CN101409586A - Method and apparatus for estimating frequency offset - Google Patents

Abstract

The invention discloses a method for estimating frequency deviation and a device thereof, wherein, the method comprises: the received data is demodulated to obtain the data sequence, the demodulated data sequence is distinguished by using the preset decision threshold, the component sequence S before demodulation is recovered for the decided data, interrelated operation is carried out on the recovered data S and the received data, and the interrelated result is obtained; chip merge is carried out on the interrelated result to obtain phase deviation, and frequency deviation is worked out according to the phase deviation. When small frequency excursion exists in a connection mode, the method and the device of the invention can more exactly estimate the frequency deviation between a sending terminal and a receiving terminal; as channel estimate result is not needed to reconstruct the signals, thus greatly reducing complexity and calculation time delay of the system. The estimated frequency deviation can meet the requirements of a receiver within the range of frequency deviation of -200 to +200Hz.

Description

A kind of method and apparatus of Frequency offset estimation

Technical field

The present invention relates to wireless communication technology field, be meant a kind of method and apparatus of Frequency offset estimation especially.

Background technology

In wireless communication system,, make to exist bigger frequency departure between carrier frequency and the local crystal oscillator because the frequency departure and the terminal of equipment room move the Doppler frequency shift that causes.Have only fast and effeciently to estimate this frequency departure, just can carry out frequency compensation, reach the frequency departure that receiver can be accepted, thereby carry out normal demodulation coding.

Referring to Fig. 1, be example with the burst structure of the business time-slot of the TD-SCDMA system of 3GPP 1.28Mcps TDD regulation, burst middle part intermediate code (midamble sign indicating number) can be used for channel estimating, and the data block on both sides is used for service data transmission.In the TD-SCDMA system, each time slot has two data blocks, and each data block has 352/Q symbol (Q is a spreading factor, Q=1,2,4,8,16), and intermediate code has 144 chips.It is T that chip continues the cycle _c=1/1.28M=781ns.

Invariant features when existing Frequency offset estimation is utilized Burst Channel, the intermediate code of reception is carried out channel estimating, generates reconstruction signal with channel estimating and local intermediate code convolution, and reconstruction signal handled obtains frequency departure and estimate.This scheme has channel estimation process, and complexity is bigger, and real-time is poor, and owing to be subjected to the restriction of local intermediate code data length, in the scope of less frequency departure, the precision of estimating frequency offset just can not meet the demands;

Another scheme adopts intermediate code to carry out channel estimating, and the demodulate/decode signal is carried out generating reconstruction signal with the channel impulse response convolution after again the coded/modulated, carries out the estimation of frequency departure with this reconstruction signal.This scheme is because many coded/modulated processes, that not only wants deal with data separates the mediation work decoding, also need the demodulate/decode signal is carried out once more generating reconstruction signal with the channel impulse response convolution again after the coded/modulated, a large amount of complexities and calculation delay have been brought, brought extra burden to receiving system, real-time is relatively poor.

Summary of the invention

In view of this, the invention reside in the method and apparatus that a kind of Frequency offset estimation is provided, above-mentioned scheme estimated accuracy is poor to solve, amount of calculation big, the problem of real-time difference.

For addressing the above problem, the invention provides a kind of method of Frequency offset estimation, comprising:

To the data r=(r that receives ₀, r ₁, L, r _k, L r _N-1) carry out demodulation obtain data sequence r '=(r ' ₀, r ' ₁..., r ' _k... r ' _N-1);

The described data that receive are meant the data before system's demodulation;

Wherein ${r_K}^{\′}=a_k{e}^{j({\mathrm{\θ}}_k+2\mathrm{\πk}{F}_d{T}_C)},$ 0≤k≤N-1; N is for receiving the number of data symbol, and k is the index of each receiving symbol, 0≤k≤N-1, a _kBe the amplitude of k receiving symbol, θ _kBe the phase modulation of k receiving symbol, F _dExpression actual frequency deviation, T _cFor chip continues the cycle;

Differentiate data sequence r ' after the demodulation with predetermined decision threshold, the data after the judgement are recovered vector sequence before the demodulation

$\hat{s}=({\hat{s}}_0,{\hat{s}}_1,L,{\hat{s}}_k,L{\hat{s}}_{N-1});$

Wherein, ${\hat{s}}_k=b_k{e}^{j{\mathrm{\θ}}_k},$ 0≤k≤N-1; B wherein _kBe the amplitude of declaring symbol K position firmly of receiving symbol, θ _kIt is the phase modulation of k receiving symbol;

To the described data that recover Carry out related operation with the described data that receive, obtain correlated results $c=\mathrm{conj}\left(\hat{s}\right)\·*r^{\′};$

Wherein: conj () represents conjugate operation, and * represents that data of corresponding positions multiplies each other respectively in two data sequences;

Described correlated results is carried out chip merge the acquisition phase deviation, calculate frequency departure by phase deviation.

