CN102332933B - Short-term burst spread spectrum signal transmitting and receiving method - Google Patents

Abstract

The invention provides a short-term burst spread spectrum signal transmitting and receiving method which comprises the following steps of: 1) enabling unique codes and carrying data to constitute a burst spread spectrum signal frame structure, wherein the unique codes play a role of a leader sequence; and 2) enabling a receiving machine to adopt a burst signal receiving method based on differential operation, and simultaneously and quickly completing burst signal detection, PN (pseudo noise) code acquisition, frequency offset estimation, frame synchronization detection and symbol timing acquisition tasks. In the method provided by the invention, only the unique codes are used in the burst leader sequence, and the all-1 or all-0 sequence is not used, so that the transmission efficiency is improved. Simultaneously, the invention further provides a burst signal receiving method based on the differential operation, which can simultaneously and quickly complete the burst signal detection, PN code acquisition, frequency offset estimation, frame synchronization detection, symbol timing acquisition and other tasks.

Description

Short-term burst spread spectrum signal transmitting and method of reseptance

Technical field

The present invention relates to the spread spectrum in radio communication, particularly, relate to a kind of short-term burst spread spectrum signal transmitting and method of reseptance.

Background technology

Spread spectrum technic is anti-interference owing to having advantages of, anti-multipath, good confidentiality, be difficult for intercepting and capturing etc. many, and is widely used in the numerous areas such as civilian, military.Wireless channel channel parameter in communication process often presents relatively violent variation in time, therefore conventionally adopts burst mode.

The difficult point of burst mode transfer is in the quick estimation of synchronization parameter and channel parameter, and the burst of employing direct sequence spread spectrum skill receives the difficulty more having strengthened fast Acquisition and quick estimation.Burst transfer pattern is to transmit data with individual data Bao Wei unit, can intermittently or send continuously packet unit.Consider under mobile channel environment discontinuous in two burst packet transmitting times, the resulting synchronizing information of the previous packet of demodulation and channel parameter information often can not be utilized by the next packet of demodulation, and each packet all needs to reappraise various parameters.In order to realize Rapid timing acquisition and to estimate various channel parameters, before the packet that normally will send at each, insert the targeting sequencing of a specific pattern, receiver utilizes targeting sequencing to carry out Timing Synchronization as auxiliary data and carrier wave recovers, during limited targeting sequencing, burst demodulation device need to be realized the quick estimation of the synchronization acquistion of fast and reliable and carrier frequency, phase information.Therefore, traditional continuous feedback control algolithms such as phase-locked loop are more difficult is applied to this short burst communication pattern.

