CN102404268B - Method for estimating and compensating doppler frequency offset in Rician channels in high-speed mobile environment - Google Patents

Method for estimating and compensating doppler frequency offset in Rician channels in high-speed mobile environment Download PDF

Info

Publication number
CN102404268B
CN102404268B CN201110366374.1A CN201110366374A CN102404268B CN 102404268 B CN102404268 B CN 102404268B CN 201110366374 A CN201110366374 A CN 201110366374A CN 102404268 B CN102404268 B CN 102404268B
Authority
CN
China
Prior art keywords
frequency deviation
frequency
channel
epsiv
sign
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Expired - Fee Related
Application number
CN201110366374.1A
Other languages
Chinese (zh)
Other versions
CN102404268A (en
Inventor
任光亮
杨丽花
翟万涛
张会宁
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Xidian Univ
Original Assignee
Xidian Univ
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Xidian Univ filed Critical Xidian Univ
Priority to CN201110366374.1A priority Critical patent/CN102404268B/en
Publication of CN102404268A publication Critical patent/CN102404268A/en
Application granted granted Critical
Publication of CN102404268B publication Critical patent/CN102404268B/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Abstract

The invention discloses a method for estimating and compensating doppler frequency offset in Rician channels in a high-speed mobile environment, which mainly solves the problem that the estimation precision of the doppler frequency offset is not high and the problem of non best multiple compensation, comprising the following steps: firstly obtaining a time domain channel estimation by receiving signals through frequency domain channel estimation algorithm or time domain channel estimation algorithm; extracting the paths which have LOS component from the time domain channel estimation; obtaining the frequency offset estimation value with high precision through executing the frequency offset estimation by using the paths with the LOS component, and determining the best frequency offset compensation multiple through the signal to interference ratio of the maximal receiving signal; finally executing the best frequency offset compensation for the receiving signal, and removing the influence of the doppler frequency offset in the receiving signal. The invention effectively reduces the system performance loss caused by the doppler frequency offset, and can be used for improving the robustness of the doppler frequency offset applied by the OFDM/OFDMA (Orthogonal Frequency Division Multiplexing/Orthogonal Frequency Division Multiple Access) system.

Description

Doppler frequency offset in Rician channels in high-speed mobile environment is estimated and compensation method
Technical field
The invention belongs to the communications field, relate to OFDM/OFDM OFDM/OFDMA system, specifically under high-speed mobile environment, in OFDM/OFDMA system, a kind of high-precision doppler frequency deviation is estimated and compensation method, can be used for reducing the performance loss that Doppler frequency deviation brings.
Background technology
Along with constantly the building up, open and use of high-speed railway, highway, the third generation mobile communication system need to and the user of high-speed mobile between carry out high speed information transmission.With the communication under inactive state or lower-speed state, compare, the high speed information transmission under high-speed moving state will face more difficulty.Under high-speed mobile environment, because mobile terminal has high translational speed, this will introduce large Doppler frequency shift.The similar rural area of the wireless propagation environment scene of High-speed Circumstance, reflector is less, direct projection path is dominant, so this large Doppler frequency shift will cause that the direct projection LOS component that receives signal occurs that large frequency shift (FS) and scattering component occur to change fast, and this will cause systematic function degradation.Therefore the impact that, overcomes large Doppler is the key that improves systematic function.
The estimation of Doppler frequency deviation and the impact that compensation technique is often used to reduce Doppler frequency shift at present, and existing many patent documentations are studied this technology, but owing to receiving the severe jamming that has been subject to scattering component in signal with the LOS component of frequency deviation, under high-speed mobile environment, directly use these algorithms no longer can obtain high-precision frequency deviation estimated value, the algorithm of low estimated accuracy will cause systematic function to decline.Due to LOS component in Rice channel from scattering component with different frequency deviations, be on LOS component with on 1 times of Doppler frequency deviation and scattering component without the impact of Doppler frequency deviation, this is different from traditional backoff algorithm by the compensation technique that makes the Doppler frequency deviation in Rice channel: if LOS component has and has larger frequency deviation value on very little frequency deviation value and scattering component in the multipath reception signal after compensation, it is more violent that this will cause that scattering component changes, and make channel estimation technique have more challenge; If have on larger frequency deviation and scattering component and have very little frequency deviation value on LOS component in the reception signal after compensation, this is by the performance degradation of the reception signal that LOS component is dominant.
