The received spread spectrum data din i is obtained, and the received spread spectrum data is compensated according to the carrier frequency and phase fed back by the carrier loop to obtain loop input data Loopdin i ;

[0009] S2, according to the scene recognition, it is known whether it belongs to a multipath scene, if it is not a multipath scene, the control unit controls the code loop to use the early minus late amplitude value method to discriminate, otherwise the early minus late amplitude value method or the E-L slope method is dynamically selected to discriminate;

[0010] S3, according to the selected phase discrimination method, a plurality of local spread spectrum codes are generated by using a local code generation unit;

[0011] S4, the loop input data Loopdin i is correlated with the plurality of local spread spectrum codes respectively, the plurality of correlation integrals are calculated and discriminated, and the code loop discrimination value CodeDiscRst i of the current epoch is obtained;

[0012] S5, the dynamic filtering unit dynamically adjusts the filtering coefficient based on the plurality of integrals of the correlation unit, so as to realize the dynamic filtering of the code loop discrimination value CodeDiscRst i , and outputs the code phase CodePhase i ;

[0013] S6, the feedback unit feeds back the code phase CodePhase i converted into the offset sampling point SampAdj i-1 and feeds back the selection of the starting sampling point of the spread spectrum data received in the next epoch. The i+1th epoch continues to be executed from step S1 until Nbits*SF pieces of received spread spectrum data are received and processed.

[0014] In some embodiments, in the step S1,

[0015] In the step S1, the i, i∈[1, Nbits]th epoch receives spread spectrum data din i The selection of the reading address is based on the code ring output sampling point offset SampAdj i-1 value of the i-1th period, and the reading address is offset by SampAdj i-1 points on the original address (i-1)*SF+1~i*SF, so that the reading address of din i is (i-1)*SF+1+SampAdj i-1 ~i*SF+SampAdj i-1 .

[0016] After the compensation of din i based on the carrier loop feedback carrier frequency CarrierFreq and phase value CarrierPhase, the loop input data Loopdin i is updated. Wherein, each piece of loop input data Loopdin i with a length of SF enters the loop for correlation, phase detection, filtering and other operations, which is called an epoch. From the start of the loop to the end of the loop operation, a total of Nbits epochs are passed. The compensation formula is:

[0017] Loopdin i = din i *(exp(-1j.*trkPhaseBuf)), wherein din i is an SF row 1 column column vector; trkPhaseBuf is a 1 row SF column row vector representing phase information composed of carrier phase and carrier frequency step, which is a 1 row SF column complex row vector.

[0018] In some embodiments, in the step S2, it also includes

[0019] S2.1 Control unit is responsible for the selection of the phase discrimination method, according to the scene recognition for non-multipath scene, then the early minus late amplitude method is used in the code ring to discriminate the phase, otherwise it is considered as a multipath scene, then the early minus late amplitude method is selected in the subsequent phase discrimination process, and when the loop state tends to be stable, Ena is set to 1, then the i+1 epoch starts, and the E-L slope method is switched to discriminate the phase, otherwise the early minus late amplitude method is used to discriminate the phase. Wherein Ena is an enablement to determine which phase discrimination method to use in a multipath scene, which is initially 0 and enabled to be 1 when the loop is stable.

[0020] S2.2 The determination of the stable state of the loop is based on: when the phase discrimination variance σ n ,n∈[1,Nseg] of each of the Nseg segments, each of which contains M epochs, is not more than 0.1, Ena is set to 1 from the initial value 0, otherwise it is always 0. Wherein Nseg and M are generally not more than 15, a segment contains M epochs, and the phase discrimination value group of the M epochs in the nth segment, n∈[1,Nseg] is expressed as:

[0021] [CodeDiscRst i-Nseg*M+(n-1)*M+1 ,CodeDiscRst i-Nseg*M+(n-1)*M+2 ,...,CodeDiscRst i-Nseg*M+n*M ]

