CN101909024B - Method and device for estimating maximum Doppler frequency offset - Google Patents

Abstract

The invention discloses a method and a device for estimating the maximum Doppler frequency offset. The method comprises the following steps of: calculating channel estimation according to received signals; calculating an amplitude spectrum of the channel estimation, and filtering the amplitude spectrum; performing folding average on the amplitude spectrum after filtering; estimating the level of noise; calculating a detection threshold; and calculating and outputting the maximum Doppler frequency offset. The method and the device solve the problems that the estimation value of an algorithm for acquiring information of the Doppler frequency offset has poor stability or low estimation accuracy, further improve the stability and can acquire higher estimation accuracy.

Description

Method and device for estimating maximum Doppler frequency offset

Technical Field

The present invention relates to the field of communications, and in particular, to a method and an apparatus for estimating maximum doppler frequency offset.

Background

Currently, Orthogonal Frequency Division Multiplexing (OFDM) has been widely applied to various Wireless communication systems, such as Digital Video Broadcasting (DVB), Wireless Local Area Network (WLAN), China Mobile Multimedia Broadcasting (CMMB), and so on. These systems are required to operate efficiently in various channel environments, for example, to accommodate various speed ranges of mobile stations, various carrier frequencies, various time delays, and Signal to Noise ratios (SNRs). Therefore, it is very important to determine the channel quality and the variation speed in advance, and it is necessary to select the system parameters adaptively as the channel environment continuously varies.

In wireless mobile communication, due to high speed movement of a transmitter or a receiver, doppler frequency spreading is caused, which affects correct reception of signals and in many adaptive link transmission systems, the system needs to obtain parameter information in a channel. For example, the two-dimensional wiener filtering channel estimation method of the OFDM system requires knowing the doppler frequency offset magnitude of the channel. The estimation result can be improved by filtering the pilot frequency through a low-pass filter with the same bandwidth as the maximum doppler frequency offset of doppler, and the required transmission power is relatively low.

In the related art, various algorithms proposed for acquiring multi-frequency offset information include: the method comprises an autocorrelation method, a level crossing rate method, a diversity switching rate method and the like, wherein the autocorrelation method is influenced by channel noise, the signal-to-noise ratio of a channel needs to be accurately estimated in advance, and the autocorrelation method has poor performance under the condition of low signal-to-noise ratio. In fact, the autocorrelation function of noise is a delta function, and when the autocorrelation method estimates the signal energy, the noise size needs to be known to remove the estimation deviation caused by the noise, even under the condition that the noise signal-to-noise ratio is accurately known, the stability of the estimated value is poor at the time of low signal-to-noise ratio; the level crossing rate method and the diversity switching rate method can only roughly estimate the Doppler frequency offset value, and the estimation precision is low.

Aiming at the problem that the algorithm estimation value for acquiring multi-frequency offset information in the related technology is poor in stability or low in estimation precision, an effective solution is not provided at present.

Disclosure of Invention

The present invention is proposed to solve the problem of poor stability or low estimation accuracy of the algorithm estimation value for obtaining multi-frequency offset information in the related art, and therefore, the main object of the present invention is to provide an improved estimation scheme for maximum doppler frequency offset to solve at least one of the above problems.

To achieve the above object, according to an aspect of the present invention, a method for estimating maximum doppler frequency offset is provided.

The method for estimating the maximum Doppler frequency offset comprises the following steps: calculating a channel estimate from the received signal; calculating an amplitude spectrum of the channel estimation, and filtering the amplitude spectrum; and folding and averaging the filtered amplitude spectrum, estimating the noise level, calculating a detection threshold value, and calculating and outputting the maximum Doppler frequency offset.

Preferably, the magnitude spectrum is filtered by the following formula: <math> <mrow> <msub> <mover> <mi>Y</mi> <mo>^</mo> </mover> <mi>sm</mi> </msub> <mrow> <mo>(</mo> <mi>k</mi> <mo>)</mo> </mrow> <mo>=</mo> <msub> <mi>&Sigma;</mi> <mrow> <mo>-</mo> <mi>L</mi> <mo>&le;</mo> <mi>n</mi> <mo>&le;</mo> <mi>L</mi> </mrow> </msub> <mover> <mi>Y</mi> <mo>^</mo> </mover> <mrow> <mo>(</mo> <mi>mod</mi> <mrow> <mo>(</mo> <mi>k</mi> <mo>-</mo> <mi>n</mi> <mo>,</mo> <mi>M</mi> <mo>)</mo> </mrow> <mo>)</mo> </mrow> <mi>h</mi> <mrow> <mo>(</mo> <mi>k</mi> <mo>)</mo> </mrow> <mo>,</mo> </mrow> </math> where h (k) is the filter coefficient, mod (·, M) is the modulo operation on M, M is the number of points for FFT, and k is greater than or equal to 0 and less than M.