The existing algorithm to Doppler frequency deviation estimation and compensation roughly can be divided into two large classes: the data-aided method of class right and wrong, as the frequency deviation algorithm for estimating based on cyclic prefix CP; One class is data-aided algorithm, as the algorithm based on pilot tone and the algorithm based on data decision etc., these algorithms be all first by certain processing from receive signal or obtain Cyclic Prefix signal or obtain pilot signal or obtain decision data signal, and then the signal of these acquisitions is carried out to relevant treatment, thereby obtain frequency deviation estimated value, finally utilize these frequency deviation estimated values directly multipath reception signal to be carried out to compensate of frequency deviation.The applicable channel model of these algorithms is the channel model that has Doppler frequency deviation.Lay is obeyed in the first footpath of this kind of channel model, and this distributes, other footpath Rayleigh distributed.According to document " Time-varying carrier offsets inmobile OFDM ", IEEE Trans.commun., vol.57, no.9, Sept.2009, S.Talbot, B.Farhang-Boroujeny, when pp.2790-2798 sets up, become the mode of Doppler frequency shift rician fading channel model, this consider LOS component be one with the constant of Doppler frequency shift, and scattering component is with the multiple decline process of doppler spectral arbitrarily.Therefore, the sampling of n on m symbol period of i subframe reception signal can be expressed as:
r i ( m , n ) = cx i ( m , n ) e j 2 πϵ i ( m ) ( mN s - N + n ) N + Σ l p = 0 L - 1 h i , l p ( m , n ) x i ( m , n - l p ) + w i ( m , n )
In formula, N and N gbe respectively the length of FFT and Cyclic Prefix, N s=N+N g, be the discrete time channel impulse response of i subframe, Lay is obeyed in this channel first footpath, and this distributes, and its Rice factor is K=|c| 2/ var[h i, 0], wherein c and h i, 0respectively LOS component and the scattering component in the first footpath, other footpaths l p=1 ..., L-1 is scattering component, Rayleigh distributed, the footpath number that L is channel, x i(m, n) and w i(m, n) is respectively time domain transmitted signal, and (its power is σ x 2) and white Gaussian noise signal (its covariance is σ w 2), ε i(m) be the normalization Doppler frequency deviation on m symbol period of i subframe.
Carry out to received signal after fast Fourier transform FFT conversion, the reception signal on l subcarrier is:
R i ( m , l ) = 1 N e j πϵ i ( m ) ( N - 1 ) n e j 2 πϵ i ( m ) ( mN s - N ) N cX i ( m , l ) sin [ πϵ i ( m ) ] sin [ πϵ i ( m ) / N ] + I i ( m , l ) + W i ( m , l )
In formula, X i(m, l) is the transmitted signal on l subcarrier of m symbol period of i subframe, and
I i ( m , l ) = H i x ( m , l ) + 1 N e j 2 πϵ i ( m ) ( mN s - N ) N c
× Σ k ∈ κ k ≠ l X i ( m , k ) sin [ π ( k + ϵ i ( m ) - l ) ] sin [ π ( k + ϵ i ( m ) - l ) / N ] e j πϵ i ( m ) ( k + ϵ i ( m ) - l ) ( N - 1 ) N
H i x ( m , l ) = 1 N Σ n = 0 N - 1 [ Σ l p = 0 L - 1 h i , l p ( m , n ) x i ( m , n - l p ) ] e - j 2 πnl N
W i ( m , l ) = 1 N Σ n = 0 N - 1 w i ( m , n ) e - j 2 πnl N
Wherein, κ is the set of useful subcarrier in OFDM/OFDMA symbol, and its number is N u.If user side translational speed considers that 350km/h and carrier frequency are 2.3GHz, now in system, maximum doppler frequency is about 800Hz.Due to a sub-frame time on the cycle variation of Doppler frequency shift be no more than 1Hz, therefore, the Doppler frequency shift becoming when a subframe is on the time can be approximated to be a constant, i.e. ε i(m) ≈ ε i.