[0022] The average phase discrimination value AvgCodeDiscRst n of the segment M is obtained by accumulating and averaging the phase discrimination values of the segment, and the phase discrimination variance σ n ,n∈[1,Nseg] of the segment, which is calculated as:

[0023]

[0024] In some embodiments, the step S3 also includes

[0025] S3.1 According to the phase discrimination method selected by the control unit in S3, the local code generation unit generates multiple spread spectrum codes, generally at least 3, which are early(CodeE), immediate(CodeP), and late(CodeL) codes. If more spread spectrum codes are generated, the early and late codes are increased in pairs, and the early code closest to the immediate code is CodeE, the early code closest to CodeE is CodeEE, and the early code closest to CodeEE is CodeEEE. The late code is also expressed as CodeX, X=..., EE, E, P, L, LL,....

[0026] S3.2 The early code represents the local spreading code of the early prompt code GapSampX', the prompt code represents the local spreading code without offset, and the late code represents the local spreading code of the late prompt code GapSampX'. The corresponding correlation calculation is used for the early path, the prompt path, and the late path, wherein GapSampX' = SampleRate * dX', X' =..., EE, E, L, LL,..., dX', dX' < 1 is the correlator spacing of the path from the prompt path, SampleRate is the loop data sampling rate, and GapSampX' is the number of local code offset points affected by the correlator spacing.

[0027] In some embodiments, the step S4 further includes

[0028] S4.1 The correlation and phase discrimination unit first completes correlation accumulation calculation and then sends the result to the phase discriminator for phase discrimination.

[0029] S4.2 The loop input data Loopdin i After correlation with the multiple spreading codes, the corresponding correlation results are obtained. The correlation results of the prompt path and the paths closest to the prompt path are represented as ECorrdout i , PCorrdout i , and LCorrdout i , and the corresponding correlation integral results are calculated as ECorrSum i , PCorrSum i , and LCorrSum i . The correlation results and the correlation integral results of the paths next closest to the prompt path are represented as EECorrdout i , LLCorrdout i , EECorrSum i , and LLCorrSum i . In this way, multiple sets of early-late path correlation results and correlation integral results can be represented.

[0030] The correlation calculation is represented as: XCorrdout i = Loopdin i * conj(CodeX)

[0031] wherein Loopdin i is an SF row 1 column vector, CodeX is a 1 row SF column vector, representing all early paths, prompt paths, and late paths, corresponding to the output XCorrdout i , and conj() represents the conjugate. The sum of the correlation results of each path is obtained as the total correlation integral: XCorrSum i=sum(XCorrdout) i ).

[0032] S4.3 obtains the multi-path correlation integral via S4.2, then uses the integral value for phase detection, and outputs the code ring phase detection value CodeDiscRst for the current epoch. i If the lead-lag amplitude method is used for phase detection, the formula is:

[0033] CodeDiscRst i =(1-d)(ECorrSum i -LCorrSum i ) / (ECorrSum i +LCorrSum i )

[0034] If the EL slope method is used for phase detection, the formula is:

[0035]

[0036] in:

[0037] d represents the relevant distance of the lagging or leading path closest to the immediate path. X E' =...,EE represents any leading path other than the leading path closest to the immediate path, X L' =...,LL represents any lag path other than the lag path closest to the instantaneous path, dX E X represents E The coherent distance between the path and the nearest leading path to the instantaneous path, dX L' X represents L' The relative distance between a road and the nearest lagging road to the instantaneous road.

[0038] In some embodiments, step S5 further includes...