Preferably, the filtered amplitude spectrum is fold averaged by the following formula: ${\hat{Y}}_{\mathrm{half}}\left(k\right)=({\hat{Y}}_{\mathrm{sm}}\left(k\right)+{\hat{Y}}_{\mathrm{sm}}\left(\mathrm{mod}(M-k,M)\right))/2,$ wherein,

mod (·, M), which is the modulo operation on M, <math> <mrow> <mn>0</mn> <mo>&le;</mo> <mi>k</mi> <mo>&le;</mo> <mfrac> <mi>M</mi> <mn>2</mn> </mfrac> <mo>.</mo> </mrow> </math>

preferably, estimating the noise level comprises: calculating the maximum value of the folded averaged amplitude spectrum: <math> <mrow> <msub> <mover> <mi>Y</mi> <mo>^</mo> </mover> <mi>max</mi> </msub> <mo>=</mo> <munder> <mi>max</mi> <mrow> <mn>0</mn> <mo>&le;</mo> <mi>k</mi> <mo>&lt;</mo> <mfrac> <mi>M</mi> <mn>2</mn> </mfrac> </mrow> </munder> <mo>{</mo> <msub> <mover> <mi>Y</mi> <mo>^</mo> </mover> <mi>half</mi> </msub> <mrow> <mo>(</mo> <mi>k</mi> <mo>)</mo> </mrow> <mo>}</mo> <mo>,</mo> </mrow> </math> wherein,

is the folded averaged amplitude spectrum; calculating the average value of the partial frequency spectrum of the folded average amplitude spectrum: $\hat{N}=\underset{\frac{M}{4}<k<\frac{M}{2}}{\mathrm{mean}}\left\{{\hat{Y}}_{\mathrm{half}}\left(k\right)\right\};$ calculating a noise level of ${\hat{N}}_0:{\hat{N}}_0=\hat{N}/{\hat{Y}}_{\max }.$

Preferably, calculating the detection threshold comprises: a detection threshold thr of <math> <mrow> <mi>thr</mi> <mo>=</mo> <msub> <mover> <mi>Y</mi> <mo>^</mo> </mover> <mi>max</mi> </msub> <mo>&CenterDot;</mo> <mi>&alpha;</mi> <mo>+</mo> <mover> <mi>N</mi> <mo>^</mo> </mover> <mo>&CenterDot;</mo> <mrow> <mo>(</mo> <mn>1</mn> <mo>-</mo> <mi>&alpha;</mi> <mo>)</mo> </mrow> <mo>,</mo> </mrow> </math> Wherein alpha is a constant less than 1 in a preset mode,

is the maximum value of the amplitude spectrum,

is the average value of part of the frequency spectrum in the folded and averaged amplitude spectrum.

Preferably, calculating the maximum doppler frequency offset comprises: searching out the maximum index with the spectrum value larger than the threshold value according to the following formula <math> <mrow> <msub> <mi>k</mi> <mi>edge</mi> </msub> <mo>=</mo> <mi>max</mi> <mo>{</mo> <mi>k</mi> <mo>;</mo> <mo>|</mo> <msub> <mover> <mi>Y</mi> <mo>^</mo> </mover> <mi>half</mi> </msub> <mrow> <mo>(</mo> <mi>k</mi> <mo>)</mo> </mrow> <mo>|</mo> <mo>></mo> <mi>thr</mi> <mo>,</mo> <mn>0</mn> <mo>&le;</mo> <mi>k</mi> <mo>&lt;</mo> <mfrac> <mi>M</mi> <mn>4</mn> </mfrac> <mo>}</mo> <mo>,</mo> </mrow> </math> Wherein,

the folded and averaged amplitude spectrum is obtained, thr is a detection threshold value, and M is the number of points of FFT (fast Fourier transform); according to the maximum index k_edgeCalculating the maximum Doppler frequency offset f_d： <math> <mrow> <msub> <mi>f</mi> <mi>d</mi> </msub> <mo>=</mo> <mfrac> <msub> <mi>k</mi> <mi>edge</mi> </msub> <mi>M</mi> </mfrac> <mo>&CenterDot;</mo> <mfrac> <mn>1</mn> <mi>T</mi> </mfrac> <mo>,</mo> </mrow> </math> Where T is the sampling time interval of the channel estimation.

Preferably, k is determined according to the maximum index_edgeCalculating the maximum Doppler frequency offset f_dPreviously, the method further comprises: the maximum index is performed according to the following formulaLinear interpolation to obtain the maximum index k after correction_edgeSo as to be based on the modified maximum index k_edgeCalculating the maximum Doppler frequency offset f_d： $k_{\mathrm{edge}}=k_{\mathrm{edge}}+(\mathrm{thr}-{\hat{Y}}_{\mathrm{half}}\left(k_{\mathrm{edge}}\right))/({\hat{Y}}_{\mathrm{half}}(k_{\mathrm{edge}}+1)-{\hat{Y}}_{\mathrm{half}}\left(k_{\mathrm{edge}}\right)).$

Preferably, k is determined according to the maximum index_edgeCalculating the maximum Doppler frequency offset f_dAfterwards, the method further comprises: and calculating and updating the filter coefficient according to the noise level and the maximum Doppler frequency offset.