As shown in Figure 1, it comprises the steps: the specific implementation of existing Doppler frequency deviation estimation and backoff algorithm
(1) from receive signal, obtain the signal of expectation, this desired signal is Cyclic Prefix signal or pilot signal or decision data signal;
(2) desired signal obtaining is carried out to relevant treatment, the algorithm of take here based on pilot tone is example, l prelevant rear signal on individual subcarrier is:
In formula, m pfor the OFDM/OFDMA symbol logo with pilot tone, l pfor the sign of pilot sub-carrier on OFDM/OFDMA symbol, D is two OFDM/OFDMA symbolic numbers of carrying out the mark space of relevant treatment, for distracter:
I ‾ l p = 1 N cexp [ jπϵ ( N - 1 ) / N ] exp [ j 2 πϵ ( ( m p + D ) N s - N ) / N ] sin πϵ sin ( πϵ / N )
X ( m p , l p ) I * ( m p , l p ) + X * ( m p + D , l p ) R * ( m p , l p ) I ( m p + D , l p ) X ( m p , l p )
With for noise correlations terms, its tool white Gaussian noise characteristic:
W ‾ l p = 1 N exp [ jπϵ ( N - 1 ) N ] exp [ j 2 πϵ ( ( m p + D ) N s - N ) N ] c sin πϵ sin ( πϵ / N ) X ( m p , l p ) W * ( m p , l p )
+ X * ( m p + D , l p ) R * ( m p , l p ) W ( m p + D , l p ) X ( m p , l p )
In formula, N and N gbe respectively the length of FFT and Cyclic Prefix, N s=N+N g, c is the LOS component in channel the first footpath.
(3) utilize step (2) to obtain frequency deviation estimated value in formula, ∠ is for getting angle arithmetic operation.
(4) utilize the frequency deviation estimated value obtaining in step (3) multipath reception signal is carried out to compensate of frequency deviation, and the reception signal after compensation is:
r ‾ ( m , n ) = r ( m , n ) e - j 2 π ϵ ^ ( mN s - N + n ) N
= 1 N Σ k ∈ κ X ( m , k ) [ c ‾ e j 2 πn ( k + ϵ ‾ ) N + H m ϵ ( n , k ) e j 2 πn ( k - ϵ ^ ) N ] + w ‾ ( m , n )
In formula, c=cexp[j2 π ε(mN s-N)/N], H m ϵ ( n , k ) = H m ( n , k ) exp ( - j 2 π ϵ ^ ( mN s - N ) / N ) , w ‾ ( m , n ) = w ( m , n ) exp ( - j 2 π ϵ ^ ( mN s - N + n ) / N ) , ϵ ‾ = ϵ - ϵ ^ .
Existing Doppler frequency deviation is estimated mainly to comprise two aspects with the shortcoming of compensation technique:
1) estimated accuracy is not high
From step (2), can see: the existence of scattering component causes receiving the larger distracter of existence signal, and existing frequency offset estimation technique is not considered the impact of this interference components on the LOS component with frequency deviation, directly from multipath reception signal, extract desired signal, and utilize this desired signal to carry out frequency deviation and estimate to process, this will cause frequency offset estimation accuracy greatly to reduce, thereby cause performance loss.The Long Term Evolution OFDM LTE-OFDMA system of take is example, and under ITU-VAx channel model, the mean square error MSE performance of prior art as shown in Figure 2.As can be seen from Figure 2, due to the impact of multiple scattering component, the estimated accuracy based on Cyclic Prefix algorithm of the prior art is the poorest, though and based on pilot tone and judgement data two kinds of frequency deviation algorithm for estimating, there is higher estimated accuracy, but its MSE performance is 10 -4to there is error substrate in left and right, utilize the systematic function that these prior aries obtain to have the loss of more dB.Therefore, reducing or overcome scattering component is the key that improves frequency offset estimation accuracy on the impact of LOS component.