[0039] The dynamic filtering unit completes the code ring phase detection output value CodeDiscRst i Perform dynamic loop filtering and output the code phase (CodePhase) of the current epoch. i The filter coefficients in FiltCoef will dynamically change with the distribution characteristics of the multipath correlation integral. The influence of the correlation integral distribution characteristics on the dynamic changes in FiltCoef refers to:

[0040] From the first epoch to the mth epoch, the following relationship has been observed in the total integral magnitude of the three paths for N consecutive epochs:

[0041] abs(PCorrSum m )>abs(ECorrSumm And abs(PCorrSum) m )>abs(LCorrSum m )or

[0042] abs(PCorrSum m )>abs(X E CorrSum m And abs(PCorrSum) m )>abs(X L CorrSum m ), m=k-N+1,k-N+2,...,k, then let FiltCoef=FiltCoef+δ1,δ1≤0.2, where abs() represents the modulus.

[0043] When the lead-lag amplitude method has abs(ECorrSum) m ) or abs(LCorrSum m The maximum, or the existence of abs(X) in the EL slope method. E CorrSum m ) or abs(X L CorrSum m If the maximum value is found, then let FiltCoef = FiltCoef + δ2, where δ2 ≤ 0.4.

[0044] Otherwise, FiltCoef is not adjusted, and code ring filtering is performed using the FiltCoef value for K epochs. After the maintenance epochs, the FiltCoef coefficients are restored to their initial values, and the correlation integral is observed again at the (i+K+1)th epoch. Here, N is the number of observation epochs, K is the number of maintenance epochs, and code ring filtering is represented as: CodePhaseFiltdout i =CodeDiscRst i *FiltCoef, and output the code phase of the current epoch. i =sum(CodePhaseFiltdout) i ), which is the sum of the filtered values ​​from the 1st to the i-th epoch.

[0045] In some embodiments, step S6 further includes the feedback unit obtaining the code phase (CodePhase) of the i-th epoch output. i Convert it into a sampling point offset SampAdj i And feeds back to the outside of the loop, acting on the selection of received spread spectrum data in the (i+1)th epoch, where SampAdj i =CodePhase i *SampleRate.

[0046] The beneficial effects of the present application: in multipath and non-multipath scenarios, the code loop phase detection can be more accurate and accelerate the code loop convergence, and work with the carrier loop, ultimately realize the precise synchronization of carrier frequency and phase, and code phase, correctly recover the modulation information contained in the spread spectrum code, improve the reliability of signal tracking, reduce the bit error rate, thereby improving the system reception performance. BRIEF DESCRIPTION OF DRAWINGS

[0047] Figure 1 Flow chart of the dynamic code loop phase detection method of the embodiment of the present application;

[0048] Figure 2 Principle diagram of the code tracking loop of the embodiment of the present application;

[0049] Figure 3 Carrier frequency fluctuation diagram of the carrier loop of the embodiment of the present application;

[0050] Figure 4 Code phase fluctuation diagram of the code loop of the embodiment of the present application. DETAILED DESCRIPTION

[0051] In order to make the technical problems, technical solutions and advantages of the embodiments of the present application more clear, the present application will be further described in detail below with reference to the drawings. It should be understood that the specific embodiments described herein are only used to explain the present application, but not to limit the present application.

[0052] The present application provides a dynamic code loop phase detection method, which can be used in a direct spread spectrum receiver, is suitable for multipath and non-multipath scenarios, and realizes coarse synchronization of signals based on the acquisition stage, and needs a tracking carrier loop and a code loop to realize precise synchronization of carrier frequency and phase, and synchronization of spread spectrum code. The dynamic code loop phase detection method of the present application has a flow chart as shown in Figure 1 , which comprises the following steps:

[0053] S1, obtaining a segment of length SF of current i, i∈[1, Nbits] epoch, offsetting SampAdj i-1 from the original base address to receive spread spectrum data din i , and compensating the received spread spectrum data according to the carrier frequency and phase fed back by the carrier loop to obtain loop input data Loopdin i ;

[0054] S2, learning whether it belongs to a multipath scenario according to scenario recognition, if it is a non-multipath scenario, the control unit controls the code loop to use the early minus late amplitude method for phase detection, otherwise, dynamically selects the early minus late amplitude method or the E-L slope method for phase detection;