Preferably, calculating and updating the filter coefficients based on the noise level and the maximum doppler shift comprises: determining noise level

Whether the value is less than a preset threshold value TH, if so, setting the filter coefficient as a unit impulseA shock function; otherwise, the sigma value of the filter coefficient is set to be sigma beta (f)_dMT) and normalizing the gaussian filter.

Preferably, after calculating and updating the filter coefficients according to the noise level and the maximum doppler frequency offset, the method further comprises: and carrying out filtering processing on the amplitude spectrum according to the updated filter coefficient.

In order to achieve the above object, according to another aspect of the present invention, there is provided an apparatus for estimating maximum doppler frequency offset.

The maximum Doppler frequency offset estimation device comprises: a first calculation module for calculating a channel estimate from the received signal; the second calculation module is used for calculating the amplitude spectrum of the channel estimation obtained by the first calculation module; the filtering module is used for filtering the amplitude spectrum obtained by the second calculating module; and the processing module is used for performing folding average on the amplitude spectrum filtered by the filtering module, estimating the noise level, calculating a detection threshold value, and calculating and outputting the maximum Doppler frequency offset.

Preferably, the above apparatus further comprises: the third calculation module is used for calculating a filter coefficient according to the noise level and the maximum Doppler frequency offset obtained by the processing module; and the updating module is used for updating by using the filter coefficient obtained by the third calculating module.

According to the invention, the amplitude spectrum of channel estimation is calculated, the filtered amplitude spectrum is subjected to folding averaging, the noise level is estimated, the detection threshold value is calculated, and the maximum Doppler frequency offset is calculated and output, so that the problem that the algorithm estimation value for acquiring multi-frequency offset information in the related technology is poor in stability or low in estimation precision is solved, the stability is further improved, and higher estimation precision can be obtained.

Drawings

The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the invention and together with the description serve to explain the invention without limiting the invention. In the drawings:

FIG. 1 is a flow chart of a method of estimating maximum Doppler shift according to an embodiment of the invention;

fig. 2 is a schematic diagram of a frame structure of a CMMB system physical layer of one of application environments according to an embodiment of the present invention;

fig. 3 is a schematic diagram of the OFDM symbol composition structure of fig. 2;

fig. 4 is a schematic diagram of the allocation of effective subcarriers of an OFDM symbol into data subcarriers, scattered pilots, and continuous pilots, and the allocation manner according to an embodiment of the present invention;

FIG. 5 is a diagram of a maximum Doppler shift estimation architecture according to an embodiment of the present invention;

fig. 6 is a block diagram of an apparatus for estimating maximum doppler frequency offset according to an embodiment of the present invention;

fig. 7 is a block diagram of a preferred structure of an apparatus for estimating maximum doppler frequency offset according to an embodiment of the present invention.

Detailed Description

Overview of the function

In consideration of the problem of poor stability or low estimation accuracy of an algorithm estimation value for acquiring multi-frequency offset information in the related art, the embodiment of the invention provides a Doppler frequency offset acquisition method of a CMMB system, wherein the Doppler frequency offset is acquired by using discrete pilot frequency of a frequency domain received signal, and the Doppler frequency offset is estimated by detecting the spectral width based on a spectral method; and selecting a proper filter according to the possible size of the Doppler value to filter the spectrum, then folding and averaging the spectrum, and folding and averaging the spectrum through detection filtering, so that the algorithm can obtain high stability in a severe environment with low signal-to-noise ratio. The estimation of the Doppler frequency offset does not need to obtain the signal-to-noise ratio of a channel, the noise has small interference on the estimation of the Doppler frequency offset, and higher estimation precision can be obtained. The method overcomes the defect of Doppler frequency offset estimation in the prior orthogonal frequency division multiplexing system.

It should be noted that the embodiments and features of the embodiments in the present application may be combined with each other without conflict. The present invention will be described in detail below with reference to the embodiments with reference to the attached drawings.

Method embodiment

According to an embodiment of the present invention, a method for estimating maximum doppler frequency offset is provided. Fig. 1 is a flowchart of a method for estimating maximum doppler frequency offset according to an embodiment of the present invention, as shown in fig. 1, the method includes the following steps 102 to 106:

step 102, calculating a channel estimate from the received signal.

And 104, calculating the amplitude spectrum of the channel estimation, and filtering the amplitude spectrum.

And 106, performing folding average on the filtered amplitude spectrum, estimating the noise level, calculating a detection threshold, and calculating and outputting the maximum Doppler frequency offset.