Preferably, the described data sequence r ' that differentiates after the demodulation with predetermined decision threshold is:

Data after the described demodulation are done hard decision, adopt 0 to be decision threshold, the data greater than 0 are judged to 1, and the data less than 0 are judged to 0.

Preferably, described correlated results is carried out chip merge the acquisition phase deviation, the process that calculates frequency departure by phase deviation is:

Each phase angle θ with described correlated results _{D, i}Deduct first θ respectively _{D, 0}Difference add up, obtain the result $\mathrm{\Δ\θ}=\underset{i=1}{\overset{N-1}{\mathrm{\Σ}}}({\mathrm{\θ}}_{d,i}-{\mathrm{\θ}}_{d,0}),$ I=1,2, L, N-1, θ _{D, i}Be related operation c as a result _kPhase place, 1≤i≤N-1;

Phase place amalgamation result Δ θ after utilization adds up obtains frequency departure ${\hat{F}}_d=\frac{\mathrm{\Δ\θ}}{\mathrm{\πTN}(N-1)};$ T is the inverse of spreading rate.

Preferably, described correlated results is carried out chip merge the acquisition phase deviation, the process that calculates frequency departure by phase deviation is:

In described related operation result's sequence, serve as to slide along sequence at interval with step-length l, obtain being spaced apart two correlated results phase differences of l, with each phase difference summation, obtain amalgamation result $\mathrm{\Δ\θ}=\underset{i=1}{\overset{N-l-1}{\mathrm{\Σ}}}({\mathrm{\θ}}_{d,i+l}-{\mathrm{\θ}}_{d,i});$

Wherein, θ _{D, i}Be related operation c as a result _kPhase place 1≤i≤N-1, step-length l be less than (N-1) arbitrarily on the occasion of;

Phase place amalgamation result Δ θ after utilization adds up obtains frequency departure ${\hat{F}}_d=\frac{\mathrm{\Δ\θ}}{2\mathrm{\πTl}(N-l)};$ T is the inverse of spreading rate.

Preferably, described correlated results is carried out chip merge the acquisition phase deviation, the process that calculates frequency departure by phase deviation is:

Between sequence after described correlated results sequence is carried out conjugate operation and the described correlated results sequence, be that the data that l obtains multiply each other, obtain the amalgamation result of correlated results product with the step interval $\mathrm{\ΔC}=\underset{i=0}{\overset{N-l-1}{\mathrm{\Σ}}}(\mathrm{conj}\left(c_i\right).*c_{i+l});$

Wherein, conj () represents conjugate operation, and * represents that data of corresponding positions multiplies each other respectively in two data sequences; C is the result of related operation;

Obtain the phase delta θ of amalgamation result Δ C, obtain frequency departure ${\hat{F}}_d=\frac{\mathrm{\Δ\θ}}{2\mathrm{\πl}T};$ T is the inverse of spreading rate.

Preferably, described step interval l is 2 or 4.

The present invention also provides a kind of device of Frequency offset estimation, comprising:

Demodulating unit is used for the data r=(r to receiving ₀, r ₁, L, r _k, L r _N-1) carry out demodulation obtain data sequence r '=(r ' ₀, r ' ₁..., r ' _k... r ' _N-1);

The described data that receive are meant the data before system's demodulation;

Wherein ${r_K}^{\′}=a_k{e}^{j({\mathrm{\θ}}_k+2\mathrm{\πk}{F}_d{T}_C),}$ 0≤k≤N-1; N is for receiving the number of data symbol, and k is the index of each receiving symbol, 0≤k≤N-1, a _kBe the amplitude of k receiving symbol, θ _kBe the phase modulation of k receiving symbol, F _dExpression actual frequency deviation, T _cFor chip continues the cycle;

Decision unit is used for the data sequence r ' after the predetermined decision threshold differentiation demodulation;

The signal reconstruction unit is used for the data after the judgement are recovered vector sequence before the demodulation $\hat{s}=({\hat{s}}_0,{\hat{s}}_1,L,{\hat{s}}_k,L{\hat{s}}_{N-1});$

Wherein, ${\hat{s}}_k=b_k{e}^{j{\mathrm{\θ}}_k},$ 0≤k≤N-1; B wherein _kBe the amplitude of declaring symbol K position firmly of receiving symbol, θ _kIt is the phase modulation of k receiving symbol;

The related operation unit is used for the data that recover described

Carry out related operation with the described data that receive, obtain correlated results $c=\mathrm{conj}\left(\hat{s}\right)\·*r^{\′};$

Wherein: conj () represents conjugate operation, and * represents that data of corresponding positions multiplies each other respectively in two data sequences;

The phase estimation unit is used for that described correlated results is carried out chip and merges the acquisition phase deviation;

The Frequency Estimation unit calculates frequency departure by phase deviation.

