A kind of method that suppresses the random access code false-alarm
Technical field
The present invention relates to communication technical field, more specifically, relate to a kind of method that suppresses the random access code false-alarm.
Background technology
For the cellular wireless mobile communication system, Random Access Channel is used for carrying out the initial synchronisation of multiple access and upper line time.Training sequence in the Random Access Channel is called random access code.In some system, use cyclically shifted sequences with Cyclic Prefix as random access code.For example, in the LTE system, use the ZC sequence, by Cyclic Prefix, convert linear convolution to circular convolution, thus the convenient realization of adopting the spectrum correlation algorithm when receiving.
In utilizing the Receiver Design of random access code, often face the false-alarm problem that the relevant peaks energy leakage brings.Accompanying drawing 1 is the receiver basic function module, comprises local sequence generation module 101, receives signal pre-processing module 102, correlation module 103 and detection and timing estimation module 104.
Receiving signal pre-processing module 102 will receive signal and go Cyclic Prefix (CP) to be transformed into afterwards the local sequence that frequency domain and local sequence generation module 101 export to carry out the complex conjugate dot product.In correlation module 103, the result is carried out anti-Fourier's variation and be transformed into time domain, can obtain corresponding time domain correlation.Then detect with timing estimation module 104 and in each detection window, find peak-peak M, estimating noise power σ
2, and detect relative threshold γ with peak value and compare.Can be peak-peak or other detection limit according to real system M.
Two kinds of detection threshold are arranged at present: relative threshold and absolute thresholding.
Relative threshold: the correlation peak/noise power of signal; γ is the relative threshold of not determining under given false-alarm targets during transmitted signal.
Absolute thresholding: the big or small δ of estimating noise power
2, with δ
2Multiply each other with relative threshold γ and can obtain absolute thresholding γ
a
If M/ is σ
2Greater than γ, then explanation accesses successfully, on the contrary failure.
In the LTE system, use the ZC sequence as random access code.The reception signal is transformed into frequency domain and local sequence is carried out the complex conjugate dot product, obtains being N
ZCThe sequence of point.N
ZCBe the length of frequency domain ZC sequence, be generally prime number, need zero padding to arrive N
IFFTPoint carries out anti-Fourier and changes N
IFFTBe generally 2 integral number power.Then correlation is asked mould square, ask as seen from Figure 2 the waveform behind the mould square.
Under the stable environment of noise power, γ
aSubstantially remain unchanged.Along with the increase of signal to noise ratio, the performance number of signal increases, and it is increasing that the secondary lobe of signal overflows meeting in the accompanying drawing 2, can spill into adjacent detection window.If the peak value of adjacent detection window is greater than γ
a, when detecting, can cause false-alarm, namely do not have the user to send in the current detection window, but detected the user.False-alarm mainly is to be caused by the nearer several adjacent detection window energy leakage of distance objective detection window, therefore needs to suppress the false-alarm that these adjacent detection window cause.
In addition, in the environment of frequency deviation (residual frequency departure and Doppler frequency shift), can produce the submaximum value in the fixing position of distance peak value.This fixing position of definition is d in the LTE system
u, the root relating to parameters of it and current use.Along with the increase of frequency deviation, apart from main peak ± d
u, ± 2d
u, ± 3d
uSubmaximum value at the place is increasing.
If adopt single window detection scheme, at distance main peak ± d
u, ± 2d
uSubmaximum value at the place can enter in the scope of other detection window.If peak value surpasses thresholding, illustrate to detect the another one user, can cause false-alarm like this.
If adopt three window Merge Scenarioses, with positive and negative submaximum value (apart from main peak value ± d
u) merge detection with main peak.Main window ± the 2d of distance
u, ± 3d
uSubmaximum value at the place may appear in another one user's the main detection window or duplicate detection window.When in comprising the detection window of these positions, detecting, if peak value surpasses thresholding, illustrate to detect the another one user, can cause false-alarm like this.
Adopted at present some technology that false-alarm is suppressed.
For example, consider the Energy Leaking of relevant peaks, in a certain distance of peak value periphery that detects, do not do detection.Although avoided to a certain extent false-alarm.But when multiple access, may can't detect the user of close together, so that user's conductively-closed of close together has caused the decline that detects performance.
For another example, after the receiving terminal random access code carries out the time domain coherent detection, when detecting certain user, find the position of the caused false-alarm of this user according to the frequency bias properties of ZC sequence.When detecting other window, can avoid these positions to reach the purpose that suppresses false-alarm.But under multi-user's scene, if some users cause that the position of false-alarm just in time is the position of the normal access of another one user, meeting has reduced multi-user's detection probability so that the user that should detect loses the chance of access.
To sum up, the existing problem that suppresses to exist in the false-alarm technology the normal access probability of reduction multi-user.