It should be noted that the detection threshold is selected as <math> <mrow> <mi>thr</mi> <mo>=</mo> <msub> <mover> <mi>Y</mi> <mo>^</mo> </mover> <mi>max</mi> </msub> <mo>&CenterDot;</mo> <mi>&alpha;</mi> <mo>+</mo> <mover> <mi>N</mi> <mo>^</mo> </mover> <mo>&CenterDot;</mo> <mrow> <mo>(</mo> <mn>1</mn> <mo>-</mo> <mi>&alpha;</mi> <mo>)</mo> </mrow> <mo>,</mo> </mrow> </math> Wherein alpha is a constant less than 1 in a preset mode,

is the maximum value of the spectrum and,

for the averaging of the partial spectrum values, the average value of the latter part of the spectrum of the folded amplitude is generally chosen.

When calculating the maximum Doppler frequency offset, firstly, the maximum index with the spectrum value larger than the threshold value is searched <math> <mrow> <msub> <mi>k</mi> <mi>edge</mi> </msub> <mo>=</mo> <mi>max</mi> <mo>{</mo> <mi>k</mi> <mo>;</mo> <mo>|</mo> <msub> <mover> <mi>Y</mi> <mo>^</mo> </mover> <mi>half</mi> </msub> <mrow> <mo>(</mo> <mi>k</mi> <mo>)</mo> </mrow> <mo>|</mo> <mo>></mo> <mi>thr</mi> <mo>,</mo> <mn>0</mn> <mo>&le;</mo> <mi>k</mi> <mo>&lt;</mo> <mfrac> <mi>M</mi> <mn>4</mn> </mfrac> <mo>}</mo> <mo>,</mo> </mrow> </math> Wherein,

for the spectrum after the average doubling, thr is a detection threshold value, and M is the number of FFT points; preferably, in order to improve the resolution, the above maximum index (edge point) may be linearly interpolated according to the following formula so as to be based on the modified maximum index k_edgeCalculating the maximum Doppler frequency offset f_d： $k_{\mathrm{edge}}=k_{\mathrm{edge}}+(\mathrm{thr}-{\hat{Y}}_{\mathrm{half}}\left(k_{\mathrm{edge}}\right))/({\hat{Y}}_{\mathrm{half}}(k_{\mathrm{edge}}+1)-{\hat{Y}}_{\mathrm{half}}\left(k_{\mathrm{edge}}\right)),$ Then, a Doppler frequency offset estimation value is output <math> <mrow> <msub> <mi>f</mi> <mi>d</mi> </msub> <mo>=</mo> <mfrac> <msub> <mi>k</mi> <mi>edge</mi> </msub> <mi>M</mi> </mfrac> <mo>&CenterDot;</mo> <mfrac> <mn>1</mn> <mi>T</mi> </mfrac> <mo>,</mo> </mrow> </math> Where T is the sampling time interval of the channel estimation.

Thereafter, a filter coefficient is calculated based on the estimated noise level and the maximum doppler frequency offset, and the filter coefficient is updated so that the magnitude spectrum is subjected to filtering processing based on the updated filter coefficient. Returning to step 102, the above steps are repeated.

Specifically, judgment

And

whether the ratio is smaller than a preset threshold value TH or not, if so, setting the filter coefficient as a unit impulse function; otherwise, the filter coefficients are set to sigma values of σ ═ β · (f)_dMT) and normalizing the gaussian filter. Coefficient of Gaussian filter is h_g(k)＝exp(-k²/(2σ²) Is more than or equal to 0 and is less than 3 sigma; normalized coefficient is h (k) h_g(k)/∑_kh_g(k)。

By the embodiment, the method for estimating the Doppler frequency offset in the mobile communication system is provided, and the method can obtain higher estimation precision which is not influenced by a channel estimation result.

The following describes in detail the implementation of the embodiments of the present invention with reference to examples.

There are many different types of OFDM systems and the implementation of the embodiments of the present invention will vary for different types of OFDM systems. The following description will be made of a case where the present invention is applied to an OFDM system such as CMMB.

Fig. 2 is a diagram illustrating a frame structure of a physical layer of a CMMB system according to one of application environments of the embodiment of the present invention, as shown in fig. 2, in the frame structure of the physical layer of the CMMB system, 1 second is equally divided into 40 slots (slot 0 to slot 39), each slot is 25ms, and each slot is composed of 1 beacon and 53 OFDM symbols (OFDM symbol 0 to OFDM symbol 52).

Fig. 3 is a schematic diagram of the OFDM symbol composition structure in fig. 2, and as shown in fig. 3, an OFDM symbol is composed of a cyclic Prefix (CP for short) and an OFDM data body. OFDM data body length (T)_U) 409.6 mus, cycle length (T)_CP) 51.2 mus, OFDM symbol length (T)_s) 460.8 μ s.