[0055] S3. Based on the selected phase detection method, generate multiple local spreading codes using the local code generation unit;

[0056] S4, the correlation and phase detection unit first processes the loop input data Loopdin i Correlated with multiple local spreading codes, calculate the multi-channel correlation integral and perform phase detection to obtain the code ring phase detection value CodeDiscRst for the current epoch. i ;

[0057] S5. The dynamic filtering unit dynamically adjusts the filtering coefficients based on the multi-channel integration of the relevant unit to achieve the code ring phase detection value CodeDiscRst. i Dynamic filtering, output code phase CodePhase i ;

[0058] S6, The feedback unit will use the code phase CodePhase i Convert to offset sampling point SampAdj i-1 It also provides feedback to control the selection of the starting sample point for receiving spread spectrum data in the next epoch. The (i+1)th epoch continues to execute from step S1 until N bits * SF of received spread spectrum data have been received and processed.

[0059] The schematic diagram of the code tracking loop in this embodiment of the invention is as follows: Figure 2 As shown.

[0060] In step S1, the i-th epoch (i ∈ [1, N bits]) receives the spread spectrum data din. i The selection of the data retrieval address is based on the offset of the sampling point SampAdj of the code ring output in the (i-1)th cycle. i-1 The value is offset by SampAdj at the original address (i-1)*SF+1 to i*SF. i-1 din after 1 point i The address for retrieving the data is (i-1)*SF+1+SampAdj i-1 ~i*SF+SampAdj i-1 Based on the carrier loop feedback carrier frequency CarrierFreq and phase value CarrierPhase, the din... i After compensation, the data is updated to loop input data Loopdin. i Among them, the loop input data Loopdin for each segment of length SF i Entering the loop to perform correlation, phase detection, filtering, and other operations is called one epoch. From the start of the loop to its end, a total of N bits of epochs have passed. The compensation formula is:

[0061] Loopdin i =din i* exp(-1j.*trkPhaseBuf), where din i is a column vector of SF rows and 1 column; trkPhaseBuf is a 1 row SF column row vector representing phase information composed of carrier phase and carrier frequency step; CodeDiscRst is a 1 row SF column complex row vector, in this embodiment, SF = 64, Nbits = 1200.

[0062] The step S2 further includes

[0063] S2.1 The control unit is responsible for the selection of the phase discrimination method. If it is identified as a non-multipath scene according to the scene, the early minus late amplitude method is used in the code ring for phase discrimination. Otherwise, it is considered as a multipath scene, and the early minus late amplitude method is selected first in the subsequent phase discrimination process. When the loop state tends to be stable, Ena is set to 1, and then the E-L slope method is switched to phase discrimination from the i+1 epoch. Otherwise, the early minus late amplitude method is used for phase discrimination. Wherein Ena is an enablement for determining which phase discrimination method to use in a multipath scene. The enablement is initially 0, and is 1 when the loop is stable.

[0064] S2.2 The basis for determining the stability of the loop state is that the phase discrimination variance σ n , n∈[1, Nseg] of each of the Nseg segments, each of which contains M epochs, is not more than 0.1. Then Ena is set to 1 from the initial value 0, otherwise it is always 0. Wherein Nseg and M are generally not more than 15, a segment contains M epochs, and the phase discrimination value group of the M epochs in the nth segment, n∈[1, Nseg] is expressed as:

[0065] [CodeDiscRst i-Nseg*M+(n-1)*M+1 , CodeDiscRst i-Nseg*M+(n-1)*M+2 ,..., CodeDiscRst i-Nseg*M+n*M ]

[0066] The average phase discrimination value AvgCodeDiscRst n of the segment is obtained by accumulating and averaging the M phase discrimination values of the segment, and the phase discrimination variance σ n , n∈[1, Nseg] of the segment is calculated as:

[0067]

[0068] In this embodiment, Nseg = 10 and M = 10

[0069] The step S3 further includes

[0070] S3.1 According to the phase discrimination method selected by the control unit in S3, the local code generation unit generates multiple spread spectrum codes, generally at least 3, which are the early code (Code E), the prompt code (Code P), and the late code (Code L). If more spread spectrum codes are generated, the early and late codes are increased in pairs, and the early code closest to the prompt code is Code E, the early code closest to Code E is Code EE, the early code closest to Code EE is Code EEE, and the late code is represented in the same manner. The multiple local codes are represented by Code X, X =..., EE, E, P, L, LL,....