Burst mode spread spectrum transceiver physical layer requires to support the communication of minimum time slot burst.In communication process, in most cases, the communication that the time slot of take is smallest transmission unit is discontinuous, and the time interval that transceiver receives any two time slots burst is indefinite.The still burst system based on targeting sequencing being most widely used at present, preamble synchronization technology has following advantage: parameter estimation procedure is simple, and the scale of realization can be accepted, and performance is good, and synchronizing speed is fast, and signal-noise ratio threshold is low.Synchronous for based on targeting sequencing, needs targeting sequencing to come settling signal to catch, the rough estimate of synchronization parameter and the indication of frame data original position.Targeting sequencing is generally just divided into two parts, comprises that a part is fixed as aided capture field and unique word (unique word) section of " 1 " or complete " 0 " entirely, referring to open source literature " Guan Yunfeng, Zhang Chaoyang, Jiao Xiangfeng, Chou Peiliang.Design and the realization of burst mode DS-CDMA-QPSK receiver.Circuits and Systems journal, 2001,6 (4): 45-50 ".The acquisition algorithm of burst spread-spectrum signal mainly contains: the fast Acquisition algorithm based on carrier signal power spectral property, by calculating the ratio of despread signal power spectrum maximum and second largest value, and this ratio and decision threshold are compared, thereby whether judgement acquisition success, referring to open source literature " Lu Jizhen, Bu Aimin.The fast Capture Technique research of high dynamically short-term burst spread spectrum signal.Semiconductor technology, 2008,33 (6): 530-533 ".For direct sequence spread spectrum communication system under short-term burst pattern, require with very low false-alarm and alarm dismissal probability, to catch this difficult problem of PN code phase in finite time open source literature " peace Jianping, Shen Yebing, Wang Aihua, Luo Changqing.PN code under a kind of low input signal-to-noise ratio and the constraint of low CFAR is caught algorithm soon.War industry's journal, 2007,28 (11): 1320-1321 " a kind of PN code fast Acquisition algorithm of the resident detection of multiple-symbol noncoherent accumulation single based on digital matched filter (DMF) has been proposed.It adds up to improve the signal to noise ratio of code phase judgement amount on the basis of the resident algorithm of traditional single with multiple-symbol, thereby greatly reduces the mean acquisition time under low input signal-to-noise ratio and the constraint of low CFAR." Zhang Guihua, king ties up roc to open source literature.A kind of method of fast Acquisition short-term burst direct sequence.Radio engineering, 2004,34 (7): 24-55 " adopt intermediate-frequency integration and video integration combines, the technology of waiting type pseudo-code cascade correlation computations, solved preferably the fast Acquisition problem of short-term burst pseudo noise code.Open source literature " Wang Zuliang, Yao Chunguang, Wang Juan, Zheng Linhua.Short frame spread spectrum communication system Processing Algorithm design and Multiple Purpose Simulation happen suddenly under complex environment.Signal is processed, 2008,24 (4): 696-699 " propose to adopt the parallel capturing method based on frequency domain, and adopt Frequency scope, Frequency stepping and become the search strategy of the time of integration; system can be worked under large Frequency Offset, can under Low SNR, work again.Open source literature " Wang Jun, peace Jianping, Song Shujuan.A kind of new high Dynamic DSSS Receiver catching method.Beijing Institute of Technology's journal, 2004,24 (5): 439-465 " take fast Acquisition as target; proposed the PN code capture method that segmented matched filter correlation method, MAXIMUM SELECTION detection method and multichannel frequency parallel search add FFT calibrating frequency technology; simulation result shows that the method can tolerate larger frequency dynamic scope, compared with completing fast code under low signal-to-noise ratio, is catching.These algorithms or do not use the targeting sequencing of burst frame, or only use complete " 1 " or complete " 0 " part of targeting sequencing in burst frame, do not consider the application of unique word, by burst signal detection, PN code, synchronous and frame synchronization has been isolated and has been come, this targeting sequencing that synchronizing process is needed is elongated, has reduced efficiency of transmission.

The present invention proposes a kind of short-term burst spread spectrum signal transmitting and method of reseptance.The present invention only uses unique word in burst targeting sequencing, and does not use complete " 1 " or complete " 0 " sequence, has improved efficiency of transmission.Meanwhile, the invention allows for the burst method of reseptance based on calculus of differences, can complete fast that burst signal detection, PN code are caught, frequency deviation is estimated, frame synchronization detects and bit timing such as catches at the task simultaneously.

Summary of the invention

The invention provides a kind of short-term burst spread spectrum signal transmitting method, and a kind of burst method of reseptance based on calculus of differences is provided, can completes fast that burst signal detection, PN code are caught, frequency deviation is estimated, frame synchronization detects and bit timing such as catches at the task simultaneously.

According to an aspect of the present invention, provide a kind of short-term burst spread spectrum signal transmitting and method of reseptance, it is characterized in that: comprise the steps:

Step 1: unique word and carrying data are formed to burst spread-spectrum signal frame structure, wherein, using unique word as targeting sequencing;

Step 2: make receiver adopt the burst method of reseptance based on calculus of differences, can complete fast that burst signal detection, PN code are caught, frequency deviation is estimated, frame synchronization detects and bit timing is caught task simultaneously.

Preferably, in described step 1, the data after framing and spreading code are carried out to spread processing, then form filtering and BPSK modulation.

Preferably, in described step 2, receiver is caught spread spectrum code acquisition and unique word simultaneously and is considered, by M unique word X ' ₀, X ' ₁... X ' _m-1information sequence is treated as a super spreading code after corresponding spread spectrum, be designated as

XC ₌x ' ₀ ^c, X ' ₁ ^c..., X ' _m-1 ^cnow, we catch by the two dimension of spreading code and unique Quick Response Code the one dimension that is converted into a super spreading code and catch.