Fig. 4 is a schematic diagram of allocating effective subcarriers of an OFDM symbol to data subcarriers, scattered pilots, and continuous pilots, and an allocation manner according to an embodiment of the present invention, and as shown in fig. 4, the allocation of effective subcarriers of an OFDM symbol to data subcarriers, scattered pilots, and continuous pilots, and an allocation manner is shown.

Fig. 5 is a schematic diagram of a maximum Doppler frequency offset estimation structure according to an embodiment of the present invention, and as shown in fig. 5, based on the above frame structure, a Doppler (Doppler) frequency offset obtaining method in an embodiment of the present invention includes the following steps:

step 1: a channel estimate is calculated from the received signal.

Specifically, a pilot subcarrier receiving signal is extracted, that is, a continuous pilot signal of a frequency domain signal received in a time slot is extracted, because the continuous pilot signal sent by the CMMB system is a fixed value 1, the received continuous pilot signal is actually a channel response of a pilot point, and each pilot subcarrier corresponds to a group of channel estimation sampling values.

Step 2: the magnitude spectrum of the channel estimate, i.e. the magnitude spectrum of the pilot signal, is calculated.

An M-point Fast Fourier Transform Algorithm (FFT) is performed on the extracted pilot signal with respect to the time direction. Each continuous pilot frequency subcarrier corresponds to a group of FFT transformation data, and the absolute value of the transformation data is taken for processing to obtain an amplitude spectrum. Adding and averaging the amplitude spectrums of different continuous pilot frequencies to obtain a time slot amplitude spectrum, and recording the time slot amplitude spectrum as

And step 3: the filtering processing is carried out on the time slot amplitude spectrum, so that the noise burrs of the amplitude spectrum can be effectively inhibited, the spectrum becomes smooth, and the detection is easy.

Let the coefficients of the filter be given as { h (k) }_-L≤k≤LTo, for

When filtering, the boundary value is processed according to cycle continuation, and filtering is carried out according to the following formula:

<math> <mrow> <msub> <mover> <mi>Y</mi> <mo>^</mo> </mover> <mi>sm</mi> </msub> <mrow> <mo>(</mo> <mi>k</mi> <mo>)</mo> </mrow> <mo>=</mo> <msub> <mi>&Sigma;</mi> <mrow> <mo>-</mo> <mi>L</mi> <mo>&le;</mo> <mi>n</mi> <mo>&le;</mo> <mi>L</mi> </mrow> </msub> <mover> <mi>Y</mi> <mo>^</mo> </mover> <mrow> <mo>(</mo> <mi>mod</mi> <mrow> <mo>(</mo> <mi>k</mi> <mo>-</mo> <mi>n</mi> <mo>,</mo> <mi>M</mi> <mo>)</mo> </mrow> <mo>)</mo> </mrow> <mi>h</mi> <mrow> <mo>(</mo> <mi>k</mi> <mo>)</mo> </mrow> </mrow> </math> 0≤k＜M

where h (k) is the filter coefficient, mod (·, M) is the modulo operation on M, which is the number of points for FFT.

And 4, step 4: and folding and averaging the filtered amplitude spectrum, estimating the noise level, calculating a detection threshold, detecting and calculating the maximum Doppler frequency offset, and outputting a Doppler frequency offset estimation value. By folding and averaging the spectrum, the probability of misjudgment of spectrum edge detection can be reduced, meanwhile, the level of noise is obtained by the amplitude spectrum, and the detection threshold value is selected in a self-adaptive manner according to the noise.

The amplitude spectrum fold average formula is as follows:

${\hat{Y}}_{\mathrm{half}}\left(k\right)=({\hat{Y}}_{\mathrm{sm}}\left(k\right)+{\hat{Y}}_{\mathrm{sm}}\left(\mathrm{mod}(M-k.M)\right))/2$

wherein,

mod (·, M), which is the modulo operation on M, <math> <mrow> <mn>0</mn> <mo>&le;</mo> <mi>k</mi> <mo>&lt;</mo> <mfrac> <mi>M</mi> <mn>2</mn> </mfrac> <mo>.</mo> </mrow> </math>

calculating the maximum of the folded average amplitude spectrum:

<math> <mrow> <msub> <mover> <mi>Y</mi> <mo>^</mo> </mover> <mi>max</mi> </msub> <mo>=</mo> <munder> <mi>max</mi> <mrow> <mn>0</mn> <mo>&le;</mo> <mi>k</mi> <mo>&lt;</mo> <mfrac> <mi>M</mi> <mn>2</mn> </mfrac> </mrow> </munder> <mo>{</mo> <msub> <mover> <mi>Y</mi> <mo>^</mo> </mover> <mi>half</mi> </msub> <mrow> <mo>(</mo> <mi>k</mi> <mo>)</mo> </mrow> <mo>}</mo> </mrow> </math>

wherein,

is the folded averaged amplitude spectrum.