[0071] S3.2 In the multiple local spread spectrum codes, the early code represents the local spread spectrum code of the prompt code GapSampX' samples, the prompt code represents the local spread spectrum code without offset, and the late code represents the local spread spectrum code of the prompt code GapSampX' samples. The corresponding correlation calculations are used for the early path, the prompt path, and the late path, where GapSampX' = SampleRate * dX', X' =..., EE, E, L, LL,..., dX', dX' < 1 is the correlator spacing of the path from the prompt path, SampleRate is the loop data sampling rate, and GapSampX' is the number of local code offset points affected by the correlator spacing. In this embodiment, when the control unit selects the early minus late amplitude phase discrimination method, three local spread spectrum codes Code E, Code P, and Code L are generated, and when the control unit selects the E-L slope method, five local spread spectrum codes Code EE, Code E, Code P, Code L, and Code LL are generated.

[0072] The step S4 also includes

[0073] S4.1 The correlation and phase discrimination unit first completes the correlation accumulation calculation, and then sends it to the phase discriminator for phase discrimination.

[0074] S4.2 The loop input data Loopdin i After correlation with the multiple spread spectrum codes, the corresponding correlation results of the prompt path and the early path and the late path closest to the prompt path are represented as ECorrdout i , PCorrdout i , and LCorrdout i , and the corresponding correlation integral results are ECorrSum i , PCorrSum i , and LCorrSum i . The correlation results and correlation integral results of the early path and the late path next to the prompt path are represented as EECorrdout i , LLCorrdout i , and EECorrSumi ,LLCorrSum i Similarly, multiple sets of leading and lagging path correlation results and correlation integral results can be represented respectively.

[0075] The relevant calculation is represented as: XCorrdout i =Loopdin i *conj(CodeX)

[0076] Among them, Loopdin i CodeX is a column vector with rows of SF and columns of 1. CodeX is a row vector with rows of SF and columns of 1. CodeX represents all lead paths, instant paths, and lag paths, with the corresponding output XCorrdout. i All are 1-row, SF-column row vectors, and `conj()` indicates taking the conjugate. The total correlation integral is obtained by summing the correlation results from each path: `XCorrSum`. i =sum(XCorrdout) i ).

[0077] S4.3 obtains the multi-path correlation integral via S4.2, then uses the integral value for phase detection, and outputs the code ring phase detection value CodeDiscRst for the current epoch. i If the lead-lag amplitude method is used for phase detection, the formula is:

[0078] CodeDiscRst i =(1-d)(ECorrSum i -LCorrSum i ) / (ECorrSum i +LCorrSum i )

[0079] If the EL slope method is used for phase detection, the formula is:

[0080]

[0081] in:

[0082] d represents the relevant distance of the lagging or leading path closest to the immediate path. X E' =...,EE represents any leading path other than the leading path closest to the immediate path, X L' =...,LL represents any lag path other than the lag path closest to the instantaneous path, dX E X represents E The coherent distance between the path and the nearest leading path to the instantaneous path, dX L' X represents L'The correlation distance between a road and the nearest lagging road. In this embodiment, SampleRate = 4, dX E =0.25, dX L' =0.25, d=0.5.

[0083] In some embodiments, step S5 further includes...