Preferably, the judging process of described decision unit is:

Data after the described demodulation are done hard decision, adopt 0 to be decision threshold, the data greater than 0 are judged to 1, and the data less than 0 are judged to 0.

Preferably,

Described phase estimation unit to the process that described correlated results carries out chip merging acquisition phase deviation is:

Each phase angle θ with described correlated results _{D, i}Deduct first θ respectively _{D, 0}Difference add up, obtain the result $\mathrm{\Δ\θ}=\underset{i=1}{\overset{N-1}{\mathrm{\Σ}}}({\mathrm{\θ}}_{d,i}-{\mathrm{\θ}}_{d,0}),$ I=1,2, L, N-1, θ _{D, i}Be related operation c as a result _kPhase place, 1≤i≤N-1;

Described Frequency Estimation unit by the process that phase deviation calculates frequency departure is: utilize the phase place amalgamation result Δ θ after adding up to obtain frequency departure ${\hat{F}}_d=\frac{\mathrm{\Δ\θ}}{\mathrm{\πTN}(N-1)};$ T is the inverse of spreading rate.

Or,

Described phase estimation unit to the process that described correlated results carries out chip merging acquisition phase deviation is:

In described related operation result's sequence, serve as to slide along sequence at interval with step-length l, obtaining step interval is two correlated results phase differences of l, with each phase difference summation, obtains amalgamation result $\mathrm{\Δ\θ}=\underset{i=1}{\overset{N-l-1}{\mathrm{\Σ}}}({\mathrm{\θ}}_{d,i+l}-{\mathrm{\θ}}_{d,i});$

Wherein, θ _{D, i}Be related operation c as a result _kPhase place 1≤i≤N-1, step-length l be less than (N-1) arbitrarily on the occasion of;

Described Frequency Estimation unit by the process that phase deviation calculates frequency departure is: utilize the phase place amalgamation result Δ θ after adding up to obtain frequency departure ${\hat{F}}_d=\frac{\mathrm{\Δ\θ}}{2\mathrm{\πTl}(N-l)};$ T is the inverse of spreading rate;

Or,

Described phase estimation unit to the process that described correlated results carries out chip merging acquisition phase deviation is:

Between sequence after described correlated results sequence is carried out conjugate operation and the described correlated results sequence, be that the data that l obtains multiply each other, obtain the amalgamation result of correlated results product with the step interval $\mathrm{\ΔC}=\underset{i=0}{\overset{N-l-1}{\mathrm{\Σ}}}(\mathrm{conj}\left(c_i\right).*c_{i+l});$ Obtain the phase delta θ of amalgamation result Δ C,

Wherein, conj () represents conjugate operation, and * represents that data of corresponding positions multiplies each other respectively in two data sequences; C is the result of related operation;

Described Frequency Estimation unit by the process that phase deviation calculates frequency departure is: utilize phase place amalgamation result Δ θ to obtain frequency departure ${\hat{F}}_d=\frac{\mathrm{\Δ\θ}}{2\mathrm{\πl}T};$ T is the inverse of spreading rate.

Preferably, the step interval l of described phase estimation unit and described Frequency Estimation unit employing is 2 or 4.

Method and apparatus of the present invention can be when there be the small frequency skew in connection mode, estimate the frequency shift (FS) between transmitting terminal and the receiving terminal more accurately, owing to do not need to utilize the channel estimation results reconstruction signal, greatly reduced the complexity and the calculation delay of system.In the scope of the frequency departure of ± 200Hz, the frequency departure that estimates all can satisfy the requirement of receiver.

Description of drawings

Fig. 1 is a TD-SCDMA business time-slot burst structure chart;

Fig. 2 adopts data reconstruction to obtain the schematic diagram that frequency departure is estimated in the embodiments of the invention;

Fig. 3 is first kind of schematic diagram of estimating phase deviation in the embodiment of the invention;

Fig. 4 is second kind of schematic diagram of estimating phase deviation in the embodiment of the invention;

Fig. 5 is the third schematic diagram of estimating phase deviation in the embodiment of the invention;

Fig. 6 is the structure chart of apparatus of the present invention embodiment.

