CN101404517B - Signal frequency deviation correction method, apparatus and system - Google Patents

Abstract

The embodiment of the invention relates to a method for correcting the frequency swing of signals and a system and a device thereof. The method for correcting the frequency swing of signals disclosed by the embodiment of the invention comprises: correcting the frequency swing of the data part of a receiving signal based on the initial frequency swing estimation value of a training sequence in the receiving signal, calculating the frequency swing estimation value of the data part based on the symbol sequence of the data part of the receiving signal after correcting the frequency swing and carrying out frequency swing correction on the data part of the subsequent receiving signal according to the calculated frequency swing estimation value. The device for correcting the frequency swing of signals comprises a first correcting module and a second correcting module. The method, device and system for correcting the frequency swing of signals disclosed by the embodiment of the invention expand the frequency discrimination operation of the signals on the whole time slot of the receiving signals, thereby ensuring the frequency discrimination estimation value used for the frequency swing correction carried out on the symbols behind the data part in the receiving signal to have a smaller error than the frequency discrimination estimation value obtained only based on TSC estimation.

Description

The methods, devices and systems of signal frequency deviation correction

Technical field

The present invention relates to the mobile communication technology field, particularly a kind of methods, devices and systems of signal frequency deviation correction.

Background technology

Rapid is popularized in the whole world like high-speed railway fast, and scooter is 200～350 kilometers/hour during its highest operation.At present, as the commercial bullet train that domestic unique employing magnetic levitation technology is built, the F-Zero of Shanghai magnetic suspension train is especially up to 431 kilometers/hour.The use of these rapids, for existing GSM has brought very big challenge, a wherein very important aspect is exactly the stronger Doppler frequency shift problem that the travelling carriage high-speed motion is brought.

In openr scene, mostly the propagation between base station and the travelling carriage is line of sight or stronger reflection footpath mode, and the athletic meeting of travelling carriage causes the frequency of its signal that receives to squint, and is called Doppler frequency shift.The frequency deviation that travelling carriage receives downstream signal is f _d, $f_d=\frac{f_c}{C}*v*\mathrm{Cos}\mathrm{\θ},$ Wherein, f _cBe carrier frequency, C is a propagation velocity of electromagnetic wave, and v is the travelling carriage translational speed, and θ is the direction of motion of travelling carriage and the angle of the direction of propagation that receives downstream signal.As carrier frequency f _cBe 1.8GHz, when the movement speed v of travelling carriage is 300km/h, maximum frequency deviation f _dPossibly reach 500Hz.

The frequency of the downstream signal that the travelling carriage locking receives is such as being f _c+ f _dWith this frequency frequency as a reference, add the frequency offset Δ f of fixing uplink and downlink frequency, i.e. f _c+ f _d+ Δ f is as the carrier frequency of the upward signal of travelling carriage transmission; In like manner, because the upward signal that the athletic meeting of travelling carriage causes sending also produces Doppler frequency shift f _d, therefore, the carrier frequency of the upward signal that the base station receives is f _c+ Δ f+2f _dAnd the base station is with f _c+ Δ f is the local frequency reception of carrying out upward signal, and the baseband signal that will cause the base station to receive like this has 2f _dFrequency deviation.Under the scene for the translational speed of the carrier frequency of the top 1.8GHz that mentions and 300km/h, the Doppler frequency shift of the upward signal that the base station receives is exactly 1KHz.

The deviation of frequency can influence communication quality, generally, the calculating of frequency deviation correction signal through automatic frequency correction (Automatic Frequency Correction, hereinafter to be referred as: AFC) algorithm is accomplished.Channel for this type of dedicated channel sends continuously adopts loop AFC, to Burst Channel or the discontinuous transmitting channel of this type of shared channel, adopts single step AFC.

In realizing process of the present invention, the inventor finds, has following problems in the prior art at least:

The AFC method of prior art, the frequency offset estimating variance is generally bigger, if the frequency offset estimating variance is big, after signal rectified a deviation, still can have very strong residual frequency departure, thereby brings bigger demodulation performance loss.

Summary of the invention

The embodiment of the invention provides a kind of methods, devices and systems of signal frequency deviation correction, can reduce the frequency offset estimating variance, and the signal that receives is carried out correcting frequency deviation more accurately.

The embodiment of the invention provides a kind of method of signal frequency deviation correction, comprising:

Based on the initial frequency deviation estimated value that the training sequence that receives in the signal calculates, the frequency deviation of correcting the data division that receives signal;

Based on the symbol sebolic addressing of the data division of the said reception signal behind the correcting frequency deviation, the frequency offset estimating value of calculated data part;

The frequency offset estimating value of the said data division that use calculates is carried out correcting frequency deviation to the data division of follow-up reception signal.

The embodiment of the invention also provides a kind of device of signal frequency deviation correction, comprising:

The first correction module is used for the initial frequency deviation estimated value based on the training sequence calculating that receives signal, the frequency deviation of correcting the data division that receives signal;

The second correction module is used for the symbol sebolic addressing based on the data division of the said reception signal behind the correcting frequency deviation, the frequency offset estimating value of calculated data part; The frequency offset estimating value of the said data division that use calculates is carried out correcting frequency deviation to the data division of follow-up reception signal.

The embodiment of the invention also provides a kind of signal frequency deviation correction system; Comprise: the device of signal frequency deviation correction described above, and apparatus for transmitting signal, wherein; Said apparatus for transmitting signal is used to send the device that the signal with certain frequency arrives said signal frequency deviation correction.