2) non-best compensate of frequency deviation
Signal after step (4) frequency deviation compensation is carried out to FFT to be processed
In formula, W ( m , l ) = 1 N Σ n = 0 N - 1 w ‾ ( m , n ) e - j 2 πnl N
H ‾ m ϵ ( k , l ) = 1 N [ c ‾ e jπ ( k - l + ϵ - ϵ ‾ ) N sin [ π ( k - l + ϵ - ϵ ‾ ) ] sin [ π ( k - l + ϵ - ϵ ‾ ) / N ] + e - j 2 π ϵ ‾ ( mN s - N ) N Σ n = 0 N - 1 H m ( n , k ) e j 2 πn ( k - l - ϵ ‾ ) N ]
According to patent " Channel estimation for rapid dispersive fading channels, " US patent, US20090103666,2009, M.Zhao, Z.N.Shi, M.Reed, the method for derivation interference power, the average signal interference ratio SIR that can obtain receiving signal is:
SIR = [ P L | sin π ϵ ‾ sin π ϵ ‾ / N | 2 + P S ( N + 2 Σ r = 1 N - 1 J 0 ( 2 π f max r T s ) cos ( 2 π ϵ ^ r / N ) ) ] / P ICI
In formula, P lfor the power of Rice channel LOS component, P l=| c| 2, c is channel the first footpath LOS component coefficient, P sfor the power of all scattering components of Rice channel and, for channel l pthe coefficient in footpath, var[] for getting variance symbol, f maxfor maximum doppler frequency, T sfor the sampling interval, J 0() is first kind zero Bessel function, P iCIbe average interference power, it can be expressed as:
P ICI = P L N u N 2 Σ p = - N u + 1 N u - 1 ( N u - | p | ) | sin π ( p + ϵ ‾ ) sin [ π ( p + ϵ ‾ ) / N ] | 2 - P L N 2 | sin π ϵ ‾ sin [ π ϵ ‾ / N ] | 2 + N u - 1 N P S
+ 2 P S N u N 2 Σ q = 1 N - 1 ( N - q ) J 0 ( 2 π f max T s q ) [ Σ p = - N u + 1 N u - 1 ( N u - | p | ) cos ( 2 πq ( p - ϵ ^ ) N ) - N u cos ( - 2 πq ϵ ^ N ) ]
N in formula unumber for useful subcarrier in OFDM/OFDMA symbol.
From the expression formula of average signal interference ratio SIR, can find out: the size of the signal interference ratio of the reception signal compensate of frequency deviation depends on the size of the factors such as Rice factor, inherent spurious frequency deviation value and Doppler frequency shift.Therefore, during compensate of frequency deviation, should consider the impact of these factors on systematic function, select best compensate of frequency deviation multiple.
Suppose to utilize existing algorithm can accurately estimate to obtain frequency deviation value, now after compensate of frequency deviation, receive the situation of change of interference power in signal as shown in Figure 3.As can be seen from Figure 3: utilizing frequency deviation estimated value directly to compensate to received signal SIR in rear reception signal will diminish, this be due to LOS component in Rice channel from scattering component with different frequency deviation values, if eliminate the frequency deviation on LOS component completely, can cause scattering component to change more violent, thereby cause larger interference.In addition, after compensate of frequency deviation, the SIR of system has the loss of more dB than the SIR without under offset frequency situation.Along with the increase of Rice factor, the signal interference ratio that receives signal becomes large.Therefore, Rice channel frequency deviation backoff algorithm has important impact to systematic function, and best compensate of frequency deviation mode is to improve the key of systematic function under Rice channel.
Summary of the invention
The present invention is directed to the deficiency of above-mentioned prior art, towards OFDM/OFDMA class wireless communication system, proposed a kind of doppler frequency offset in Rician channels in high-speed mobile environment and estimated and compensation method, to improve the estimated accuracy of frequency deviation value, and choose best compensate of frequency deviation multiple, improve systematic function.