Embodiment

For clearly demonstrating the scheme among the present invention, provide preferred embodiment below and be described with reference to the accompanying drawings.

Referring to Fig. 2, Fig. 2 is the schematic diagram of the inventive method embodiment, comprises that data demodulates, hard decision device, signal reconstruction, related operation, data merge, each process of Frequency Estimation.At TD-SCDMA system or similar system terminal, adopt the process of data reconstruction estimating frequency offset may further comprise the steps in the embodiment of the invention:

Step 1: to the data r=(r that receives ₀, r ₁, L, r _k, L r _N-1) carry out demodulation; Data after the demodulation be r '=(r ' ₀, r ' ₁..., r ' _k... r ' _N-1);

The described data that receive are meant the data before system's demodulation;

Wherein ${r_K}^{\′}=a_k{e}^{j({\mathrm{\θ}}_k+2\mathrm{\πk}{F}_d{T}_C)},$ 0≤k≤N-1; Wherein N is for receiving the number of data symbol, and k is the index of each receiving symbol, 0≤k≤N-1, a _kBe the amplitude of k receiving symbol, θ _kBe the phase modulation of k receiving symbol, F _dExpression actual frequency deviation, T _cFor chip continues the cycle;

Step 2: the data after the described demodulation are done hard decision, adopt 0 to be decision threshold, the data greater than 0 are judged to 1, and the data less than 0 are judged to 0;

Step 3:, recover not to be subjected to fundamental component before the demodulation that frequency departure influences according to modulation constellation to the data after the judgement

$\hat{s}=({\hat{s}}_0,{\hat{s}}_1,L,{\hat{s}}_k,L{\hat{s}}_{N-1});$ Wherein, ${\hat{s}}_k=b_k{e}^{j{\mathrm{\θ}}_k},$ 0≤k≤N-1; B wherein _kBe the amplitude of declaring symbol K position firmly of receiving symbol, θ _kIt is the phase modulation of k receiving symbol;

Step 4: to the data that recover

Carry out relevant with the reception data;

Correlated results is $c=\mathrm{conj}\left(\hat{s}\right)\·*r^{\′}$

Promptly $c=(\mathrm{conj}\left({\hat{s}}_0\right)\&CenterDot;{\mathrm{<figure-callout\; id="0"\; label="r"\; filenames="A20081018241500193.png,A20081018241500201.png"\; state="\{\{state\}\}">r</figure-callout>}}_0,\mathrm{conj}\left({\hat{s}}_1\right)\&CenterDot;{\mathrm{<figure-callout\; id="1"\; label="r"\; filenames="A20081018241500193.png,A20081018241500201.png"\; state="\{\{state\}\}">r</figure-callout>}}_1,L,\mathrm{conj}\left({\hat{s}}_{N-1}\right)\&CenterDot;r_{N-1})$

Wherein: conj () represents conjugate operation, and * represents that data of corresponding positions multiplies each other respectively in two data sequences; C is the result of related operation, c=(c ₀, c ₁, L, c _N-1).

Step 5: described correlated results is carried out chip merge, the sequence after being combined obtains phase deviation, and wherein phase deviation can obtain by function arg tg () computing;

Step 6: utilize the result of described phase deviation, obtain frequency departure;

For step 5 and step 6, can adopt multiple mode to obtain phase deviation and frequency departure.Provide phase deviation and the three kinds of schemes of frequency departure of obtaining below:

Scheme 1:

Utilize the result of related operation, obtain phase deviation:

Referring to Fig. 4, the phase place amalgamation result is: $\mathrm{\&Delta;\&theta;}=\underset{i=1}{\overset{N-1}{\mathrm{\&Sigma;}}}({\mathrm{\&theta;}}_{d,i}-{\mathrm{\&theta;}}_{d,0});$

Each phase angle θ with correlated results _{D, i}Deduct θ respectively _{D, 0}Difference accumulation result Δ θ, i=1,2, L, N-1, θ _{D, i}Be related operation c as a result _kPhase place, 1≤i≤N-1;

Phase place amalgamation result after utilization adds up obtains frequency departure ${\hat{F}}_d=\frac{\mathrm{\&Delta;\&theta;}}{\mathrm{\&pi;TN}(N-1)};$ T is the inverse of spreading rate.