Preferably, in described step 2, in super spread spectrum code acquisition process, adopt moving average method to estimate the noise power of differential output signal, and self adaptation arrange decision threshold accordingly.Difference output is got real part and is compared as judgment variables and threshold value after integration, carries out super spread spectrum code acquisition detection.

Preferably, in described step 2, after having completed the catching of super spreading code (being that burst signal detection, PN code are caught and frame synchronization detects), according to the symmetry of differential output signal, carry out bit synchronization and catch.

Preferably, in described step 2, after completion bit synchronization acquistion, the sequence (now can obtain maximum signal to noise ratio) of utilizing optimum sampling point to form is carried out frequency deviation estimation.

Preferably, described step 2 comprises the steps:

The first sub-step: analog signal becomes digital signal after ADC, supposes every chip sampling IV sample, i.e. sampling period T _s=T _c/ N, T _cfor the chip cycle.

The second sub-step: digital sampled signal is with speed f _s=1/T _scarry out matched filtering, filtering output is as follows:

$r\left(k\right)=\sqrt{E_c}g\left[(k-\mathrm{uN})T_s\right]x\left[(k-\mathrm{uN})T_s\right]c\left[(k-\mathrm{uN})T_s\right]e^{j\left\{\mathrm{\Δ\ω}\right[(k-\mathrm{uN})T_s]+\mathrm{\φ}\}}+n\left(k{T}_s\right)$

E wherein _cit is chip energy, g (t) is the base band pulse after chip matched filtering, it is raised cosine (RC) pulse, x (t)=the ± 1st, user sends information, c (t) is user's spreading code, and Δ ω=2 π Δ f is the frequency departure that Doppler frequency displacement and carrier beat cause jointly, and φ is the carrier phase of receiving terminal, it comprises the phase place that the carrier wave first phase of making a start and transmission channel cause and changes, and n (t) is that one-sided power spectrum density is N ₀white complex gaussian noise, average is 0, variance is

matched filtering output is delivered to shift register and is stored.

The 3rd sub-step: the current sample of the matched filtering of take output is pointer, take N as extracting than extracting, and obtains the extraction sequence of a point of every chip.

The 4th sub-step: extract sequence and super spreading code X _ccarry out related operation, its output is as follows:

$y\left(k\right)=\sqrt{E_c}g\left[(k-\mathrm{uN})T_s\right]x\left[(k-\mathrm{uN})T_s\right]c\left[(k-\mathrm{uN})T_s\right]e^{j\left\{\mathrm{\Δ\ω}\right[(k-\mathrm{uN})T_s]+\mathrm{\φ}\}}+n^{\′}\left(k{T}_s\right)$

N ' (kT wherein _s)=xc (k) n (kT _s).Carrying out time delay is lT again _scalculus of differences, difference output is as follows:

$\begin{array}{c}{d}_1\left(k\right)=y\left(k\right)*y^{*}(k-l)\\ =E_c{g}^{\′}\left[(k-\mathrm{uN})T_s\right]x^{\′}\left[(k-\mathrm{uN})T_s\right]c^{\′}\left[(k-\mathrm{uN})T_s\right]x^{\′}c\left(k\right)e^{j\left(\mathrm{\Δ\ω}{T}_s\right)}+n^{\″}\left(k{T}_s\right)\end{array}$

g′[(k-uN)T _s]=g[(k-uN)T _s]*g[(k-uN-l)T _s]，

x[(k-uN)T _s]=x[(k-uN)T _s]*x*](k-uN-l)T _s]，

c′[(k-uN)T _s]=c[(k-uN)T _s]*c ^*[(k-uN-l)T _s]，

X ' _c(k)=xc (k) * x _c(k-l), wherein,

N " (kT _s) be difference output noise.

The positive integer of l for being divided exactly by N.