For achieving the above object, the present invention provides following two kinds of technical schemes:
Technical scheme 1, obtains time domain channel by channel estimation in frequency domain algorithmic transformation and estimates, comprising:
(1) frequency deviation estimating step:
(1a) extract the footpath with direct projection component LOS
First, extract frequency domain and receive the OFDM/OFDM OFDM/OFDMA frequency pilot sign in signal, and utilize frequency pilot sign and the frequency pilot sign in transmitted signal in this reception signal to carry out channel estimating, obtain pilot frequency locations domain channel response m wherein pfor the sign with frequency pilot sign, k psign for pilot sub-carrier position;
Secondly, to pilot frequency locations domain channel response carry out interpolation, obtain the channel frequency domain response on other data subcarriers on whole frequency pilot sign and the current upper domain channel response estimated value with frequency pilot sign is designated as k wherein dfor the sign of data subcarrier position, k is the sign of the sub-carrier positions that takies of system;
Again, to domain channel response carry out IFFT conversion process, obtain time domain channel estimated value h ^ m p = [ h ^ ( m p , 0 ) , . . . , h ^ ( m p , L - 1 ) ] , In formula, the estimated value of time domain channel the first footpath coefficient, the estimated value of time domain channel L-1 footpath coefficient, for the mean value of whole OFDM/ODMA symbol upper signal channel response, and be positioned at the centre position of this OFDM/OFDMA symbol;
Finally, from time domain channel estimated value middle extraction is with the footpath of LOS component, i.e. the first footpath: h ^ los ( m p ) = h ^ ( m p , 0 ) ;
(1b) footpath with LOS component that utilizes estimation to obtain carry out frequency deviation estimation, obtain Doppler frequency deviation estimated value:
ϵ ^ = N 2 π N s DN m Σ m p ∈ κ p arg { h ^ los ( m p + D ) h ^ los * ( m p ) }
In formula, κ pfor the OFDM/OFDMA character position set with pilot tone, m pfor the sign of the OFDM/OFDMA symbol with pilot tone, N s=N+N g, N and N gbe respectively the length of FFT and Cyclic Prefix, arg{} represents to get angle symbol manipulation, and D is two symbolic numbers of carrying out the OFDM/OFDMA mark space of relevant treatment, N mfor carrying out the OFDM/OFDMA group number of relevant treatment, for position m p+ D place is with the footpath coefficient estimated value of LOS component, represent conjugate;
(2) best Doppler frequency deviation compensation process:
(2a) the signal interference ratio SIR of signal after calculating compensate of frequency deviation
Suppose for the frequency deviation value of compensate of frequency deviation to be wherein ρ ∈ [0 1] is compensation multiple, for the estimated value of frequency deviation, after compensate of frequency deviation, the signal interference ratio SIR of signal is:
SIR = [ P L | sin π ϵ ‾ sin π ϵ ‾ / N | 2 + P S ( N + 2 Σ r = 1 N - 1 J 0 ( 2 π f max r T s ) cos ( 2 πρ ϵ ^ r / N ) ) ] / P ICI
In formula, f maxfor maximum doppler frequency, T sfor the sampling interval, J 0() is first kind zero Bessel function, ε is true normalization frequency deviation, P lfor the power of Rice channel LOS component, P l=| c| 2, c is channel the first footpath LOS component coefficient, P sfor the power of all scattering components of Rice channel and, for channel l pthe coefficient in footpath, var[] for getting variance symbol, P iCIfor average interference power:
P ICI = P L N u N 2 Σ p = - N u + 1 N u - 1 ( N u - | p | ) | sin π ( p + ϵ ‾ ) sin [ π ( p + ϵ ‾ ) / N ] | 2 - P L N 2 | sin π ϵ ‾ sin [ π ϵ ‾ / N ] | 2 + N u - 1 N P S
+ 2 P S N u N 2 Σ q = 1 N - 1 ( N - q ) J 0 ( 2 π f max T s q ) [ Σ p = - N u + 1 N u - 1 ( N u - | p | ) cos ( 2 πq ( p - ϵ ^ ) N ) - N u cos ( - 2 πq ϵ ^ N ) ]
In formula, N unumber for useful subcarrier in OFDM/OFDMA symbol;
(2b) according to the size of the signal interference ratio SIR of signal after compensation, determine best compensate of frequency deviation multiple ρ, compensate of frequency deviation multiple ρ corresponding when soon SIR gets maximum is as best compensation multiple.