Can know by above technical scheme; The methods, devices and systems of the signal frequency deviation correction of the embodiment of the invention; Can the frequency discrimination operation of signal be expanded on the whole time slot that receives signal; Be about to receive the TSC symbol sebolic addressing of signal and the symbol sebolic addressing of data division all is used for carrying out the frequency offset estimating of signal, thereby the employed frequency offset estimating value of the correcting frequency deviation that the symbol of the back of the data division in can guaranteeing to received signal carries out is than only estimating that based on TSC the frequency offset estimating initial value that obtains has littler error.

Description of drawings

Fig. 1 is the time slot format of the reception signal of the embodiment of the invention in the gsm system Traffic Channel;

Fig. 2 is the schematic flow sheet of first embodiment of the method for signal frequency deviation correction of the present invention;

Fig. 3 is the schematic flow sheet of second embodiment of the method for signal frequency deviation correction of the present invention;

Fig. 4 is the schematic flow sheet of the 3rd embodiment of the method for signal frequency deviation correction of the present invention;

Fig. 5 is the schematic flow sheet that embodiment of the invention cross product frequency discrimination method calculates the initial frequency deviation estimated value;

Fig. 6 A is the sketch map of 4 state trellis figure of the embodiment of the invention;

The sketch map that Fig. 6 B selects for embodiment of the invention survivor branch;

Fig. 7 is the local survivor path of the embodiment of the invention;

Fig. 8 is the structural representation of first embodiment of the device of signal frequency deviation correction of the present invention;

Fig. 9 is the structural representation of second embodiment of the device of signal frequency deviation correction of the present invention;

Figure 10 is the structural representation of signal frequency deviation correction system implementation example of the present invention.

Embodiment

Below through specific embodiment and combine accompanying drawing that the present invention is done further detailed description.

Because in common frequency correction process; The training sequence code of using (Train SequenceCode; Hereinafter to be referred as: TSC) shared length is shorter in receiving the time slot (Burst) of signal because its amount of information is not enough, can cause frequency discrimination after the frequency offset estimating variance bigger.Such as to global system for mobile communications (Global System for Mobile Communications; Hereinafter to be referred as: GSM); Fig. 1 is the time slot format of the reception signal of the embodiment of the invention in the gsm system Traffic Channel, and is as shown in Figure 1, and the TSC symbol of 26bit is only arranged in each time slot; Only adopt the TSC symbol of this 26bit to carry out frequency offset estimating, can cause the frequency offset estimating variance behind the frequency discrimination very big.The frequency offset estimating variance is big, after rectifying a deviation so to received signal, still can have very strong residual frequency departure, thereby brings the loss of the performance of bigger signal demodulation.In the following embodiment of the invention; Not only utilized the TSC symbol to carry out frequency offset estimating; Also introduced to adopt and handled (Per-Survivor Processing by survivor path; Hereinafter to be referred as: PSP) method data (DATA) part is to received signal carried out frequency offset estimating, thereby has reduced the variance of frequency offset estimating.

Fig. 2 is for the schematic flow sheet of first embodiment of the method for signal frequency deviation correction of the present invention, and is as shown in Figure 2, comprises the steps:

Step 201, the initial frequency deviation estimated value of calculating based on the training sequence (TSC symbol sebolic addressing) that receives in the signal, the frequency deviation of correcting the data division that receives signal;

Step 202, based on the symbol sebolic addressing of the data division of the reception signal behind the correcting frequency deviation, the frequency offset estimating value of calculated data part;

Step 203, use the frequency offset estimating value of the data division calculate, the data division of follow-up reception signal is carried out correcting frequency deviation.

The method of the signal frequency deviation correction that present embodiment provides; Can the frequency discrimination operation of signal be expanded on the whole time slot that receives signal; Be about to receive the TSC symbol sebolic addressing of signal and the symbol sebolic addressing of data division all is used for carrying out the frequency offset estimating of signal, thereby the employed frequency offset estimating value of the correcting frequency deviation that the symbol of the back of the data division in can guaranteeing to received signal carries out is than only estimating that based on TSC the frequency offset estimating initial value that obtains has littler error.

Fig. 3 is for the schematic flow sheet of second embodiment of the method for signal frequency deviation correction of the present invention, and is as shown in Figure 3, comprises the steps:

Step 301, calculate the initial frequency deviation estimated value based on the training sequence (TSC symbol sebolic addressing) that receives in the signal;

Receiving equipment, like portable terminal, receive signal after, can calculate the initial frequency deviation estimated value based on the TSC symbol sebolic addressing on the known TSC sequence of this receiving equipment and the time slot as shown in Figure 1 that receives;

Step 302, utilize initial p symbol of initial frequency deviation estimated value data division (DATA) to received signal to carry out correcting frequency deviation;

In the embodiment of the invention, after receiving equipment receives signal, partly be that bit of a bit carries out correcting frequency deviation to the DATA on the time slot as shown in Figure 1; But because the calculating of frequency offset estimating value is comparatively complicated, therefore can set one-period, promptly can take every p symbol to utilize same frequency offset estimating value to carry out correcting frequency deviation for the DATA part that receives signal; In this step 302, promptly with the initial frequency deviation estimated value preceding p the symbol of DATA carried out correcting frequency deviation, wherein p is a positive integer;

Step 303, based on the frequency offset estimating value of N the symbol of preceding N-q sign computation of the data division that receives signal, wherein, N=p+1, p+2, p+3 ..., q is the positive integer littler than N;

Step 304, utilize p the symbol from N sign-on of the frequency offset estimating value data division to received signal of N symbol to carry out correcting frequency deviation;

In above-mentioned steps 302, adopted preceding p symbol of initial frequency deviation estimated value data portion to carry out correcting frequency deviation, suppose p=5; In step 303 and step 304,, utilize the frequency offset estimating value of the 6th symbol that 5 symbols from the 6th sign-on are carried out correcting frequency deviation based on the frequency offset estimating value of the 6th symbol of preceding 5 sign computation behind the correcting frequency deviation; By that analogy; If N=16 can pass through selectable q value, suppose q=7; Promptly adopt DATA part several 9 symbols before the 16th symbol; The symbol sebolic addressing of promptly forming from 15 symbols of the 7th symbol to the calculates the frequency offset estimating value of the 16th symbol, thereby to back 5 symbols from the 16th sign-on, promptly 20 symbols of the 16th symbol to the carry out correcting frequency deviation; The correcting frequency deviation of the total data during final receiving equipment can be accomplished to the received signal.