[0084] The dynamic filtering unit completes the code ring phase detection output value CodeDiscRst i Perform dynamic loop filtering and output the code phase (CodePhase) of the current epoch. i The filter coefficients in FiltCoef will dynamically change with the distribution characteristics of the multipath correlation integral. The influence of the correlation integral distribution characteristics on the dynamic changes in FiltCoef refers to:

[0085] From the first epoch to the mth epoch, the following relationship has been observed in the total integral magnitude of the three paths for N consecutive epochs:

[0086] abs(PCorrSum m )>abs(ECorrSum m And abs(PCorrSum) m )>abs(LCorrSum m )or

[0087] abs(PCorrSum m )>abs(X E CorrSum m And abs(PCorrSum) m )>abs(X L CorrSum m ), m=k-N+1,k-N+2,...,k, then let FiltCoef=FiltCoef+δ1,δ1≤0.2, where abs() represents the modulus.

[0088] When the lead-lag amplitude method has abs(ECorrSum) m ) or abs(LCorrSum m The maximum, or the existence of abs(X) in the EL slope method. E CorrSum m ) or abs(X L CorrSum m If the maximum value is found, then let FiltCoef = FiltCoef + δ2, where δ2 ≤ 0.4.

[0089] Otherwise, FiltCoef is not adjusted, and code ring filtering is performed using the FiltCoef value for K epochs. After the maintenance epochs, the FiltCoef coefficients are restored to their initial values, and the correlation integral is observed again at the (i+K+1)th epoch. Here, N is the number of observation epochs, K is the number of maintenance epochs, and code ring filtering is represented as: CodePhaseFiltdout i =CodeDiscRst i *FiltCoef, and output the code phase of the current epoch. i =sum(CodePhaseFiltdout) i ), which is the sum of the filtered values ​​from the 1st to the i-th epoch.

[0090] In this embodiment, δ1 = 0.2, δ2 = 0.4, N = 10, and K = 10 are set.

[0091] Step S6 also includes

[0092] The feedback unit obtains the code phase (CodePhase) output at the i-th epoch. i Convert it into a sampling point offset SampAdj i This is fed back to the outside of the loop and acts on the selection of the received spread spectrum data in the (i+1)th epoch. Where SampAdj... i =CodePhase i *SampleRate.

[0093] In this embodiment, the scene is identified as two paths, each with a delay and attenuation, and a phase (degrees) of 0, -3dB, 0, and 2.1*10. -6 -3dB, 180. The signal-to-noise ratio (SNR) is set to 4dB, with oversampling at 8x and a sampling rate of 800kHz.

[0094] Figure 3 This is a carrier ring frequency fluctuation diagram of this embodiment. It compares two different phase detection mechanisms: the "dynamic code ring phase detection method" in this invention and "phase detection using only the lead-hysteresis method without dynamic filtering." It can be seen that the dynamic code ring mechanism accelerates the convergence of the carrier ring frequency, thereby promoting code ring convergence, and the converged carrier frequency value has relatively small jitter.

[0095] Figure 4 This is a code ring phase fluctuation diagram according to an embodiment of the present invention. Comparing the code ring phase detection values ​​under the "dynamic code ring phase detection method" and "phase detection using only the lead-hysteresis method without dynamic filtering", it can be seen that the former has less jitter and the phase detection value is more accurate.

[0096] The background section of this document can include information about the problem or environment of the invention, not necessarily the prior art. Thus, the content of the background section is not an admission that the prior art is pertinent or that "what is already known has been thoroughly searched, found to be true, accurate and reliable." In fact, it is admitted that the content of the background section can not be pertinent or accurate in view of the prior art.