Technical scheme 2, directly obtains time domain channel by time domain channel algorithm for estimating and estimates, comprising:
1) frequency deviation estimating step:
1a) extract the footpath with LOS component
The OFDM/OFDMA time-domain signal receiving is adopted to time domain channel algorithm for estimating, obtains time domain channel estimated value: h ^ m p = [ h ^ ( m p , 0 ) , . . . , h ^ ( m p , L - 1 ) ] , Wherein the estimated value of time domain channel the first footpath coefficient, the estimated value of time domain channel L-1 footpath coefficient, for the mean value of whole OFDM/ODMA symbol upper signal channel response, and be positioned at the centre position of this OFDM/OFDMA symbol, m psign for the OFDM/OFDMA symbol with pilot tone;
From time domain channel estimated value middle extraction is with the footpath of LOS component, i.e. the first footpath: h ^ los ( m p ) = h ^ ( m p , 0 ) ;
The footpath with LOS component that 1b) utilizes estimation to obtain carry out frequency deviation estimation, obtain Doppler frequency deviation estimated value:
ϵ ^ = N 2 π N s DN m Σ m p ∈ κ p arg { h ^ los ( m p + D ) h ^ los * ( m p ) }
In formula, κ pfor the OFDM/OFDMA character position set with pilot tone, m pfor the sign of the OFDM/OFDMA symbol with pilot tone, N s=N+N g, N and N gbe respectively the length of FFT and Cyclic Prefix, arg{} represents to get angle symbol manipulation, and D is two symbolic numbers of carrying out the OFDM/OFDMA mark space of relevant treatment, N mfor carrying out the OFDM/OFDMA group number of relevant treatment, for position m p+ D place is with the footpath coefficient estimated value of LOS component, represent conjugation;
2) best Doppler frequency deviation compensation process:
2a) the signal interference ratio SIR of signal after calculating compensate of frequency deviation
Suppose for the frequency deviation value of compensate of frequency deviation to be wherein ρ ∈ [0 1] is compensation multiple, for the estimated value of frequency deviation, after compensate of frequency deviation, the signal interference ratio SIR of signal is:
SIR = [ P L | sin π ϵ ‾ sin π ϵ ‾ / N | 2 + P S ( N + 2 Σ r = 1 N - 1 J 0 ( 2 π f max r T s ) cos ( 2 πρ ϵ ^ r / N ) ) ] / P ICI
In formula, f maxfor maximum doppler frequency, T sfor the sampling interval, J 0() is first kind zero Bessel function, ε is true normalization frequency deviation, P lfor the power of Rice channel LOS component, P l=| c| 2, c is channel the first footpath LOS component coefficient, P sfor the power of all scattering components of Rice channel and, for channel l pthe coefficient in footpath, var[] for getting variance symbol manipulation, P iCIfor average interference power:
P ICI = P L N u N 2 Σ p = - N u + 1 N u - 1 ( N u - | p | ) | sin π ( p + ϵ ‾ ) sin [ π ( p + ϵ ‾ ) / N ] | 2 - P L N 2 | sin π ϵ ‾ sin [ π ϵ ‾ / N ] | 2 + N u - 1 N P S
+ 2 P S N u N 2 Σ q = 1 N - 1 ( N - q ) J 0 ( 2 π f max T s q ) [ Σ p = - N u + 1 N u - 1 ( N u - | p | ) cos ( 2 πq ( p - ϵ ^ ) N ) - N u cos ( - 2 πq ϵ ^ N ) ]
In formula, N unumber for useful subcarrier in OFDM/OFDMA symbol;
2b) according to the size of the signal interference ratio SIR of signal after compensation, determine best compensate of frequency deviation multiple ρ, compensate of frequency deviation multiple ρ corresponding when soon SIR gets maximum is as best compensation multiple.