Scheme 2:

Utilize the result of related operation, obtain phase deviation, satisfy:

Utilize the result of related operation, obtain phase deviation: $\mathrm{\&Delta;\&theta;}=\underset{i=1}{\overset{N-l-1}{\mathrm{\&Sigma;}}}({\mathrm{\&theta;}}_{d,i+l}-{\mathrm{\&theta;}}_{d,i})$

Referring to Fig. 5, in related operation result's sequence, obtaining step interval is two correlated results phase differences of l, slides along sequence, and each phase difference summation with obtaining obtains amalgamation result Δ θ, and the computing of plural phase place is got in " ∠ " expression among the figure.

Wherein, Δ θ is the amalgamation result with respect to phase deviation, θ _{D, i}Be related operation c as a result _kPhase place 1≤i≤N-1, step-length l get than N-1 little arbitrarily on the occasion of, the suggestion l preferably get 2 or 4;

According to phase deviation, obtain frequency departure, satisfy: ${\hat{F}}_d=\frac{\mathrm{\&Delta;\&theta;}}{2\mathrm{\&pi;Tl}(N-l)};$ T is the inverse of spreading rate.

Scheme 3:

Utilize the result of related operation, obtain phase deviation:

Between the sequence and correlated series behind the correlated results conjugate operation, the data of getting step interval and be l multiply each other, and slide along two sequences, obtain the amalgamation result of correlated results product $\mathrm{\&Delta;C}=\underset{i=0}{\overset{N-l-1}{\mathrm{\&Sigma;}}}(\mathrm{conj}\left(c_i\right).*c_{i+l});$

Wherein, conj () represents conjugate operation, and * represents that data of corresponding positions multiplies each other respectively in two data sequences; C is the result of related operation, c=(c ₀, c ₁, L, c _N-1).

Wherein, l gets the arbitrary value littler than N-1, and suggestion l preferably gets 2 or 4.

According to phase deviation, obtain frequency departure, satisfy:

${\hat{F}}_d=\frac{\mathrm{\&Delta;\&theta;}}{2\mathrm{\&pi;l}T}$

Wherein, Δ θ is the phase place of amalgamation result Δ C, and T is the inverse of spreading rate.

Method of the present invention can estimate the frequency shift (FS) between transmitting terminal and the receiving terminal more accurately when there is the small frequency skew in connection mode, owing to do not need to utilize the channel estimation results reconstruction signal, greatly reduced the complexity and the calculation delay of system.In the scope of the frequency departure of ± 200Hz, the frequency departure that estimates all can satisfy the requirement of receiver.

Describe method of the present invention above in detail, various ways arranged, provide preferred device embodiment below,, comprising referring to Fig. 6 for the device of realizing the inventive method:

Demodulating unit is used for the data r=(r to receiving ₀, r ₁, L, r _k, L r _N-1) carry out demodulation obtain data sequence r '=(r ' ₀, r ' ₁..., r ' _k... r ' _N-1);

The described data that receive are meant the data before system's demodulation;

Wherein ${r_K}^{\&prime;}=a_k{e}^{j({\mathrm{\&theta;}}_k+2\mathrm{\&pi;k}{F}_d{T}_C)},$ 0≤k≤N-1; N is for receiving the number of data symbol, and k is the index of each receiving symbol, 0≤k≤N-1, a _kBe the amplitude of k receiving symbol, θ _kBe the phase modulation of k receiving symbol, F _dExpression actual frequency deviation, T _cFor chip continues the cycle;

Decision unit is used for the data sequence r ' after the predetermined decision threshold differentiation demodulation;

The signal reconstruction unit is used for the data after the judgement are recovered vector sequence before the demodulation

$\hat{s}=({\hat{s}}_0,{\hat{s}}_1,L,{\hat{s}}_k,L{\hat{s}}_{N-1});$

Wherein, ${\hat{s}}_k=b_k{e}^{j{\mathrm{\&theta;}}_k},$ 0≤k≤N-1; B wherein _kBe the amplitude of declaring symbol K position firmly of receiving symbol, θ _kIt is the phase modulation of k receiving symbol;

The related operation unit is used for the data that recover described

Carry out related operation with the described data that receive, obtain correlated results $c=\mathrm{conj}\left(\hat{s}\right)\&CenterDot;*r^{\&prime;};$

Wherein: conj () represents conjugate operation, and * represents that data of corresponding positions multiplies each other respectively in two data sequences;

The phase estimation unit is used for that described correlated results is carried out chip and merges the acquisition phase deviation;

The Frequency Estimation unit calculates frequency departure by phase deviation.