The method of the signal frequency deviation correction that present embodiment provides; Expanded and utilized the data division that receives signal to calculate the frequency offset estimating value; Make the frequency offset estimating value of using when carrying out correcting frequency deviation to received signal have than only estimating that based on the TSC symbol sebolic addressing frequency offset estimating value that obtains has littler error; Therefore; The frequency offset estimating variance of the frequency offset estimating value that the method through this signal frequency deviation correction obtains can significantly reduce, thereby can carry out correcting frequency deviation comparatively accurately to received signal, to reduce the receiving equipment loss of demodulation performance to received signal that frequency deviation causes.

Fig. 4 is for the schematic flow sheet of the 3rd embodiment of the method for signal frequency deviation correction of the present invention, and is as shown in Figure 4, comprises the steps:

Step 401, carry out channel estimating and slot synchronization to received signal, reception signal and channel estimation value after obtaining synchronously;

Receiving equipment is carrying out in the process of signal processing to received signal, at first will carry out channel estimating and slot synchronization to received signal, with reception signal and the channel estimation value after obtaining synchronously.

Step 402, the TSC symbol sebolic addressing in the reception signal after is synchronously carried out frequency discrimination, obtain the initial frequency deviation estimated value;

Calculate the initial frequency deviation estimated value with the cross product frequency discrimination method below.Fig. 5 is the schematic flow sheet that embodiment of the invention cross product frequency discrimination method calculates the initial frequency deviation estimated value, and is as shown in Figure 5.Wherein, the TSC symbol sebolic addressing receiving symbol partly that receives in the signal is:

$r_k={\mathrm{<figure-callout\; id="2"\; label="e\; j"\; filenames="H200810226092XE00032.png,H200810226092XE00041.png"\; state="\{\{state\}\}">e</figure-callout>}}^j\underset{l=0}{\overset{L}{\mathrm{\&Sigma;}}}{h}_l{I}_{k-l}+n_k---\left(1\right)$

Wherein, k is constantly, T _sBe symbol period, Δ f is the frequency offset of uplink and downlink frequency,

Be frequency deviation modulation signal, h ₁Be channel factors, I _K-lBe the discrete message symbol sebolic addressing that receives, n _kBe white Gaussian noise, L is a channel factors length.

Step 501, utilize channel estimation value

and send symbol

, convolution obtains reference signal:

$s_k^{\mathrm{ref}}=\underset{l=0}{\overset{L}{\mathrm{\&Sigma;}}}{\hat{h}}_l{\hat{I}}_{k-l}---\left(2\right)$

Wherein, Channel estimation value

obtains when carrying out channel estimating to received signal; Sending symbol

is the TSC symbol sebolic addressing, and this TSC symbol sebolic addressing is that receiving equipment is known;

Step 502, will receive signal r _kAnd reference signal

Conjugate multiplication can obtain difference frequency signal

${\hat{x}}_k=r_k{\left(s_k^{\mathrm{ref}}\right)}^{*}---\left(3\right)$

Step 503, two certain difference frequency signals at interval carried out cross product calculate, obtain cross product value R (m):

$R\left(m\right)=\frac{1}{L_0-m}\underset{k=m}{\overset{L_0-1}{\mathrm{\&Sigma;}}}{x}_k{x}_{k-m}^{*}---\left(4\right)$

Wherein m is the cross product interval, L ₀Expression is used for the length of the difference frequency symbol sebolic addressing of cross product calculating.

Step 504, based on the cross product value R (m) that step 503 obtains, can obtain initial frequency deviation estimated value

through asking phase place

$\mathrm{\&Delta;}\hat{f}=\frac{1}{2\mathrm{\&pi;m}{T}_s}\arg \left(R\left(m\right)\right)---\left(5\right)$

Wherein phase place, T are asked in arg () expression _sBe symbol period.

The method of carrying out frequency discrimination based on cross product has a lot, such as Fitz algorithm, Luise algorithm etc.Can carry out the method for frequency discrimination based on cross product with above-mentioned in the embodiment of the invention, but be not limited to the method, calculate the initial frequency deviation estimated value.

Step 403, utilize initial p symbol of initial frequency deviation estimated value data division to received signal to carry out correcting frequency deviation, p is the correcting frequency deviation cycle, and p is a positive integer;

Step 404, utilize the method for grid search, confirm the survivor branch under the corresponding trellis state of N symbol;

Carry out correcting frequency deviation except initial p the symbol that adopts initial frequency deviation estimated value that known TSC symbol sebolic addressing the calculates data division in to received signal in the embodiment of the invention; Also to utilize the symbol sebolic addressing of data division to calculate the frequency offset estimating value; To bring in constant renewal in the frequency offset estimating value, the successive character of data portion carries out correcting frequency deviation.Yet the symbol sebolic addressing of the data division in the reception signal is not known, need in grid search, find DATA part symbol sebolic addressing partly based on the PSP method, to calculate the frequency offset estimating value according to the symbol sebolic addressing that obtains.