The 5th sub-step: calculus of differences is through MN _cthe cumulative operation of length, computing formula is as follows:

$y_{d_1}\left(k\right)=\underset{k=0}{\overset{M{N}_c-1}{\mathrm{\Σ}}}{d}_1\left(k\right)$

It is exported a road and gets real part and obtain judgment variables

computing formula is as follows:

$y_{R,d_l}\left(k\right)=\mathrm{Re}\left[y_{d_1}\left(k\right)\right]$

Ref wherein) represent to get real part.Power is asked on another road, through the laggard line slip average operation of time delay, and obtains difference output noise power

computing formula is as follows:

${\mathrm{\σ}}_{w,n}^2=\frac{1}{L}\underset{m=1}{\overset{L}{\mathrm{\Σ}}}{\left|y_{d_l}(k-m)\right|}^2$

Wherein, L is sliding window length.Finally, according to difference output noise power, adaptive threshold T is set _h.

The 6th sub-step: judgment variables with adaptive threshold T _hcompare,

$\left[\begin{array}{cc}H=H_1& \mathrm{if}{y}_{R,d_1}\left(k\right)\&GreaterEqual;T_h\\ H=H_0& \mathrm{if}{y}_{R,d_l}\left(k\right)<T_h\end{array}\right]$

Wherein, H ₁represent that super spread spectrum code acquisition (being that burst signal detection, PN code are caught and frame synchronization detection) event occurs, H ₀represent that capturing events does not occur.

The 7th sub-step: after having completed the catching of super spreading code (being that burst signal detection, PN code are caught and frame synchronization detects), by judgment variables

deliver to the shift register { D of 3 states ₁, D2, D3} is after signal capture, according to shift register { D ₁, D ₂, D ₃state carry out bit timing and catch.If D ₂>D ₁and D ₂>D ₃, represent that bit timing caught, now optimum sampling point is D ₂corresponding sampling instant.

The 8th sub-step: after bit timing has been caught, the sequence (now can obtain maximum signal to noise ratio) of utilizing optimum sampling point to form is carried out frequency deviation estimation, shown in formula specific as follows:

$\mathrm{\&Delta;f}=\mathrm{arctg}\left(\underset{k=0}{\overset{{\mathrm{MN}}_c-1}{\mathrm{\&Sigma;}}}{d}_1\left(k\right)\right)/\left(2\mathrm{\&pi;l}{T}_s\right).$

In order to realize goal of the invention, the method that the present invention proposes comprises:

L, burst spread-spectrum signal frame structure are comprised of unique word and carrying data.

2, receiver is caught spread spectrum code acquisition and unique word simultaneously and is considered, information sequence after spread spectrum corresponding to M unique word is treated as a super spreading code, the two dimension of spreading code and unique word is caught to the one dimension that is converted into a super spreading code and catch.

3, in super spread spectrum code acquisition process, adopt moving average method to estimate the noise power of differential output signal, and self adaptation arrange decision threshold accordingly.Difference output is got real part and is compared as judgment variables and threshold value after integration, carries out super spread spectrum code acquisition detection.

4, after having completed the catching of super spreading code (being that burst signal detection, PN code are caught and frame synchronization detects), according to the symmetry of differential output signal, carry out bit synchronization and catch.

5,, after completion bit synchronization acquistion, the sequence (now can obtain maximum signal to noise ratio) of utilizing optimum sampling point to form is carried out frequency deviation estimation.

Accompanying drawing explanation

Figure l illustrates burst spread-spectrum signal frame structure provided by the invention.

Fig. 2 illustrates short-term burst spread spectrum signal method of reseptance provided by the invention.

Specific implementation method

Below in conjunction with accompanying drawing, further illustrate the present invention.

Figure l is burst spread-spectrum signal frame structure provided by the invention.Burst spread-spectrum signal frame structure is comprised of unique word and carrying data, wherein, and using unique word as targeting sequencing.Data after framing and spreading code are carried out to spread processing, then form filtering and BPSK modulation.Wherein, described should be understood to unique word as targeting sequencing: the present invention only uses unique word in burst targeting sequencing, and does not use complete " l " or complete " 0 " sequence.

Fig. 2 is short-term burst spread spectrum signal method of reseptance provided by the invention.Receiver adopts the burst method of reseptance based on calculus of differences, can complete fast that burst signal detection, PN code are caught, frequency deviation is estimated, frame synchronization detects simultaneously and bit timing such as catches at the task.Whole process comprises the steps:

The first sub-step: analog signal becomes digital signal after ADC, supposes every chip sampling N sample, i.e. sampling period T _s=T _c/ N, T _cfor the chip cycle.