Summary of the invention
The embodiment of the invention proposes a kind of method that suppresses the random access code false-alarm, in the situation that guarantee the normal access probability of multi-user, effectively suppresses false-alarm.
The technical scheme of the embodiment of the invention is as follows:
A kind of method that suppresses the random access code false-alarm, the method comprises:
Peak value corresponding to random access code in the measurement target detection window;
Described peak value is deducted non-target detection window and is revealed correction peak value corresponding to energy acquisition random access code;
Judge according to described correction peak value and absolute thresholding whether the user that described target detection window occurs is false-alarm, if false-alarm, then mark in described target detection window.
Further comprise before the peak value that random access code is corresponding in the described measurement target detection window: all detection window are carried out pre-detection, in detecting the detection window of peak value, determine the target detection window.
Describedly determine that in detecting the detection window of peak value the target detection window comprises:
When high s/n ratio, described peak value is greater than absolute thresholding, distance detect have a detection window at least in K the nearer detection window of the detection window of peak value peak value greater than absolute thresholding, the detection window of then determining to detect peak value is the target detection window, K is less than or equal to 4.
Describedly determine that in detecting the detection window of peak value the target detection window comprises:
In the situation that frequency displacement occurs, described peak value is greater than absolute thresholding, and each detection window of determining to detect peak value is the target detection window.
It is to be determined by at least one product that has detected inhibiting factor corresponding to random access code is corresponding in user's the non-target detection window peak value and described non-target detection window that described non-target detection window is revealed energy.
Further comprise before the peak value that random access code is corresponding in the described measurement target detection window: random access lead code of transmitting terminal emission, measure the peak value that each detection window occurs;
At high s/n ratio and when having at the most 2 adjacent windows the user to occur, described inhibiting factor comprises the first inhibiting factor and the second inhibiting factor;
The first inhibiting factor equals A
K-1With A
KRatio, A
K-1The energy that the previous detection window label of target detection window is the K-1 detection window, A
KIt is the peak value of target detection window;
The second inhibiting factor equals A
K-2With A
KRatio, A
K-2That the previous label of K-1 detection window is the peak value of the detection window of K-2;
Corresponding first inhibiting factor of detection window that the detection window label is K+1 behind K-1 detection window and the target detection window;
Corresponding second inhibiting factor of detection window that the detection window label is K+2 after K-2 detection window and the K+1 detection window.
Further comprise before the peak value that random access code is corresponding in the described measurement target detection window: transmitting terminal is launched random access lead code, measures the peak value of each detection window;
At high s/n ratio and when having at least 3 adjacent windows the user to occur, described inhibiting factor comprises the first inhibiting factor and the second inhibiting factor;
The first inhibiting factor equals A
K-1With A
KRatio, A
K-1The peak value that the previous detection window label of target detection window is the K-1 detection window, A
KIt is the peak value of target detection window;
The second inhibiting factor equals zero;
Corresponding first inhibiting factor of detection window that the detection window label is K+1 behind K-1 detection window and the target detection window;
Corresponding second inhibiting factor of detection window that the detection window label is K+2 after K-2 detection window and the K+1 detection window.
Further comprise before the peak value that random access code is corresponding in the described measurement target detection window: transmitting terminal is launched random access lead code, measures the peak value of each detection window;
At high s/n ratio and when having at least 3 adjacent windows the user to occur, described inhibiting factor comprises the first inhibiting factor and the second inhibiting factor;
The first inhibiting factor equals A
K-1With A
KRatio, A
K-1The energy that the previous detection window label of target detection window is the K-1 detection window, A
KIt is the peak value of target detection window;
The second inhibiting factor equals A
K-2With A
KRatio, A
K-2That the previous label of K-1 detection window is the peak value of the detection window of K-2;
Corresponding first inhibiting factor of detection window that the detection window label is K+1 behind K-1 detection window and the target detection window;
Corresponding second inhibiting factor of detection window that the detection window label is K+2 after K-2 detection window and the K+1 detection window.
Further comprise between the peak value that random access code is corresponding in the described measurement target detection window: transmitting terminal is launched random access lead code, measures the peak value of each detection window;
In the situation that frequency displacement occurs, described inhibiting factor comprises at least one sequence inhibiting factor;
The described sequence factor equals A
K+dWith A
KRatio, A
K+dThat the distance objective detection window is the peak value of the detection window of d, A
KIt is the peak value of target detection window.
Describedly judge according to described correction peak value and absolute thresholding whether the user that described target detection window occurs is that false-alarm comprises:
Described correction peak value is less than absolute thresholding, and then the user of described target detection window appearance is false-alarm;
Otherwise, determine that the user that described target detection window occurs is not false-alarm.