For the ease of explanation, Fig. 6 A is the sketch map of 4 state trellis figure of the embodiment of the invention.Figure is an example with this 4 state trellis, shown in Fig. 6 A, and stain, promptly each node on this grid chart is represented the state of register, comes mark with 0,1,2,3 respectively, k, k+1 ..., k+4 is each moment point.The bit value of the symbol that solid line between each state of register and dotted line are represented respectively to import is 0 and the transfer of the state of 1 o'clock register, and the bit of the symbol of the numeral on the line " 0/00 " expression input is that 0 o'clock output bit is " 00 ".The grid search method promptly solves the data that receive the signal through the transfer of register from a state to another state of searching on this 4 state trellis figure, thereby obtains symbol sebolic addressing, for example adopts Viterbi (Viterbi) algorithm.And the transfer of register from a state to another state that search is confirmed is survivor branch, and this survivor branch specifically can be through the relatively acquisition of branch metric, and branch metric is represented the probability that shifts between different conditions.

The sketch map that Fig. 6 B selects for embodiment of the invention survivor branch is shown in Fig. 6 B, for example for the corresponding trellis state S of N symbol _{K, 0}Under the selection of survivor branch.K is state S constantly _{K, 0}The selection course of survivor branch, S here _{K, i}Expression k i state constantly.λ (S among the figure _K-1,0→ S _{K, 0}) be from k-1 moment state S _K-1,0Transfer to k state S constantly _{K, 0}Branch metric, it is by the data and the S that receive in the signal _K-1,0Decision.M _{K-1, i}(i=0,1,2,3) are state S _{K-1, i}The cumulative metric of (i=0,1,2,3) is from 0 to k-1 moment state S _{K-1, i}The summation of the branch metric of all survivor branch on the survivor path of (i=0,1,2,3).Suppose M _K-1,0+ λ (S _K-1,0→ S _{K, 0})<m _K-1,2+ λ (S _K-1,2→ S _{K, 0}), the S of branch then _K-1,2→ S _{K, 0}Be exactly S _{K, 0}Survivor branch, S _{K, 0}Cumulative metric M _{K, 0}=M _K-1,2+ λ (S _K-1,2→ S _{K, 0}).Other states to synchronization are made same treatment, just can find each survivor branch of each state constantly.In embodiments of the present invention, with k moment state S _{K, 0}The state corresponding as N symbol is the selection that example is carried out survivor branch.

Step 405, recall the local survivor path under the corresponding trellis state of N symbol along survivor branch;

Wherein, local survivor path is made up of the survivor branch under the trellis state of preceding N-q symbol correspondence of the data division that receives signal.

In the present embodiment, promptly along k moment state S _{K, 0}Survivor branch S _K-1,2→ S _{K, 0}Recall, confirm k-1 state S constantly _K-1,2Survivor branch, date back to the corresponding k-m of N-q symbol state S constantly always _{K-m, i}, promptly confirm k-m+1 state S constantly _{K-m+1, i}Survivor branch; Then all survivor branch are connected, just obtained N-q symbol to the local survivor path between N the symbol.Fig. 7 is the local survivor path of the embodiment of the invention, wherein the k moment the 0th state S _{K, 0}Survivor branch be b ₀₀, along this survivor branch b ₀₀Recall, obtain S _{K, 0}Local survivor path, promptly shown in the overstriking path 701 among Fig. 7.

The PSP method is in the process of grid search, on the survivor branch of each each state constantly or survivor path, carries out the processing such as parameter Estimation.Being presented as in embodiments of the present invention; Be based on the local survivor path that adopts the grid search method to obtain on the above-mentioned 4 state trellis figure that give an example; Utilize this PSP method, the symbol sebolic addressing of the data division of the reception signal that obtains is carried out frequency offset estimating, promptly frequency discrimination is handled.

Step 406, according to the local survivor path that obtains, obtain the symbol sebolic addressing under the trellis state of N symbol correspondence, symbol sebolic addressing is that N-q symbol is to N symbol;

Promptly illustrated like above-mentioned step 404, through grid search decoded data symbol.The sequence of setting all states on the local survivor path 701 as shown in Figure 7 does

Can be obtained the symbol of the data division in the reception signal of corresponding each state by the survivor branch of each state on this Fig. 7, the assemble of symbol of the data division in the reception signal that each survivor branch is obtained then obtains k state S constantly _{K, 0}The corresponding symbol sequence

Promptly receive the symbol between the N-q to N of data division of signal.

Step 407, reception signal and channel estimation value according to symbol sebolic addressing, after synchronously calculate the difference frequency signal under this trellis state;

Reception signal and channel estimation value according to the symbol sebolic addressing that obtains, after synchronously, the computational methods according to the cross product frequency discrimination of above-mentioned Fig. 4 calculate reference signal by (2) formula earlier $s_k^{\mathrm{Ref}}=\underset{l=0}{\overset{L}{\mathrm{\&Sigma;}}}{\hat{h}}_l{\hat{I}}_{k-l};$ Again by receiving signal r _kAnd reference signal Calculate state S according to (3) formula _{K, 0}Difference frequency signal ${\hat{x}}_{k,0}=r_k{\left(\underset{l=0}{\overset{L}{\mathrm{\&Sigma;}}}{\hat{h}}_l{\hat{I}}_{k-l}\right)}^{*}.$ Wherein

Be the symbol sebolic addressing that adopts the data division that above-mentioned grid search method obtains.