Calculating the average value of the frequency point values of the folded average amplitude spectrum part:

$\hat{N}=\underset{\frac{M}{4}<k<\frac{M}{2}}{\mathrm{mean}}\left\{{\hat{Y}}_{\mathrm{half}}\left(k\right)\right\}$

computing

And

is recorded as the ratio of

${\hat{N}}_0=\hat{N}/{\hat{Y}}_{\max }$

And use the ratio

The noise level is measured and retained for use in the next step.

Calculating a detection threshold value:

<math> <mrow> <mi>thr</mi> <mo>=</mo> <msub> <mover> <mi>Y</mi> <mo>^</mo> </mover> <mi>max</mi> </msub> <mo>&CenterDot;</mo> <mi>&alpha;</mi> <mo>+</mo> <mover> <mi>N</mi> <mo>^</mo> </mover> <mo>&CenterDot;</mo> <mrow> <mo>(</mo> <mn>1</mn> <mo>-</mo> <mi>&alpha;</mi> <mo>)</mo> </mrow> </mrow> </math>

the factor α is a positive number not greater than 1, and α is 0.5, for example.

Detecting the position of the edge point:

<math> <mrow> <msub> <mi>k</mi> <mi>edge</mi> </msub> <mo>=</mo> <mi>max</mi> <mo>{</mo> <mi>k</mi> <mo>;</mo> <mo>|</mo> <msub> <mover> <mi>Y</mi> <mo>^</mo> </mover> <mi>half</mi> </msub> <mrow> <mo>(</mo> <mi>k</mi> <mo>)</mo> </mrow> <mo>|</mo> <mo>></mo> <mi>thr</mi> <mo>,</mo> <mn>0</mn> <mo>&le;</mo> <mi>k</mi> <mo>&lt;</mo> <mfrac> <mi>M</mi> <mn>4</mn> </mfrac> <mo>}</mo> </mrow> </math>

k obtained_edgeAre integer values. To function

And linear interpolation is carried out, and the resolution ratio of the edge points is improved. Correction k_edgeThe value of (c):

$k_{\mathrm{edge}}=k_{\mathrm{edge}}+\frac{\mathrm{thr}-{\hat{Y}}_{\mathrm{half}}\left(k_{\mathrm{edge}}\right)}{{\hat{Y}}_{\mathrm{half}}(k_{\mathrm{edge}}+1)-{\hat{Y}}_{\mathrm{half}}\left(k_{\mathrm{edge}}\right)}$

calculating the maximum Doppler frequency offset by the following formula:

<math> <mrow> <msub> <mi>f</mi> <mi>d</mi> </msub> <mo>=</mo> <mfrac> <msub> <mi>k</mi> <mi>edge</mi> </msub> <mi>M</mi> </mfrac> <mo>&CenterDot;</mo> <mfrac> <mn>1</mn> <mi>T</mi> </mfrac> </mrow> </math>

wherein T is an OFDM symbol duration; then, the maximum Doppler frequency offset f is output_d。

And 5: and 4, calculating a filter coefficient according to the noise level and the maximum Doppler frequency offset estimated in the step 4, and updating the filter coefficient. And returning to the step 1, and repeating the steps.

Determining noise level

Whether it is large or notAt a predetermined threshold TH, if

Above a predetermined threshold TH, no filtering is required, i.e. the coefficients are set to h (0) 1, h (k) 0, where k ≠ 0;

otherwise, the filter coefficients are set to gaussian filter coefficients.

The size of the gaussian filter parameter σ is determined as follows:

σ＝β·(f_dMT)

the value of the constant β is a positive number smaller than 1, and may be set in advance.

The gaussian filter coefficients are:

<math> <mrow> <msub> <mi>h</mi> <mi>g</mi> </msub> <mrow> <mo>(</mo> <mi>k</mi> <mo>)</mo> </mrow> <mo>=</mo> <mi>exp</mi> <mrow> <mo>(</mo> <mo>-</mo> <mfrac> <msup> <mi>k</mi> <mn>2</mn> </msup> <mrow> <mn>2</mn> <msup> <mi>&sigma;</mi> <mn>2</mn> </msup> </mrow> </mfrac> <mo>)</mo> </mrow> </mrow> </math> 0≤|k|＜3σ

and normalizing the gaussian filter coefficients:

<math> <mrow> <mi>h</mi> <mrow> <mo>(</mo> <mi>k</mi> <mo>)</mo> </mrow> <mo>=</mo> <mfrac> <mrow> <msub> <mi>h</mi> <mi>g</mi> </msub> <mrow> <mo>(</mo> <mi>k</mi> <mo>)</mo> </mrow> </mrow> <mrow> <munder> <mi>&Sigma;</mi> <mi>k</mi> </munder> <msub> <mi>h</mi> <mi>g</mi> </msub> <mrow> <mo>(</mo> <mi>k</mi> <mo>)</mo> </mrow> </mrow> </mfrac> </mrow> </math>

and returning to the step 1, and repeating the steps 1 to 5.