[0097] The above further describes the present application in connection with specific / preferred embodiments but is not to be taken as a restriction on the overall scope of the application. One of ordinary skill in the art, after having familiarized himself / herself with the teachings of the present application, can make several modifications or variations of these described embodiments without departing from the scope of the present application. Such modifications or variations are intended to be included within the scope of the present application. In the description of the present application, the expressions "one embodiment", "some embodiments", "preferred embodiments", "example", "specific example" or "some examples" are intended to mean that the specific feature, structure, material or characteristic being described in connection with this embodiment or example is included in at least one embodiment or example of the present application. The illustrative expressions of the above terms in the present description are not necessarily intended to refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics being described can be combined in any suitable manner in any one or more embodiments or examples. The person skilled in the art can combine and combine the different embodiments or examples described in the present description and the features of the different embodiments or examples, without mutual contradiction. Although embodiments of the present application and their advantages have been described in detail, it should be understood that various changes, substitutions and alterations can be made herein without departing from the scope of the patent application.

Claims

1. A dynamic code ring phase discrimination method, comprising the following steps: S1, obtaining current first Epoch a length of , on the basis of the original address offset sampling point Offset receiving spread spectrum data , and according to the carrier loop feedback carrier frequency and phase, the receiving spread spectrum data is compensated, and the loop input data is obtained ; S2, according to the scene recognition, it is known whether it is a multipath scene, if it is not a multipath scene, the control unit controls the code ring to use the early minus lag amplitude method to discriminate phase, otherwise, the early minus lag amplitude method or the E-L slope method is selected to discriminate phase; specifically comprising: S2.1 Control unit is responsible for the selection of phase identification method, according to the scene recognition for non-multipath scene, then the early minus lag amplitude method is used in the code ring to identify the phase, otherwise it is considered as a multipath scene, then the early minus lag amplitude method is selected in the subsequent phase identification process, and when the loop state tends to be stable, the enable is set to 1, and the E-L slope method is switched to identify the phase from the beginning of the epoch, otherwise the early minus lag amplitude method is used to identify the phase all the time, wherein is the enable to determine which phase identification method to use in the multipath scene, and the enable is initially 0, and the enable is 1 when the loop is stable; is the enable to determine which phase identification method to use in the multipath scene, and the enable is initially 0, and the enable is 1 when the loop is stable;​ S2.2 The criterion for determining the stability of the loop state is: when the loop is continuously stable... paragraph, each paragraph Phase identification variance per epoch If none of them exceed 0.1, then The value is set from 0 to 1, otherwise it remains 0; where and No more than 15, a segment containing The first epoch, the Section The phase value set for each epoch is expressed as follows: to the segment The average phase detection value of the segment is obtained by accumulating the phase detection values of the segment The phase detection variance of the segment The phase detection variance calculation is represented as: ； S3, according to the selected phase discrimination method, a plurality of local spread spectrum codes are generated by using a local code generation unit; S4, the correlation and phase discrimination unit will loop input data correlation with multiple local spreading codes, respectively, to obtain multiple correlation integral value and phase discrimination, get the current epoch code ring phase discrimination value ; S5, the dynamic filter unit dynamically adjusts the filter coefficient according to the correlation and phase discrimination unit multipath correlation integral value, to realize the dynamic filtering of the code ring phase discrimination value , output code phase ; S6、feedback unit converts code phase to offset sampling point and feeds back, controls selection of starting sampling point of spread spectrum data received in next epoch; the first epoch continues to be executed from step S1 until spread spectrum data is received and processed.

2. The dynamic code ring phase detection method of claim 1, wherein: In step S1, the first Epoch receives spread spectrum data The selection of the read address is based on the first cycle code ring output offset sample point value, and the read address is offset by points from the original address The read address is ; ; represents complex element-wise multiplication, Carrier loop feedback carrier frequency with phase value , spread spectrum data compensation, update for loop input data , where each piece of long loop input data into the loop for correlation, phase detection, filtering operation called an epoch, loop start to loop end of the run a total of epoch, compensation formula: ， where the spread spectrum data is a column vector of length 1, -1 j in j is the complex imaginary unit; is a complex row vector of length 1 representing the phase information consisting of the carrier phase and the carrier frequency step. column.