Tool of the present invention has the following advantages:
(a) the present invention is directed in Rice channel and only have LOS component to there is the feature of Doppler frequency deviation, to the footpath containing LOS component, carry out separated with scattering footpath, only use with the footpath of LOS component and carry out frequency deviation estimation, reduce greatly the impact that scattering component is estimated frequency deviation, effectively improved the estimated accuracy of Doppler frequency deviation value.
(b) the present invention is directed to LOS component and the feature of scattering footpath with different frequency deviations in Rice channel, utilize the frequency deviation estimated value of best multiple to compensate to received signal, weighed and received LOS component and the impact of scattering component on systematic function in signal.
Simulation result shows: compared with prior art, the present invention has reduced the performance loss that under high velocity environment, in Rice channel, Doppler frequency deviation brings greatly, the frequency deviation that can directly apply in the OFDM/OFDMA type systematic under high-speed mobile environment is estimated and compensation, effectively improves the robustness of system to Doppler frequency deviation.
Object of the present invention, technical scheme, advantage can further illustrate by following drawings and embodiments.
Accompanying drawing explanation
Fig. 1 is the flow chart of existing frequency deviation estimation and compensation technique;
Fig. 2 be existing several frequency deviation algorithm for estimating under VAx channel model MSE Performance Ratio;
Fig. 3 receives the situation of change of interference power in signal before and after prior art compensate of frequency deviation under VAx channel model;
Fig. 4 is implementing procedure figure of the present invention;
Fig. 5 be the present invention and prior art under RAx.y channel model MSE Performance Ratio;
Fig. 6 is to be under 10 environment at VAx channel and Rice factor, receives the situation of change of interference power in signal during different compensate of frequency deviation multiple ρ.
Embodiment
One, know-why of the present invention
1) frequency deviation estimation principle
Owing to receiving the LOS component with Doppler frequency deviation in signal, be subject to the severe jamming of scattering component in channel, if directly utilize this reception signal to carry out frequency deviation, estimated to cause estimated accuracy greatly to decline.Therefore from receive signal, by separating with the footpath of LOS component, be, a kind of effective way that improves frequency offset estimation accuracy.The present invention by extract the footpath containing LOS component from time domain channel is estimated, is used for carrying out frequency deviation estimation just.
2) best Doppler frequency deviation compensation principle
Due to direct projection component in Rice channel from scattering component with different frequency deviation values, therefore, if have and there is larger inherent spurious frequency deviation value on very little inherent spurious frequency deviation and scattering component on the LOS component in the reception signal after compensation, this will cause that scattering footpath channel variation is more violent, cause precision of channel estimation to decline, thereby systematic function is declined; If have on larger frequency deviation value and scattering component and have very little frequency deviation value on the LOS component in the reception signal after compensation, this is by the performance degradation of the reception signal that causes LOS component to be dominant.Therefore the compensate of frequency deviation technology in Rice channel has important impact to systematic function, it is one of key technology improving systematic function.The present invention determines best compensation multiple according to the size of signal sir value after reception compensate of frequency deviation.
Two, embodiment of the present invention
With reference to accompanying drawing 4, the present invention provides following two embodiment according to channel estimating from frequency-domain and time-domain different angles.
Embodiment 1, from frequency domain angle, obtains time domain channel and estimates:
Step 1, extracts frequency domain and receives the OFDM/OFDM OFDM/OFDMA frequency pilot sign in signal, and utilize frequency pilot sign and the frequency pilot sign in transmitted signal in this reception signal to carry out channel estimating, obtains pilot frequency locations domain channel response: in formula, m pfor the sign with frequency pilot sign, k pfor the sign of pilot sub-carrier position, Y (m p, k p) be position (m p, k p) frequency domain located receives signal, X (m p, k p) be position (m p, k p) the frequency domain transmitted signal located.