It should be noted that, a gaussian filter is selected in step 5, and other types of window functions can also be selected and used, and similarly, the width of the window function is selected according to the previous doppler estimation value.

Device embodiment

According to an embodiment of the present invention, an apparatus for estimating maximum doppler frequency offset is provided. Fig. 6 is a block diagram of an apparatus for estimating maximum doppler frequency offset according to an embodiment of the present invention, as shown in fig. 6, the apparatus includes: the first calculating module 2, the second calculating module 4, the filtering module 6 and the processing module 8, which are described below.

A first calculation module 2, configured to calculate a channel estimate according to a received signal; the second calculation module 4 is connected to the first calculation module 2 and is used for calculating the amplitude spectrum of the channel estimation obtained by the first calculation module 2; the filtering module 6 is connected to the second calculating module 4 and is used for filtering the amplitude spectrum obtained by the second calculating module 4; and the processing module 8 is connected to the filtering module 6 and is used for performing folding averaging on the amplitude spectrum filtered by the filtering module 6, estimating the noise level, calculating the detection threshold value, and calculating and outputting the maximum Doppler frequency offset.

Fig. 7 is a block diagram of a preferred structure of an apparatus for estimating maximum doppler frequency offset according to an embodiment of the present invention, as shown in fig. 7, the apparatus preferably further includes:

a third calculating module 72, connected to the processing module 8, for calculating a filter coefficient according to the noise level and the maximum doppler frequency offset obtained by the processing module 8; and an updating module 74, connected to the third calculating module 72, for updating by using the filter coefficient obtained by the third calculating module 72, so that the filtering module 6 performs filtering processing according to the updated filter coefficient.

In summary, the embodiments of the present invention provide a method for detecting a doppler U-shaped spectral width, which adaptively selects a filter of parameters according to a spectral width and a noise level for filtering, performs spectrum folding averaging, and adaptively selects a detection threshold according to the noise level, so that noise interference is small, the method can operate in a low signal-to-noise ratio channel, and the accuracy of estimating the doppler frequency offset is high.

It will be apparent to those skilled in the art that the modules or steps of the present invention described above may be implemented by a general purpose computing device, they may be centralized on a single computing device or distributed across a network of multiple computing devices, and they may alternatively be implemented by program code executable by a computing device, such that they may be stored in a storage device and executed by a computing device, or fabricated separately as individual integrated circuit modules, or fabricated as a single integrated circuit module from multiple modules or steps. Thus, the present invention is not limited to any specific combination of hardware and software.

The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims (10)

1. A method for estimating maximum Doppler frequency offset is characterized by comprising the following steps:

calculating a channel estimate from the received signal;

calculating the amplitude spectrum of the channel estimation, and filtering the amplitude spectrum;

folding and averaging the filtered amplitude spectrum, estimating the noise level of the folded and averaged amplitude spectrum according to the maximum value of the folded and averaged amplitude spectrum and the average value of the partial frequency spectrum of the folded and averaged amplitude spectrum, calculating the detection threshold value of the folded and averaged amplitude spectrum according to the maximum value of the folded and averaged amplitude spectrum and the average value of the partial frequency spectrum of the folded and averaged amplitude spectrum, and calculating the maximum Doppler frequency offset of the folded and averaged amplitude spectrum according to the folded and averaged amplitude spectrum and the detection threshold value;

and calculating and updating a filter coefficient according to the noise level and the maximum Doppler frequency offset.

2. The method according to claim 1, characterized in that the magnitude spectrum is filtered by the following formula:

<math> <mrow> <msub> <mover> <mi>Y</mi> <mo>^</mo> </mover> <mi>sm</mi> </msub> <mrow> <mo>(</mo> <mi>k</mi> <mo>)</mo> </mrow> <mo>=</mo> <msub> <mi>&Sigma;</mi> <mrow> <mo>-</mo> <mi>L</mi> <mo>&le;</mo> <mi>n</mi> <mo>&le;</mo> <mi>L</mi> </mrow> </msub> <mover> <mi>Y</mi> <mo>^</mo> </mover> <mrow> <mo>(</mo> <mi>mod</mi> <mrow> <mo>(</mo> <mi>k</mi> <mo>-</mo> <mi>n</mi> <mo>,</mo> <mi>M</mi> <mo>)</mo> </mrow> <mo>)</mo> </mrow> <mi>h</mi> <mrow> <mo>(</mo> <mi>k</mi> <mo>)</mo> </mrow> <mo>,</mo> </mrow> </math> where h (k) is the filter coefficient, mod (·, M) is the modulo operation on M, M is the number of points for FFT, k is greater than or equal to 0<M。