The second sub-step: digital sampled signal is with speed f _s=l/T _scarry out matched filtering, filtering output is as follows:

$r\left(k\right)=\sqrt{E_c}g\left[(k-\mathrm{uN})T_s\right]x\left[(k-\mathrm{uN})T_s\right]c\left[(k-\mathrm{uN})T_s\right]e^{j\left\{\mathrm{\&Delta;\&omega;}\right[(k-\mathrm{uN})T_s]+\mathrm{\&phi;}\}}+n\left(k{T}_s\right)$

E wherein _cit is chip energy, g (t) is the base band pulse after chip matched filtering, it is raised cosine (RC) pulse, x (t)=the ± 1st, user sends information, c (t) is user's spreading code, and Δ ω=2 π Δ f is the frequency departure that Doppler frequency displacement and carrier beat cause jointly, and φ is the carrier phase of receiving terminal, it comprises the phase place that the carrier wave first phase of making a start and transmission channel cause and changes, and n (t) is that one-sided power spectrum density is N ₀white complex gaussian noise, average is 0, variance is

.Matched filtering output is delivered to shift register and is stored.

The 3rd sub-step: the current sample of the matched filtering of take output is pointer, take J7v as extracting than extracting, and obtains the extraction sequence of a point of every chip.

The 4th sub-step: extract sequence and super spreading code X _ccarry out related operation, its output is as follows:

$y\left(k\right)=\sqrt{E_c}g\left[(k-\mathrm{uN})T_s\right]x\left[(k-\mathrm{uN})T_s\right]c\left[(k-\mathrm{uN})T_s\right]e^{j\left\{\mathrm{\&Delta;\&omega;}\right[(k-\mathrm{uN})T_s]+\mathrm{\&phi;}\}}+n^{\&prime;}\left(k{T}_s\right)$

N ' (kT wherein _s)=xc (k) n (kT _s) to carry out time delay be lT again _scalculus of differences, difference output is as follows:

$\begin{array}{c}{d}_1\left(k\right)=y\left(k\right)*y^{*}(k-l)\\ =E_c{g}^{\&prime;}\left[(k-\mathrm{uN})T_s\right]x^{\&prime;}\left[(k-\mathrm{uN})T_s\right]c^{\&prime;}\left[(k-\mathrm{uN})T_s\right]x^{\&prime;}c\left(k\right)e^{j\left(\mathrm{\&Delta;\&omega;}{T}_s\right)}+n^{\&Prime;}\left(k{T}_s\right)\end{array}$

g′[(k-uN)T _s]=g[(k-uN)T _s]*g[(k-uN-l)T _s],

x′[(k-uN)T _s]=x[(k-uN)T _s]*x*[(jk-uN-l)T _s],

Wherein, c ' [(k-uN) T _s]=c[(k-uN) T _s] * c ^*[(k-uN-l) T _s],

x′ _C(k)=x _C(k)*xc(k-l),

N " (kT _s) be difference output noise.

The positive integer of l for being divided exactly by N.

The 5th sub-step: calculus of differences is through MN _cthe cumulative operation of length, computing formula is as follows:

$y_{d_1}\left(k\right)=\underset{k=0}{\overset{M{N}_c-1}{\mathrm{\&Sigma;}}}{d}_1\left(k\right)$

It is exported a road and gets real part and obtain judgment variables computing formula is as follows:

$y_{R,d_1}\left(k\right)=\mathrm{Re}\left[y_{d_1}\left(k\right)\right]$

Ref wherein) represent to get real part.Power is asked on another road, through the laggard line slip average operation of time delay, and obtains difference output noise power

computing formula is as follows:

${\mathrm{\&sigma;}}_{w,n}^2=\frac{1}{L}\underset{m=1}{\overset{L}{\mathrm{\&Sigma;}}}{\left|y_{d_l}(k-m)\right|}^2$ Wherein, L is sliding window length.Finally, according to difference output noise power, adaptive threshold T is set _h.

The 6th sub-step: judgment variables with adaptive threshold T _hcompare,

$\left[\begin{array}{cc}H=H_1& \mathrm{if}{y}_{R,d_1}\left(k\right)\&GreaterEqual;T_h\\ H=H_0& \mathrm{if}{y}_{R,d_l}\left(k\right)<T_h\end{array}\right]$

Wherein, H ₁represent that super spread spectrum code acquisition (being that burst signal detection, PN code are caught and frame synchronization detection) event occurs, H ₀represent that capturing events does not occur.