Phase estimation unit and Frequency Estimation unit have three kinds of implementations:

Mode 1

The phase estimation unit to the process that described correlated results carries out chip merging acquisition phase deviation is:

Each phase angle θ with described correlated results _{D, i}Deduct first θ respectively _{D, 0}Difference add up, obtain the result $\mathrm{\&Delta;\&theta;}=\underset{i=1}{\overset{N-1}{\mathrm{\&Sigma;}}}({\mathrm{\&theta;}}_{d,i}-{\mathrm{\&theta;}}_{d,0}),$ I=1,2, L, N-1, θ _{D, i}Be related operation c as a result _kPhase place, 1≤i≤N-1;

The Frequency Estimation unit by the process that phase deviation calculates frequency departure is: utilize the phase place amalgamation result Δ θ after adding up to obtain frequency departure ${\hat{F}}_d=\frac{\mathrm{\&Delta;\&theta;}}{\mathrm{\&pi;TN}(N-1)};$ T is the inverse of spreading rate.

Mode 2

The phase estimation unit to the process that described correlated results carries out chip merging acquisition phase deviation is:

In described related operation result's sequence, serve as to slide at interval with step-length l along sequence, obtain being spaced apart the amalgamation result of two correlated results phase differences of l $\mathrm{\&Delta;\&theta;}=\underset{i=1}{\overset{N-l-1}{\mathrm{\&Sigma;}}}({\mathrm{\&theta;}}_{d,i+l}-{\mathrm{\&theta;}}_{d,i});$

Wherein, θ _{D, i}Be related operation c as a result _kPhase place 1≤i≤N-1, step-length l be less than (N-1) arbitrarily on the occasion of;

The Frequency Estimation unit by the process that phase deviation calculates frequency departure is: utilize the phase place amalgamation result Δ θ after adding up to obtain frequency departure ${\hat{F}}_d=\frac{\mathrm{\&Delta;\&theta;}}{2\mathrm{\&pi;Tl}(N-l)};$ T is the inverse of spreading rate.

Mode 3

The phase estimation unit to the process that described correlated results carries out chip merging acquisition phase deviation is:

Between sequence after described correlated results sequence is carried out conjugate operation and the described correlated results sequence, be that the data that l obtains multiply each other with the step interval, step-length l be less than (N-1) arbitrarily on the occasion of; Obtain the amalgamation result of correlated results product $\mathrm{\&Delta;C}=\underset{i=0}{\overset{N-l-1}{\mathrm{\&Sigma;}}}(\mathrm{conj}\left(c_i\right).*c_{i+l});$ Obtain the phase delta θ of amalgamation result Δ C,

Wherein, conj () represents conjugate operation, and * represents that data of corresponding positions multiplies each other respectively in two data sequences; C is the result of related operation;

The Frequency Estimation unit by the process that phase deviation calculates frequency departure is: utilize phase place amalgamation result Δ θ to obtain frequency departure ${\hat{F}}_d=\frac{\mathrm{\&Delta;\&theta;}}{2\mathrm{\&pi;l}T};$ T is the inverse of spreading rate.

In mode 2 and mode 3, step-length l can be preferably 2 or 4 for the arbitrary value littler than N-1.

Device of the present invention can estimate the frequency shift (FS) between transmitting terminal and the receiving terminal more accurately when there is the small frequency skew in connection mode, owing to do not need to utilize the channel estimation results reconstruction signal, greatly reduced the complexity and the calculation delay of system.In the scope of the frequency departure of ± 200Hz, the frequency departure that estimates all can satisfy the requirement of receiver.

For the method and apparatus of being set forth among each embodiment of the present invention, within the spirit and principles in the present invention all, any modification of being done, be equal to replacement, improvement etc., all should be included within protection scope of the present invention.

Claims (10)

1, a kind of method of Frequency offset estimation is characterized in that, comprising:

To the data r=(r that receives ₀, r ₁, L, r _k, L _N-1) carry out demodulation obtain data sequence r '=(r ' ₀, r ' ₁..., r ' _k... r ' _N-1);

Wherein ${r_K}^{\&prime;}=a_k{e}^{j({\mathrm{\&theta;}}_k+2\mathrm{\&pi;k}{F}_d{T}_C)},$ 0≤k≤N-1; N is for receiving the number of data symbol, and k is the index of each receiving symbol, 0≤k≤N-1, a _kBe the amplitude of k receiving symbol, θ k is the phase modulation of k receiving symbol, F _dExpression actual frequency deviation, T _cFor chip continues the cycle;

Differentiate data sequence r ' after the demodulation with predetermined decision threshold, the data after the judgement are recovered vector sequence before the demodulation

$\hat{s}=({\hat{s}}_0,{\hat{s}}_1,L,{\hat{s}}_k,L{\hat{s}}_{N-1});$

Wherein, ${\hat{s}}_k=b_k{e}^{j{\mathrm{\&theta;}}_k},$ 0≤k≤N-1; B wherein _kBe the amplitude of declaring symbol K position firmly of receiving symbol, θ _kIt is the phase modulation of k receiving symbol;

To the described data that recover

Carry out related operation with the described data that receive, obtain correlated results $c=\mathrm{conj}\left(\hat{s}\right)\&CenterDot;*r^{\&prime;};$

Wherein: conj () represents conjugate operation, and * represents that data of corresponding positions multiplies each other respectively in two data sequences;

Described correlated results is carried out chip merge the acquisition phase deviation, calculate frequency departure by phase deviation.