3. The method according to claim 1, characterized in that the filtered amplitude spectrum is fold averaged by the following formula:

${\hat{Y}}_{\mathrm{half}}\left(k\right)=({\hat{Y}}_{\mathrm{sm}}\left(k\right)+{\hat{Y}}_{\mathrm{sm}}\left(\mathrm{mod}(M-k,M)\right))/2,$ wherein,

mod (·, M) is the modulo operation on M for the filtered amplitude spectrum,

4. the method of claim 1, wherein estimating the noise level comprises:

calculating the maximum value of the folded averaged amplitude spectrum:wherein,

the magnitude spectrum after folding averaging;

calculating an average value of the partial frequency spectrum of the magnitude spectrum after folding averaging:wherein M is the number of points of FFT transformation;

calculating the noise level as

5. The method of claim 1, wherein calculating the detection threshold comprises:

the detection threshold value thr isWherein alpha is a constant less than 1 in a preset mode,

is the maximum value of the amplitude spectrum,

is the average value of partial frequency spectrum in the amplitude spectrum after folding average.

6. The method of claim 1, wherein calculating the maximum doppler frequency shift comprises:

searching out the maximum index with the spectrum value larger than the threshold value according to the following formula <math> <mrow> <msub> <mi>k</mi> <mi>edge</mi> </msub> <mo>=</mo> <mi>max</mi> <mo>{</mo> <mi>k</mi> <mo>;</mo> <mo>|</mo> <msub> <mover> <mi>Y</mi> <mo>^</mo> </mover> <mi>half</mi> </msub> <mrow> <mo>(</mo> <mi>k</mi> <mo>)</mo> </mrow> <mo>|</mo> <mo>></mo> <mi>thr</mi> <mo>,</mo> <mn>0</mn> <mo>&le;</mo> <mi>k</mi> <mo>&lt;</mo> <mfrac> <mi>M</mi> <mn>4</mn> </mfrac> <mo>}</mo> <mo>,</mo> </mrow> </math> Wherein,

for the amplitude spectrum after the folding average, thr is the detection threshold value, and M is the number of points of FFT transformation;

according to the maximum index k_edgeCalculating the maximum Doppler frequency offset f_d：

Where T is the sampling time interval of the channel estimation.

7. The method of claim 6, wherein k is determined according to the maximum index_edgeCalculating the maximum Doppler frequency offset f_dPreviously, the method further comprises:

performing linear interpolation on the maximum index according to the following formula to obtain the corrected maximum index k_edgeSo as to be based on said modified maximum index k_edgeCalculating the maximum Doppler frequency offset f_d：

$k_{\mathrm{edge}}=k_{\mathrm{edge}}+(\mathrm{thr}-{\hat{Y}}_{\mathrm{half}}\left(k_{\mathrm{edge}}\right))\text{/}({\hat{Y}}_{\mathrm{half}}(k_{\mathrm{edge}}+)-{\hat{Y}}_{\mathrm{half}}\left(k_{\mathrm{edge}}\right)).$

8. The method of claim 7, wherein computing and updating filter coefficients based on the noise level and the maximum doppler shift comprises:

determining the noise level

Whether the coefficient is smaller than a preset threshold value TH or not, if so, setting the filter coefficient as a unit impulse function; otherwise, the sigma value of the filter coefficients is set to σ = β · (f)_dMT) and normalizing said gaussian filter; where M is the number of points of the FFT, T is an OFDMA symbol duration, and β is a positive number less than 1.

9. The method of claim 1, wherein after calculating and updating filter coefficients based on the noise level and the maximum doppler shift, the method further comprises:

and carrying out filtering processing on the amplitude spectrum according to the updated filter coefficient.

10. An apparatus for estimating maximum doppler frequency offset, comprising:

a first calculation module for calculating a channel estimate from the received signal;

the second calculation module is used for calculating the amplitude spectrum of the channel estimation obtained by the first calculation module;

the filtering module is used for filtering the amplitude spectrum obtained by the second calculating module;

the processing module is used for performing folding averaging on the amplitude spectrum filtered by the filtering module, estimating the noise level of the amplitude spectrum after folding averaging according to the maximum value of the amplitude spectrum after folding averaging and the average value of the partial frequency spectrum of the amplitude spectrum after folding averaging, calculating the detection threshold value of the amplitude spectrum after folding averaging according to the maximum value of the amplitude spectrum after folding averaging and the average value of the partial frequency spectrum of the amplitude spectrum after folding averaging, and calculating the maximum Doppler frequency offset of the amplitude spectrum after folding averaging according to the amplitude spectrum after folding averaging and the detection threshold value;

a third calculating module, configured to calculate a filter coefficient according to the noise level and the maximum doppler frequency offset obtained by the processing module;

and the updating module is used for updating by using the filter coefficient obtained by the third calculating module.

Images