The 7th sub-step: after having completed the catching of super spreading code (being that burst signal detection, PN code are caught and frame synchronization detects), by judgment variables

deliver to the shift register of 3 states: { D ₁: D ₂, D ₃), after signal capture, according to shift register { D ₁, D ₂, D ₃state carry out bit timing and catch.If D ₂>D ₁and D ₂>D ₃, represent that bit timing caught, now optimum sampling point is D ₂corresponding sampling instant.

The 8th sub-step: after bit timing has been caught, the sequence (now can obtain maximum signal to noise ratio) of utilizing optimum sampling point to form is carried out frequency deviation estimation, shown in formula specific as follows:

$\mathrm{\&Delta;f}=\mathrm{arctg}\left(\underset{k=0}{\overset{{\mathrm{MN}}_c-1}{\mathrm{\&Sigma;}}}{d}_1\left(k\right)\right)/\left(2\mathrm{\&pi;l}{T}_s\right).$

Claims (5)

1. short-term burst spread spectrum signal transmitting and a method of reseptance, is characterized in that: comprise the steps:

Step 1: unique word and carrying data are formed to burst spread-spectrum signal frame structure, wherein, using unique word as targeting sequencing;

Step 2: make receiver adopt the burst method of reseptance based on calculus of differences, complete fast that burst signal detection, PN code are caught, frequency deviation is estimated, frame synchronization detects and bit timing is caught task simultaneously;

In described step 2, receiver is caught spread spectrum code acquisition and unique word simultaneously and is considered, by M unique word x ' ₀, x ' ₁..., x ' _m-1information sequence is treated as a super spreading code after corresponding spread spectrum, is designated as x _c=x ' ₀c, x ' ₁c ..., x ' _m-1c};

In described step 2, in super spread spectrum code acquisition process, adopt moving average method to estimate the noise power of differential output signal, and self adaptation arrange decision threshold accordingly; Difference output is got real part and is compared as judgment variables and threshold value after integration, carries out super spread spectrum code acquisition detection.

2. the method for claim 1, is characterized in that: in described step 1, the data after framing and spreading code are carried out to spread processing, then form filtering and BPSK modulation.

3. the method for claim 1, is characterized in that: in described step 2, after having completed the catching of super spreading code, according to the symmetry of differential output signal, carry out bit synchronization and catch.

4. the method for claim 1, is characterized in that: in described step 2, after completion bit synchronization acquistion, the sequence of utilizing optimum sampling point to form is carried out frequency deviation estimation.

5. the method for claim 1, is characterized in that: described step 2 comprises the steps:

The first sub-step: analog signal becomes digital signal after ADC, supposes every chip sampling N sample, i.e. sampling period T _s=T _c/ N, T _cfor the chip cycle;

The second sub-step: digital sampled signal is with speed f _s=1/T _scarry out matched filtering, filtering output is as follows:

$r\left(k\right)=\sqrt{E_{c}g}\left[(k-\mathrm{uN})T_s\right]x\left[(k-\mathrm{uN})T_s\right]c\left[(k-\mathrm{uN})T_s\right]e^{j\left\{\mathrm{\&Delta;\&omega;}\right[(k-\mathrm{uN})T_s]+\mathrm{\&phi;}\}}+n\left({\mathrm{kT}}_s\right)$

E wherein _cit is chip energy, g (t) is the base band pulse after chip matched filtering, it is raised cosine (RC) pulse, x (t)=the ± 1st, user sends information, c (t) is user's spreading code, and Δ ω=2 π Δ f is the frequency departure that Doppler frequency displacement and carrier beat cause jointly, and φ is the carrier phase of receiving terminal, it comprises the phase place that the carrier wave first phase of making a start and transmission channel cause and changes, and n (t) is that one-sided power spectrum density is N ₀white complex gaussian noise, average is 0, variance is

matched filtering output is delivered to shift register and is stored;

The 3rd sub-step: the current sample of the matched filtering of take output is pointer, take N as extracting than extracting, and obtains the extraction sequence of a point of every chip;