2, method according to claim 1 is characterized in that, the described data sequence r ' that differentiates after the demodulation with predetermined decision threshold is:

Data after the described demodulation are done hard decision, adopt 0 to be decision threshold, the data greater than 0 are judged to 1, and the data less than 0 are judged to 0.

3, method according to claim 1 is characterized in that, described correlated results is carried out chip merge the acquisition phase deviation, and the process that calculates frequency departure by phase deviation is:

Each phase angle θ with described correlated results _{D, i}Deduct first θ respectively _{D, 0}Difference add up, obtain the result $\mathrm{\&Delta;\&theta;}=\underset{i=1}{\overset{N-1}{\mathrm{\&Sigma;}}}({\mathrm{\&theta;}}_{d,i}-{\mathrm{\&theta;}}_{d,0}),$ I=1,2, L, N-1, θ _{D, i}Be related operation c as a result _kPhase place, 1≤i≤N-1;

Phase place amalgamation result Δ θ after utilization adds up obtains frequency departure ${\hat{F}}_d=\frac{\mathrm{\&Delta;\&theta;}}{\mathrm{\&pi;TN}(N-1)};$ T is the inverse of spreading rate.

4, method according to claim 1 is characterized in that, described correlated results is carried out chip merge the acquisition phase deviation, and the process that calculates frequency departure by phase deviation is:

In described related operation result's sequence, serve as to slide along sequence at interval with step-length 1, obtain being spaced apart two correlated results phase differences of 1, with each phase difference summation, obtain amalgamation result $\mathrm{\&Delta;\&theta;}=\underset{i=1}{\overset{N-l-1}{\mathrm{\&Sigma;}}}({\mathrm{\&theta;}}_{d,i+l}-{\mathrm{\&theta;}}_{d,i});$

Wherein, θ _{D, i}Be related operation c as a result _kPhase place 1≤i≤N-1, step-length 1 be less than (N-1) arbitrarily on the occasion of;

Phase place amalgamation result Δ θ after utilization adds up obtains frequency departure ${\hat{F}}_d=\frac{\mathrm{\&Delta;\&theta;}}{2\mathrm{\&pi;Tl}(N-l)};$ T is the inverse of spreading rate.

5, method according to claim 1 is characterized in that, described correlated results is carried out chip merge the acquisition phase deviation, and the process that calculates frequency departure by phase deviation is:

Between sequence after described correlated results sequence is carried out conjugate operation and the described correlated results sequence, be that 1 data that obtain multiply each other, obtain the amalgamation result of correlated results product with step interval $\mathrm{\&Delta;C}=\underset{i=0}{\overset{N-l-1}{\mathrm{\&Sigma;}}}(\mathrm{conj}\left(c_i\right).*c_{i+l});$

Wherein, conj () represents conjugate operation, and * represents that data of corresponding positions multiplies each other respectively in two data sequences; C is the result of related operation;

Obtain the phase delta θ of amalgamation result Δ C, obtain frequency departure ${\hat{F}}_d=\frac{\mathrm{\&Delta;\&theta;}}{2\mathrm{\&pi;lT}};$ T is the inverse of spreading rate.

6, method according to claim 5 is characterized in that, described step-length 1 is 2 or 4.

7, a kind of device of Frequency offset estimation is characterized in that, comprising:

Demodulating unit is used for the data r=(r to receiving ₀, r ₁, L, r _k, L r _N-1) carry out demodulation obtain data sequence r '=(r ' ₀, r ' ₁..., r ' _k... r ' _N-1);

Wherein ${r_K}^{\&prime;}=a_k{e}^{j({\mathrm{\&theta;}}_k+2\mathrm{\&pi;k}{F}_d{T}_C)},$ 0≤k≤N-1; N is for receiving the number of data symbol, and k is the index of each receiving symbol, 0≤k≤N-1, a _kBe the amplitude of k receiving symbol, θ _kBe the phase modulation of k receiving symbol, F _dExpression actual frequency deviation, T _cFor chip continues the cycle;