Step 408, the difference frequency signal of difference frequency signal under this trellis state and the initial moment corresponding states of local survivor path is carried out cross product, obtain the cross product value under the corresponding trellis state of N symbol;

Cross product value under the trellis state that N-1 symbol is corresponding on the said survivor path under cross product value under the corresponding trellis state of step 409, said N symbol that will obtain and this trellis state is carried out smothing filtering;

In the present embodiment, step 408 and step 409 are specially, and establish among Fig. 7, and k is state S constantly _{K, 0}Local survivor path on k-m state S constantly _{K-m, 1}On difference frequency signal do

According to (4) formula, and adopt a kind of method of smothing filtering,, obtain S such as adopting Alpha (alpha) filter _{K, 0}On the cross product value

${\hat{R}}_{k,0}=\mathrm{\&alpha;}{\hat{R}}_{k-\mathrm{1,0}}+(1-\mathrm{\&alpha;}){\hat{x}}_{k,0}{\hat{x}}_{k-m,{\hat{s}}_{k-m}}^{*}---\left(6\right)$

Need to prove that in step 408, can adopt above-mentioned steps to obtain based on the cross product value under the trellis state of each data symbol correspondence on the survivor path, promptly each data symbol all need calculate the cross product value; And the process of in the step 409 the cross product value of some data symbols being carried out smothing filtering can be that each data symbol is all carried out, but in order to simplify computing, preferably the cross product value of that symbol that will calculate the frequency offset estimating value is carried out smothing filtering.In addition, it should be noted that here to describe that other frequency discrimination method such as Fitz, Luise method etc., also is applicable to the embodiment of the invention fully with simple monospace cross product method.

Step 410, according to the cross product value, calculate the frequency offset estimating value under the corresponding trellis state of N symbol;

In this step 410, be the frequency offset estimating value that can calculate each data symbol according to the cross product value that aforementioned calculation obtains in theory, but for fear of a large amount of processing expenditure of asking phase calculation to cause, cycle that can set-up and calculated frequency offset estimating value.That is, if N the trellis state that symbol is corresponding, k is state S constantly _{K, 0}The frequency offset estimating value update cycle T of moment corresponding k for setting _UpdateMoment of integral multiple, this update cycle T _UpdatePromptly corresponding p symbol then according to the cross product value, obtains the frequency offset estimating value under the trellis state of this moment N symbol correspondence through asking phase place; Other corresponding constantly state is then continued to use previous moment on its local survivor path, continues to use the frequency offset estimating value under the state corresponding on the k-2 local survivor path of the moment constantly like k-1.K among Fig. 7 is state S constantly _{K, 0}The frequency offset estimating value be:

$2\mathrm{\&pi;\&Delta;}{\hat{f}}_{k,0}{T}_s=\left\{\begin{array}{ccc}\frac{1}{m}\arg \left({\hat{R}}_{k,0}\right),& \mathrm{if}& k\mathrm{mod}{T}_{\mathrm{update}}=0\\ 2\mathrm{\&pi;\&Delta;}{\hat{f}}_{k-\mathrm{1,0}}{T}_s,& \mathrm{else}& \end{array}\right.---\left(7\right)$

Step 411, utilize p the symbol from N sign-on of the frequency offset estimating value data division to received signal of N symbol to carry out correcting frequency deviation.

For example in the application of the selection of following survivor branch, promptly can use the method that adopts above-mentioned steps 410 to calculate the frequency offset estimating values to rectify a deviation to received signal.Suppose for k moment state S among Fig. 7 _{K, 0}, have 2 branches to get into this state S _{K, 0}, be designated as b ₀₀And b ₀₂, its initial state is respectively S _K-1,0And S _K-1,2, when selecting survivor branch, can applying step 410 calculate the method for frequency offset estimating values, the frequency offset estimating value that adopts k-1 to calculate constantly in Branch Computed tolerance

R to received signal _kCarry out correcting frequency deviation, and then the branch metric that calculates is used to select survivor branch, this branch metric calculation is following:

$\mathrm{\&lambda;}\left(b_{00}\right)={|r_k-{\mathrm{<figure-callout\; id="2"\; label="e\; j"\; filenames="H200810226092XE00032.png,H200810226092XE00041.png"\; state="\{\{state\}\}">e</figure-callout>}}^j\underset{l=0}{\overset{L}{\mathrm{\&Sigma;}}}{\hat{h}}_l{I}_{k-l}\left(b_{00}\right)|}^2$

，k>0(8)

$\mathrm{\&lambda;}\left(b_{02}\right)={|r_k-{\mathrm{<figure-callout\; id="2"\; label="e\; j"\; filenames="H200810226092XE00032.png,H200810226092XE00041.png"\; state="\{\{state\}\}">e</figure-callout>}}^j\underset{l=0}{\overset{L}{\mathrm{\&Sigma;}}}{\hat{h}}_l{I}_{k-l}\left(b_{02}\right)|}^2$

Wherein offset is the time offset in the correcting frequency deviation, and its foundation of confirming is in order to eliminate the influence of first phase.In the calculating of formula (8) to branch metric; If calculate k=0 branch metric constantly; Being first symbol of data portion will carry out correcting frequency deviation the time, then generally is to adopt the initial frequency deviation estimated value that partly obtains based on the TSC symbol sebolic addressing to carry out correcting frequency deviation.

The method of the signal frequency deviation correction that present embodiment provides; Can will adopt frequency discrimination method to calculate the frequency offset estimating value expands on the whole time slot that receives signal; The TSC symbol sebolic addressing and the data division that are about to receive signal all are used for carrying out the calculating of frequency offset estimating; Especially data portion was carried out based on the moment of PSP method at the integral multiple of the update cycle of setting, and the cross product value that adopts frequency discrimination to obtain calculates the frequency offset estimating value of renewal; And the frequency offset estimating value that calculates is used for the frequency deviation of the reception signal of state corresponding on next survivor branch is constantly corrected, to reduce the receiving equipment loss of demodulation performance to received signal that frequency deviation causes.

Fig. 8 is the structural representation of first embodiment of the device of signal frequency deviation correction of the present invention, and as shown in Figure 8, this device comprises: the first correction module 80 and the second correction module 80 '.Wherein the first correction module 80 is used for based on the initial frequency deviation estimated value that receives the training sequence calculating in the signal, the frequency deviation of correcting the data division that receives signal; The second correction module 80 ' be used for is based on the symbol sebolic addressing of the data division of the reception signal behind the correcting frequency deviation, the frequency offset estimating value of calculated data part; The frequency offset estimating value of the data division that use calculates is carried out correcting frequency deviation to the data division of follow-up reception signal.