The 4th sub-step: extract sequence and super spreading code X _ccarry out related operation, its output is as follows:

$y\left(k\right)=\sqrt{E_{c}g}\left[(k-\mathrm{uN})T_s\right]x\left[(k-\mathrm{uN})T_s\right]c\left[(k-\mathrm{uN})T_s\right]e^{j\left\{\mathrm{\&Delta;\&omega;}\right[(k-\mathrm{uN})T_s]+\mathrm{\&phi;}\}}+n^{\&prime;}\left({\mathrm{kT}}_s\right)$

N'(kT wherein _s)=xc (k) n (kT _s); Carrying out time delay is lT again _scalculus of differences, difference output is as follows:

$\begin{array}{c}{d}_1\left(k\right)=y{\left(k\right)}^{*}{y}^{*}(k-1)\\ =E_c{g}^{\&prime;}\left[(k-\mathrm{uN})T_s\right]x^{\&prime;}\left[(k-\mathrm{uN})T_s\right]c^{\&prime;}\left[(k-\mathrm{uN})T_s\right]x^{\&prime;}c\left(k\right)e^{j\left(\mathrm{\&Delta;\&omega;l}{T}_s\right)}+n^{\&prime;\&prime;}\left(k{T}_s\right)\end{array}$

Wherein, g ' [(k-uN) T _s]=g[(k-uN) T _s] * g[(k-uN-l) T _s],

x′[(k-uN)T _s]=x[(k-uN)T _s]*x ^*[(k-uN-l)T _s,

c′[(k-uN)T _s]=c[(k-uN)T _s]*c ^*[(k-uN-l)T _s],

x′ _C(k)=x _C(k) ^*x _C(k-l)，

N " (kT _s) be difference output noise;

The positive integer of l for being divided exactly by N;

The 5th sub-step: calculus of differences is through the cumulative operation of MNC length, and computing formula is as follows:

$y_{d_l}\left(k\right)=\underset{k=0}{\overset{{\mathrm{MN}}_c-1}{\mathrm{\&Sigma;}}}{d}_l\left(k\right)$

It is exported a road and gets real part and obtain judgment variables computing formula is as follows:

$y_{R,d_l}\left(k\right)=\mathrm{Re}\left[y_{d_l}\left(k\right)\right]$

Wherein Re () represents to get real part; Power is asked on another road, through the laggard line slip average operation of time delay, and obtains difference output noise power

computing formula is as follows:

${\mathrm{\&sigma;}}_{w,n}^2=\frac{1}{L}\underset{m=1}{\overset{L}{\mathrm{\&Sigma;}}}{\left|y_{d_l}(k-m)\right|}^2$

Wherein, L is sliding window length; Finally, according to difference output noise power, adaptive threshold T is set _h;

The 6th sub-step: judgment variables

with adaptive threshold T _hcompare,

$\left\{\begin{array}{cc}H=H_1& \mathrm{if}{y}_{R,d_l}\left(k\right)\&GreaterEqual;T_h\\ H=H_0& \mathrm{if}{y}_{R,d_l}\left(k\right)<T_h\end{array}\right.$

Wherein, H ₁represent that super spread spectrum code acquisition (being that burst signal detection, PN code are caught and frame synchronization detection) event occurs, H ₀represent that capturing events does not occur;

The 7th sub-step: after having completed the catching of super spreading code (being that burst signal detection, PN code are caught and frame synchronization detects), by judgment variables

deliver to the shift register { D of 3 states ₁, D ₂, D ₃, after signal capture, according to shift register { D ₁, D ₂, D ₃state carry out bit timing and catch; If D ₂>D ₁and D ₂>D ₃, represent that bit timing caught, now optimum sampling point is D ₂corresponding sampling instant;

The 8th sub-step: after bit timing has been caught, the sequence (now can obtain maximum signal to noise ratio) of utilizing optimum sampling point to form is carried out frequency deviation estimation, shown in formula specific as follows:

$\mathrm{\&Delta;f}=\mathrm{arctg}\left(\underset{k=0}{\overset{M{N}_c-1}{\mathrm{\&Sigma;}}}{d}_l\left(k\right)\right)/\left(2\mathrm{\&pi;l}{T}_S\right).$