Decision unit is used for the data sequence r ' after the predetermined decision threshold differentiation demodulation; The signal reconstruction unit is used for the data after the judgement are recovered vector sequence before the demodulation

$\hat{s}=({\hat{s}}_0,{\hat{s}}_1,L,{\hat{s}}_k,L{\hat{s}}_{N-1});$

Wherein, ${\hat{s}}_k=b_k{e}^{j{\mathrm{\&theta;}}_k},$ 0≤k≤N-1; B wherein _kBe the amplitude of declaring symbol K position firmly of receiving symbol, θ _kIt is the phase modulation of k receiving symbol;

The related operation unit is used for the data that recover described

Carry out related operation with the described data that receive, obtain correlated results $c=\mathrm{conj}\left(\hat{s}\right)\&CenterDot;*r^{\&prime;};$

Wherein: conj () represents conjugate operation, and * represents that data of corresponding positions multiplies each other respectively in two data sequences;

The phase estimation unit is used for that described correlated results is carried out chip and merges the acquisition phase deviation;

The Frequency Estimation unit calculates frequency departure by phase deviation.

8, device according to claim 7 is characterized in that, the judging process of described decision unit is:

Data after the described demodulation are done hard decision, adopt 0 to be decision threshold, the data greater than 0 are judged to 1, and the data less than 0 are judged to 0.

9, device according to claim 7 is characterized in that,

Described phase estimation unit to the process that described correlated results carries out chip merging acquisition phase deviation is:

Each phase angle θ with described correlated results _{D, i}Deduct first θ respectively _{D, 0}Difference add up, obtain the result $\mathrm{\&Delta;\&theta;}=\underset{i=1}{\overset{N-1}{\mathrm{\&Sigma;}}}({\mathrm{\&theta;}}_{d,i}-{\mathrm{\&theta;}}_{d,0}),$ I=1,2, L, N-1, θ _{D, i}Be related operation c as a result _kPhase place, 1≤i≤N-1;

Described Frequency Estimation unit by the process that phase deviation calculates frequency departure is: utilize the phase place amalgamation result Δ θ after adding up to obtain frequency departure ${\hat{F}}_d=\frac{\mathrm{\&Delta;\&theta;}}{\mathrm{\&pi;TN}(N-1)};$ T is the inverse of spreading rate;

Or,

Described phase estimation unit to the process that described correlated results carries out chip merging acquisition phase deviation is:

In described related operation result's sequence, serve as to slide at interval with step-length 1 along sequence, obtain step interval and be two correlated results phase differences of 1, with each phase difference summation, obtain amalgamation result $\mathrm{\&Delta;\&theta;}=\underset{i=1}{\overset{N-l-1}{\mathrm{\&Sigma;}}}({\mathrm{\&theta;}}_{d,i+l}-{\mathrm{\&theta;}}_{d,i});$

Wherein, θ _{D, i}Be related operation c as a result _kPhase place 1≤i≤N-1, step-length 1 be less than (N-1) arbitrarily on the occasion of;

Described Frequency Estimation unit by the process that phase deviation calculates frequency departure is: utilize the phase place amalgamation result Δ θ after adding up to obtain frequency departure ${\hat{F}}_d=\frac{\mathrm{\&Delta;\&theta;}}{2\mathrm{\&pi;Tl}(N-l)};$ T is the inverse of spreading rate;

Or,

Described phase estimation unit to the process that described correlated results carries out chip merging acquisition phase deviation is:

Between sequence after described correlated results sequence is carried out conjugate operation and the described correlated results sequence, be that 1 data that obtain multiply each other, obtain the amalgamation result of correlated results product with step interval $\mathrm{\&Delta;C}=\underset{i=0}{\overset{N-l-1}{\mathrm{\&Sigma;}}}(\mathrm{conj}\left(c_i\right).*c_{i+l});$ Obtain the phase delta θ of amalgamation result Δ C,

Wherein, conj () represents conjugate operation, and * represents that data of corresponding positions multiplies each other respectively in two data sequences; C is the result of related operation;

Described Frequency Estimation unit by the process that phase deviation calculates frequency departure is: utilize phase place amalgamation result Δ θ to obtain frequency departure ${\hat{F}}_d=\frac{\mathrm{\&Delta;\&theta;}}{2\mathrm{\&pi;lT}};$ T is the inverse of spreading rate.

10, device according to claim 9 is characterized in that, the step interval 1 that described phase estimation unit and described Frequency Estimation unit adopt is 2 or 4.

Images