Wherein the first correction module 80 can specifically comprise: based on first computing module 81 that receives the training sequence calculating initial frequency deviation estimated value in the signal; And initial p the symbol that utilizes initial frequency deviation estimated value data division to received signal carry out the first correcting frequency deviation module 82 of correcting frequency deviation, and p is the correcting frequency deviation cycle, and p is a positive integer.The second correction module 80 ' can specifically comprise: based on second computing module 83 of the frequency offset estimating value of N the symbol of preceding N-q sign computation of the data division that receives signal, wherein, and N=p+1, p+2, p+3 ..., q is the positive integer littler than N; Utilize p the symbol from N sign-on of the frequency offset estimating value data division to received signal of N symbol to carry out the second correcting frequency deviation module 84 of correcting frequency deviation.

The device of the signal frequency deviation correction that present embodiment provides has been expanded and has been utilized the data division that receives signal to calculate the frequency offset estimating value, and the detailed process of handling to received signal through each module can realize through the description of the concrete grammar step among the method embodiment of above-mentioned signal frequency deviation correction.

The device of the signal frequency deviation correction that present embodiment provides has been expanded and has been utilized the data division that receives signal to calculate the frequency offset estimating value; Make the frequency offset estimating value of using when carrying out correcting frequency deviation to received signal have than only estimating that based on the TSC symbol sebolic addressing frequency offset estimating value that obtains has littler error; Therefore; The frequency offset estimating variance of the frequency offset estimating value that the method through this signal frequency deviation correction obtains can significantly reduce; Thereby can carry out correcting frequency deviation comparatively accurately to received signal, to reduce the receiving equipment loss of demodulation performance to received signal that frequency deviation causes.

Fig. 9 is the structural representation of second embodiment of the device of signal frequency deviation correction of the present invention, and as shown in Figure 9, this device comprises: the first correction module 80 and the second correction module 80 '.Wherein the first correction module 80 comprises first computing module 81 and the first correcting frequency deviation module 82, the second rectify a deviation modules 80 ' comprise again second computing module 83 and the second correcting frequency deviation module 84 again.

Wherein second computing module 83 can also specifically comprise: utilize the method for grid search, confirm the survivor branch determination module 831 of the survivor branch under the corresponding trellis state of N symbol; Recall the survivor path acquisition module 832 of the local survivor path under the corresponding trellis state of N symbol along survivor branch, this part survivor path is made up of the survivor branch under the trellis state of preceding N-q symbol correspondence of the data division that receives signal; According to local survivor path, obtain the symbol sebolic addressing acquisition module 833 of the symbol sebolic addressing under the corresponding trellis state of N symbol, this symbol sebolic addressing is that N-q symbol is to N symbol; Reception signal and channel estimation value according to symbol sebolic addressing, after synchronously calculate the difference frequency signal computing module 834 of the difference frequency signal under this trellis state; The difference frequency signal of difference frequency signal under this trellis state and the initial moment corresponding states of local survivor path is carried out cross product, obtain the cross product value computing module 835 of the cross product value under the corresponding trellis state of N symbol; According to the cross product value, calculate the frequency offset estimating computing module 836 of the frequency offset estimating value under the corresponding trellis state of N symbol.

Wherein second computing module 83 can also comprise: smothing filtering module 837, the cross product value that is used on the survivor path under cross product value and this trellis state under the trellis state that N the symbol that obtains through cross product value computing module 835 is corresponding under the corresponding trellis state of N-1 symbol is carried out smothing filtering.

The device of the signal frequency deviation correction that present embodiment provides can will adopt frequency discrimination method to calculate the frequency offset estimating value and expand on the whole time slot that receives signal, and the detailed process of handling to received signal through each module can realize through the description of the concrete grammar step among the method embodiment of above-mentioned signal frequency deviation correction.

The device of the signal frequency deviation correction that present embodiment provides; Can will adopt frequency discrimination method to calculate the frequency offset estimating value expands on the whole time slot that receives signal; The TSC symbol sebolic addressing and the data division that are about to receive signal all are used for carrying out the calculating of frequency offset estimating; Especially data portion was carried out based on the moment of PSP method at the integral multiple of the update cycle of setting, and the cross product value that adopts frequency discrimination to obtain calculates the frequency offset estimating value of renewal; And the frequency offset estimating value that calculates is used for the frequency deviation of the reception signal of state corresponding on next survivor branch is constantly corrected, to reduce the receiving equipment loss of demodulation performance to received signal that frequency deviation causes.

An alternative embodiment of the invention also provides a kind of signal frequency deviation correction system; This system comprises: the device of signal frequency deviation correction described above, and apparatus for transmitting signal, wherein; Apparatus for transmitting signal is used for sending signal with certain frequency device to this signal frequency deviation correction.Below in conjunction with accompanying drawing this system is described.

Figure 10 is the structural representation of signal frequency deviation correction system implementation example of the present invention, and shown in figure 10, this system comprises: the apparatus for transmitting signal 9 that sends the signal with certain frequency; And the signal that receives said apparatus for transmitting signal transmission with certain frequency; And carry out the signal receiver 8 of correcting frequency deviation to received signal; This signal receiver 8 can be the device of aforesaid signal frequency deviation correction, or the network element of the device of integrated aforesaid signal frequency deviation correction.

Wherein, this signal receiver 8 can comprise: based on the initial frequency deviation estimated value that the training sequence that receives in the signal calculates, correct the first correction module 80 of the frequency deviation of the data division that receives signal; And based on the symbol sebolic addressing of the data division of the reception signal behind the correcting frequency deviation, the frequency offset estimating value of calculated data part; The frequency offset estimating value of the data division that use calculates, to the data division of follow-up reception signal carry out the second correction module 80 of correcting frequency deviation '.