From technique scheme, can find out, in embodiments of the present invention, peak value corresponding to random access code in the measurement target detection window at first; Then described peak value is deducted non-target detection window and is revealed correction peak value corresponding to energy acquisition random access code; Judge according to described correction peak value and absolute thresholding whether the user that described target detection window detects is false-alarm, if described target detection window is false-alarm, mark in described target detection window then.Because according to revising peak value and absolute thresholding can judge whether the target detection window is false-alarm, can guarantee effectively to suppress false-alarm in the situation of the normal access probability of multi-user.
Description of drawings
Fig. 1 is the structural representation of receiver basic function module;
Fig. 2 is the waveform schematic diagram of asking behind the mould square;
Fig. 3 is the method flow schematic diagram that the embodiment of the invention suppresses the random access code false-alarm.
Embodiment
For making the purpose, technical solutions and advantages of the present invention express clearlyer, the present invention is further described in more detail below in conjunction with drawings and the specific embodiments.
In embodiments of the present invention, no matter be in the situation that high s/n ratio or in frequency deviation, the energy that the non-target detection window of deduction is revealed in peak value just can obtain to revise peak value, judges according to revising peak value whether the user of target detection window appearance is false-alarm again.Can guarantee so effectively to suppress false-alarm in normal multi-user's the situation of access probability.
That the method flow schematic diagram that suppresses the random access code false-alarm specifically may further comprise the steps referring to accompanying drawing 3:
Peak value corresponding to random access code in step 301, the measurement target detection window.
At first, all detection window are carried out pre-detection, namely detect peak value or other detection limit in each detection window, such as energy etc.In detecting the detection window of peak value, determine the target detection window.Wherein, this peak value is greater than absolute thresholding.When high s/n ratio, distance detect have a detection window at least in K the detection window of detection window of peak value peak value greater than absolute thresholding, the detection window of then determining to detect peak value is the target detection window.In the LTE system, rule of thumb be worth general K less than or equal to 4.If certainly do not consider the problem of amount of calculation, for the accuracy that guarantees to calculate, then the value of K is the bigger the better.In the situation that frequency displacement occurs, determining here comprises that peak value is the target detection window greater than each detection window of thresholding.
Afterwards, corresponding peak value and the record of random access code in the measurement target detection window.
Step 302, described peak value are deducted non-target detection window and are revealed correction peak value corresponding to energy acquisition random access code;
Because therefore the energy that might have non-target detection window to reveal in the target detection window needs to revise the peak value that step 301 is measured.The energy that the peak value that random access code is corresponding is deducted non-target detection window leakage is correction peak value corresponding to random access code.Wherein, to reveal energy be that to have detected the product of inhibiting factor corresponding to random access code is corresponding in user's the non-target detection window peak value and this non-target detection window by at least one definite for non-target detection window.
Step 303, judge according to described correction peak value and absolute thresholding whether the user that described target detection window occurs is false-alarm, if false-alarm, then mark in described target detection window.
If the correction peak value that step 302 obtains, determines then that the target detection window is false-alarm, carries out mark less than absolute thresholding in this target detection window; Otherwise, determine it is not false-alarm.
Describe technical scheme of the present invention in detail below in conjunction with specific embodiment.
Embodiment one
In the situation that high s/n ratio, the corresponding detection window of root parameters u of current random access code has N (the detection window index is 1,2,3......, N).At first all detection window are carried out pre-detection, (flag=1 represents to detect peak value to obtain user's judgement sign flag; Flag=0 represents not detect peak value).
The user appears in K detection window, its pre-detection sign flag
K=1, K detection window may be real user, also may be because the leakage of its adjacent detection window energy causes.
The size of K user's peak value (or other detection limit) is Maxvalue
K, its revised peak value (or other detection limit) size is designated as Maxvalue_r
K, Maxvalue_r
KWith Maxvalue
KAfter partly deducting, middle possible adjacent window energy leakage obtains.
Transmitting terminal is launched a random access lead code, measures the peak value of each detection window.Through after processing, the index of target detection window is K at receiving terminal, and the size of its peak value (or other detection limit) is A
KBecause the distance of user and base station is different, the energy that causes being leaked to adjacent detection window can be slightly variant.The user nearer apart from the base station, the energy that the energy that is leaked to the K-1 detection window is leaked to the K+1 detection window relatively is larger, and the peak value that is leaked to the K-1 detection window is A
K-1, the first inhibiting factor G1=A so
K-1/ A
KThe energy that the energy that is leaked to the K-2 detection window is leaked to the K+2 detection window relatively is larger, and the peak value that is leaked to the K-2 detection window is A
K-2, the second inhibiting factor G2=A so
K-2/ A
K
When high s/n ratio and N 〉=1 o'clock, N is the number of detection window, and described inhibiting factor comprises G1 and G2.Corresponding first inhibiting factor of detection window that the detection window label is K+1 behind K-1 detection window and the target detection window; Corresponding second inhibiting factor of detection window that the detection window label is K+2 after K-2 detection window and the K+1 detection window.