3. The dynamic code ring phase detection method of claim 1, wherein: The step S3 comprises: S3.1 According to the phase identification method selected by the control unit in S2, the local code generating unit generates multiple spread spectrum codes, including the early path CodeP and the late path CodeL; if more spread spectrum codes are generated, the early path and the late path are increased in pairs, the early code closest to the immediate code is CodeE, the early code closest to CodeE is CodeEE, the early code closest to CodeEE is CodeEEE, and the representation of the late code is also the same, wherein CodeX represents the multiple local codes, ; In S3.2 multi-channel local spreading codes, the lead code represents the lead-instant code. The local spreading code is represented by an instant code, where the instant code indicates a non-offset local spreading code, and the hysteresis code indicates a hysteresis instant code. Each local spreading code is used accordingly for the calculation of leading, instantaneous, and lagging paths, among which... , , , The distance between the path and the corresponding path is the spacing between the path's correlators. This refers to the sampling rate of the loop data. This represents the number of local code offset points affected by the correlator spacing.

4. The dynamic code ring phase detection method of claim 1, wherein: The step S4 comprises: S4.1, the correlation and phase discrimination unit first completes correlation accumulation calculation, and then sends into the phase discriminator to discriminate phase; S4.2 Loop input data The corresponding correlation results are obtained after correlation with the multiple spread spectrum codes, and the correlation results corresponding to the current path and the preceding path and the lagging path closest to the current path are represented as , , , and the corresponding correlation integral results are calculated , , ; the correlation results of the preceding path and the lagging path closest to the current path are represented as , , , , and so on, and the multiple sets of preceding and lagging path correlation results and correlation integral results are represented respectively; The correlation calculation is represented as: ; wherein is a column vector of 1 row, is a row vector of 1 row is a row vector of 1 row is a row vector of 1 row is a row vector of 1 row denotes taking the conjugate; the correlation results of each path are accumulated and summed to obtain the total correlation integral of each path: ; S4.3 Obtain the multi-path correlation integrals from step S4.2, then use the integral values to phase-detect, output the code loop phase-detection value of the current epoch If the early minus late amplitude method is used for phase detection, the formula is: If the E-L slope method is used to discriminate phase, the formula is: ， wherein: , denotes the relevant distance to the nearest lag or lead road from the instant road, , denotes all but the nearest lead road from the instant road, , denotes all but the nearest lag road from the instant road, denotes the relevant distance between the road and the nearest lead road from the instant road, denotes the relevant distance between the road and the nearest lag road from the instant road.

5. The dynamic code ring phase detection method of any one of claims 1 to 4, wherein: In the step S5, the dynamic filtering unit completes the dynamic loop filtering of the code ring phase discrimination output value , and outputs the code phase of the current epoch ; the filtering coefficient will dynamically change with the distribution characteristics of the multi-path correlation integration; the distribution characteristics of the correlation integration affect the dynamic change. From the 1st to the 3rd epoch, there is a relationship between the total 3-path integrated amplitudes of successive epochs of: ​ and or and , then let where denotes modulo; When the lead-lag amplitude method exists or the maximum, or the E-L slope method exists or the maximum, then let ; Otherwise Not adjusted, and with Value maintained Code loop filtering is performed on the epoch, and the epoch is maintained after Coefficient recovery initial value, and the observation of the correlation integral is observed again at the Epoch, wherein The number of observation epochs is The number of maintenance epochs is code loop filtering, which is expressed as: And output the code phase of the current epoch The filtering value of the first to the Epoch is accumulated.

6. A method of dynamic code phase discrimination as claimed in any one of claims 1 to 4, characterized by: In step S6, the feedback unit obtains the first... The code phase of each epoch output Convert it into offset sampling points And feed back to the outside of the loop, acting on the first Each epoch is used for selecting the received spread spectrum data, where SampleRate is the sampling rate.

7. A computer readable storage medium storing a computer program, characterized in that: The computer program is executed by the processor, and the dynamic code ring phase discrimination method in any one of claims 1 to 6 is realized.

Images