Wherein the first correction module 80 can specifically comprise: based on first computing module 81 that receives the training sequence calculating initial frequency deviation estimated value in the signal; And initial p the symbol that utilizes initial frequency deviation estimated value data division to received signal carry out the first correcting frequency deviation module 82 of correcting frequency deviation, and p is the correcting frequency deviation cycle, and p is a positive integer.The second correction module 80 ' can specifically comprise: based on second computing module 83 of the frequency offset estimating value of N the symbol of preceding N-q sign computation of the data division that receives signal, wherein, and N=p+1, p+2, p+3 ..., q is the positive integer littler than N; Utilize p the symbol from N sign-on of the frequency offset estimating value data division to received signal of N symbol to carry out the second correcting frequency deviation module 84 of correcting frequency deviation.

This signal receiver 8 can also comprise: processing module 85; Be used for the reception signal that is sent by apparatus for transmitting signal 9 that this signal receiver 8 receives is carried out channel estimating and slot synchronization; Reception signal and channel estimation value after obtaining synchronously calculate the initial frequency deviation estimated value with the training sequence in the reception signal after obtaining synchronously by first computing module 81.

Wherein second computing module 83 can comprise: utilize the method for grid search, confirm the survivor branch determination module 831 of the survivor branch under the corresponding trellis state of N symbol; Recall the survivor path acquisition module 832 of the local survivor path under the corresponding trellis state of N symbol along survivor branch, this part survivor path is made up of the survivor branch under the trellis state of preceding N-q symbol correspondence of the data division that receives signal; According to local survivor path, obtain the symbol sebolic addressing acquisition module 833 of the symbol sebolic addressing under the corresponding trellis state of N symbol, this symbol sebolic addressing is that N-q symbol is to N symbol; Reception signal and channel estimation value according to symbol sebolic addressing, after synchronously calculate the difference frequency signal computing module 834 of the difference frequency signal under this trellis state; The difference frequency signal of difference frequency signal under this trellis state and the initial moment corresponding states of local survivor path is carried out cross product, obtain the cross product value computing module 835 of the cross product value under the corresponding trellis state of N symbol; According to the cross product value, calculate the frequency offset estimating computing module 836 of the frequency offset estimating value under the corresponding trellis state of N symbol.

Wherein second computing module 83 can also comprise: smothing filtering module 837, the cross product value that is used on the survivor path under cross product value and this trellis state under the trellis state that N the symbol that obtains through cross product value computing module 835 is corresponding under the corresponding trellis state of N-1 symbol is carried out smothing filtering.

The apparatus for transmitting signal in the signal frequency deviation correction system that present embodiment provides and the concrete course of work of signal receiver can realize through the description of the concrete grammar step among the method embodiment of above-mentioned signal frequency deviation correction.

The signal frequency deviation correction system that present embodiment provides can expand to the frequency discrimination operation to signal on the time slot of whole reception signal; Be about to receive the TSC symbol sebolic addressing of signal and the symbol sebolic addressing of data division all is used for carrying out the frequency offset estimating of signal, thereby the employed frequency offset estimating value of the correcting frequency deviation that the symbol of the back of the data division in can guaranteeing to received signal carries out is than only estimating that based on TSC the frequency offset estimating initial value that obtains has littler error.

One of ordinary skill in the art will appreciate that all or part of flow process that realizes in the foregoing description method; Be to instruct relevant hardware to accomplish through computer program; Described program can be stored in a computer and can obtain in the storage medium; This program can comprise the flow process like the embodiment of above-mentioned each side method when carrying out.Wherein, described storage medium can be magnetic disc, CD, read-only storage memory body (Read-Only Memory, ROM) or at random store memory body (Random AccessMemory, RAM) etc.

What should explain at last is: above embodiment is only in order to explaining technical scheme of the present invention, but not to its restriction; Although with reference to previous embodiment the present invention has been carried out detailed explanation, those of ordinary skill in the art is to be understood that: it still can be made amendment to the technical scheme that aforementioned each embodiment put down in writing, and perhaps part technical characterictic wherein is equal to replacement; And these are revised or replacement, do not make the spirit and the scope of the essence disengaging various embodiments of the present invention technical scheme of relevant art scheme.

Claims (9)

1. the method for a signal frequency deviation correction is characterized in that, comprising:

Based on the initial frequency deviation estimated value that the training sequence that receives in the signal calculates, correct the frequency deviation that receives in the signal with the data division of said training sequence in same time slot;

Based on the symbol sebolic addressing of the data division of the said reception signal behind the correcting frequency deviation, the frequency offset estimating value of calculated data part;

The frequency offset estimating value of the said data division that use calculates is carried out correcting frequency deviation to the data division of follow-up reception signal;

Wherein, the said initial frequency deviation estimated value of calculating based on the training sequence that receives in the signal, the frequency deviation of correcting the data division that receives signal comprises:

Training sequence based on receiving in the signal calculates the initial frequency deviation estimated value;

Utilize said initial frequency deviation estimated value that initial p symbol of the data division of said reception signal carried out correcting frequency deviation;

The symbol sebolic addressing of said data division based on the said reception signal behind the correcting frequency deviation, the frequency offset estimating value of calculated data part comprises:

Based on the frequency offset estimating value of N the symbol of preceding N-q sign computation of N symbol of the data division of the said reception signal behind the correcting frequency deviation, wherein, N=p+1, p+2, p+3 ..., p is the correcting frequency deviation cycle, and p is a positive integer, and q is the positive integer littler than N;

The frequency offset estimating value of the said data division that said use calculates, the data division of follow-up reception signal is carried out correcting frequency deviation comprise:

The frequency offset estimating value of utilizing said N symbol is carried out correcting frequency deviation to p the symbol from N sign-on of the data division of said reception signal.