N=4 has at most three adjacent windows preliminary examinations to measure the user, so:
Maxvalue_r
K=Maxvalue?
K-Maxvalue?
K+1*G1*flag?
K+1
-Maxvalue?
K-1*G1*flag?
K-1
-Maxvalue?
K+2*G2*flag?
K+2
N=3 has at most two adjacent windows preliminary examinations to measure the user, so:
Maxvalue_r
K=Maxvalue?
K-Maxvalue?
K+1*G1*flag?
K+1
-Maxvalue?
K-1*G1*flag?
K-1
N=2 has at most an adjacent windows preliminary examination to measure the user, so:
Maxvalue_r
K=Maxvalue?
K-Maxvalue?
K+1*G1*flag?
K+1
Wherein the flag value is 0 or 1, be 1 to be illustrated in that the user appears in this detection window in the pre-judgement, and be that the user does not appear in 0 expression.Considered adjacent windows possible influence each other after, carry out conclusive judgement: if Maxvalue_r
K<γ
α, then determine flag
KThe=0th, false-alarm; Otherwise, the pre-judgement result before keeping.
flag?
K=1。
Embodiment two
Example two all is applicable in the situation of high s/n ratio with embodiment one.In embodiment two, have 3 adjacent windows at least, i.e. N 〉=4.Because in the situation that concrete, when inhibiting factor comprised G1 and G2, amount of calculation was too large.For fear of a large amount of calculating, only consider affecting G2 and can make G2=0 between the adjacent windows.Identical among definite method of G1 and the embodiment one.Certainly do not consider amount of calculation, guaranteeing in the situation of precision, adopting the scheme among the embodiment one best.
Embodiment three
In the environment of frequency deviation (residual frequency departure and Doppler frequency shift), the detection window corresponding to the root parameters u detects, and obtains user's judgement sign flag.If flag
K=1, peak is Maxindex
KMaxindex
k± (d
u, 2d
u... .nd
u) all may cause other detection window false-alarm.But along with the increase of n, the energy of leakage can be more and more less.Therefore when using single window to detect, mainly consider apart from main peak value ± d
uThe energy leakage at place is on the impact of other detection window, and all users that can obtain flag=1 may cause the location sets Flasealarm_index of false-alarm.If need consideration ± 2d
u, 3d
u... .nd
uLocational energy leakage, can expand Flasealarm_index.
Detect if use three windows to merge, obtain user's judgement sign flag.When using three windows to merge detection, mainly consider apart from main peak value ± 2d
uThe energy leakage at place can obtain flag to the impact of other detection window
KAll users of=1 may cause the location sets Flasealarm_index of false-alarm.If need consideration ± 3d
u... .nd
uLocational energy leakage, can expand Flasealarm_index.
Certain user's flag
K=1, whether the false-alarm position of at first judging other user among the Flasealarm_index target approach detection window.If exist, number and the corresponding peak value thereof of record false-alarm position remove the energy that leaks into the target detection window; If there is no, then finish.Wherein, comprised the position that false-alarm may occur among the Flasealarm_index, the physical location of target detection window also is known, can judge whether the target approach detection window.
The size of K user's peak value (or other detection limit) is Maxvalue
k, its revised peak value (or other detection limit) size is designated as Maxvalue_r
kThere is M position to enter into the scope of target detection window, so
G
mFor there being the sequence inhibiting factor under the offset frequency situation.In the residential quarter of LTE low-speed mode, the user moves and can cause Doppler frequency shift, and the frequency deviation that produces is more stable.Only have a user to launch random access lead code at transmitting terminal, through after processing, the index of target detection window is K at receiving terminal, and the size of its peak value (or other detection limit) is A
K, the position is Maxindex
KUnder a certain fixedly frequency deviation, apart from Maxindex
KPositive d
uThe size of locational peak value (or other detection limit) be
, so
In like manner can calculate apart from Maxindex
KOther integral multiple d
uLocational sequence inhibiting factor.The sequence inhibiting factor can be done suitable adjustment according to the channel scene of reality and user's translational speed.
After the impact of having considered frequency deviation, carry out conclusive judgement: if Maxvalue_r
K<γ
α, then determine flag
KThe=0th, false-alarm; Otherwise, the pre-judgement structure flag before keeping
K=1.
The above is preferred embodiment of the present invention only, is not for limiting protection scope of the present invention.Within the spirit and principles in the present invention all, any modification of doing, be equal to replacement, improvement etc., all should be included within protection scope of the present invention.