2. the method for signal frequency deviation correction according to claim 1 is characterized in that, saidly calculates the initial frequency deviation estimated value and comprises based on receiving training sequence in the signal:

Said reception signal is carried out channel estimating and slot synchronization, reception signal and channel estimation value after the acquisition synchronously;

Training sequence in the said reception signal after synchronously carries out frequency discrimination, obtains said initial frequency deviation estimated value.

3. the method for signal frequency deviation correction according to claim 2 is characterized in that, the frequency offset estimating value of N the symbol of preceding N-q sign computation of said data division based on the said reception signal behind the correcting frequency deviation comprises:

Utilize the method for grid search, confirm the survivor branch under the corresponding trellis state of said N symbol;

Recall the local survivor path under the corresponding trellis state of said N symbol along said survivor branch, said local survivor path is made up of the survivor branch under the trellis state of preceding N-q symbol correspondence of the data division of said reception signal;

According to said local survivor path, obtain the symbol sebolic addressing under the corresponding trellis state of said N symbol, said symbol sebolic addressing is that N-q symbol is to N symbol;

According to said symbol sebolic addressing, said reception signal and said channel estimation value after synchronously, calculate the difference frequency signal under this trellis state;

The difference frequency signal of difference frequency signal under said this trellis state and the initial moment corresponding states of said local survivor path is carried out cross product, obtain the cross product value under the corresponding trellis state of said N symbol;

According to said cross product value, calculate the frequency offset estimating value under the corresponding trellis state of said N symbol.

4. the method for signal frequency deviation correction according to claim 3; It is characterized in that, between the frequency offset estimating value under the said trellis state that obtains cross product value and said N the symbol correspondence of said calculating under the corresponding trellis state of said N symbol, also comprise:

Cross product value under the trellis state that N-1 symbol is corresponding on the said survivor path under cross product value under the trellis state that said N the symbol that obtains is corresponding and this trellis state is carried out smothing filtering.

5. according to the method for claim 3 or 4 described signal frequency deviation corrections, it is characterized in that said according to said cross product value, the frequency offset estimating value of calculating under the corresponding trellis state of said N symbol comprises:

If the trellis state moment corresponding of said N symbol correspondence is the moment of the integral multiple of the frequency offset estimating value update cycle of setting; Then, obtain the frequency offset estimating value under the corresponding trellis state of said N symbol of this moment through asking phase place according to said cross product value.

6. the device of a signal frequency deviation correction is characterized in that, comprising:

The first correction module is used for the initial frequency deviation estimated value calculated based on the training sequence that receives signal, corrects the frequency deviation that receives in the signal with the data division of said training sequence in same time slot;

The second correction module is used for the symbol sebolic addressing based on the data division of the said reception signal behind the correcting frequency deviation, the frequency offset estimating value of calculated data part; The frequency offset estimating value of the said data division that use calculates is carried out correcting frequency deviation to the data division of follow-up reception signal;

Wherein, the said first correction module comprises:

First computing module is used for calculating the initial frequency deviation estimated value based on the training sequence that receives signal;

The first correcting frequency deviation module is used to utilize said initial frequency deviation estimated value that initial p symbol of the data division of said reception signal carried out correcting frequency deviation;

The said second correction module comprises:

Second computing module is used for the frequency offset estimating value based on N the symbol of preceding N-q sign computation of N symbol of the data division of said reception signal, wherein, and N=p+1; P+2, p+3 ...; P is the correcting frequency deviation cycle, and p is a positive integer, and q is the positive integer littler than N;

The second correcting frequency deviation module, the frequency offset estimating value that is used to utilize said N symbol is carried out correcting frequency deviation to p the symbol from N sign-on of the data division of said reception signal.

7. the device of signal frequency deviation correction according to claim 6 is characterized in that, said second computing module comprises:

The survivor branch determination module is used to utilize the method for grid search, confirms the survivor branch under the corresponding trellis state of said N symbol;

The survivor path acquisition module; Be used for recalling the local survivor path under the corresponding trellis state of said N symbol along said survivor branch, said local survivor path is made up of the survivor branch under the trellis state of preceding N-q symbol correspondence of the data division of said reception signal;

The symbol sebolic addressing acquisition module is used for according to said local survivor path, obtains the symbol sebolic addressing under the corresponding trellis state of said N symbol, and said symbol sebolic addressing is that N-q symbol is to N symbol;

The difference frequency signal computing module, be used for according to said symbol sebolic addressing and through said reception signal is carried out channel estimating and slot synchronization obtain synchronously after reception signal and channel estimation value, calculate the difference frequency signal under this trellis state;

Cross product value computing module is used for the difference frequency signal of difference frequency signal under said this trellis state and the initial moment corresponding states of said local survivor path is carried out cross product, obtains the cross product value under the corresponding trellis state of said N symbol;

The frequency offset estimating computing module is used for according to said cross product value, calculates the frequency offset estimating value under the corresponding trellis state of said N symbol.

8. the device of signal frequency deviation correction according to claim 7 is characterized in that, said second computing module also comprises:

The smothing filtering module, the cross product value that is used on the said survivor path under cross product value and this trellis state under the trellis state that said N the symbol that obtains is corresponding under the corresponding trellis state of N-1 symbol is carried out smothing filtering.

9. a signal frequency deviation correction system is characterized in that, comprising: like the device of each described signal frequency deviation correction of claim 6 to 8; And apparatus for transmitting signal; Wherein, said apparatus for transmitting signal is used to send the device that the signal with certain frequency arrives said signal frequency deviation correction.

Images