Step 2, to pilot frequency locations domain channel response carry out interpolation, obtain the channel frequency domain response on other data subcarriers on whole frequency pilot sign the interpolation algorithm here can adopt any in linear interpolation, Gaussian interpolation and Cubic interpolation, wherein:
Linear interpolation formula:
H ^ ( m p , k d ) = H ^ ( m p , k p + m ) = ( 1 - m N f ) H ^ p ( m p , k p ) + m N f H ^ p ( m p , k p + N f )
Gaussian interpolation formula:
H ^ ( m p , k d ) = H ^ ( m p , k p + m )
= 1 2 [ ( m N f ) 2 - m N f ] H ^ p ( m p , k p - N f )
+ [ 1 - ( m N f ) 2 ] H ^ p ( m p , k p ) + 1 2 [ ( m N f ) 2 + m N f ] H ^ p ( m p , k p + N f )
Cubic interpolation formula:
H ^ ( m p , k d ) = H ^ ( m p , k p + m ) = ( 1 + 2 m N f ) ( m - N f - N f ) 2 H ^ p ( m p , k p )
+ ( 1 + 2 ( m - N f ) - N f ) ( m N f ) 2 H ^ p ( m p , k p + N f ) + m ( m - N f - N f ) 2 H ^ p * ( m p , k p )
+ ( m - N f ) ( m N f ) 2 H ^ p * ( m p , k p + N f )
In formula, data subcarrier position (the m drawing for interpolation p, k d) domain channel response, m pfor the sign with frequency pilot sign, k pfor the sign of pilot sub-carrier position, k dfor the sign of data subcarrier position, N fbe the sub-carrier number at two adjacent pilot sub-carrier intervals, m gets different value makes interpolation result traverse all data subcarrier positions, and m ∈ [1 N f-1], for position (m p, k p) channel estimation value located, position (the m obtaining for interpolation p, k p+ m) channel estimation value of locating, for position (m p, k p+ N f) channel estimation value located, for position (m p, k p-N f) channel estimation value located, represent conjugation, represent conjugation.
Step 3, to domain channel response carry out IFFT conversion process, obtain time domain channel estimated value h ^ m p = [ h ^ ( m p , 0 ) , . . . , h ^ ( m p , L - 1 ) ] , In formula, the estimated value of time domain channel the first footpath coefficient, the estimated value of time domain channel L-1 footpath coefficient, for the mean value of whole OFDM/ODMA symbol upper signal channel response, and be positioned at the centre position of this OFDM/OFDMA symbol.
Step 4, from time domain channel estimated value middle extraction is with the footpath of LOS component, i.e. the first footpath: h ^ los ( m p ) = h ^ ( m p , 0 ) .
Step 5, the footpath with LOS component that utilizes estimation to obtain carry out frequency deviation estimation, obtain Doppler frequency deviation estimated value:
ϵ ^ = N 2 π N s DN m Σ m p ∈ κ p arg { h ^ los ( m p + D ) h ^ los * ( m p ) }
In formula, κ pfor the OFDM/OFDMA character position set with pilot tone, m pfor the sign of the OFDM/OFDMA symbol with pilot tone, N s=N+N g, N and N gbe respectively the length of FFT and Cyclic Prefix, arg{} represents to get angle symbol manipulation, and D is two symbolic numbers of carrying out the OFDM/OFDMA mark space of relevant treatment, N mfor carrying out the OFDM/OFDMA group number of relevant treatment, for position m p+ D place is with the footpath coefficient estimated value of LOS component, represent conjugate.
Step 6, estimates frequency deviation ρ doubly as compensate of frequency deviation value be used for compensating reception signal, ρ ∈ [01] wherein, calculates the average interference power P that receives signal after compensate of frequency deviation iCI:
P ICI = P L N u N 2 Σ p = - N u + 1 N u - 1 ( N u - | p | ) | sin π ( p + ϵ ‾ ) sin [ π ( p + ϵ ‾ ) / N ] | 2 - P L N 2 | sin π ϵ ‾ sin [ π ϵ ‾ / N ] | 2
+ N u - 1 N P S + 2 P S N u N 2 Σ q = 1 N - 1 ( N - q ) J 0 ( 2 π f max T s q ) [ Σ p = - N u + 1 N u - 1 ( N u - | p | ) cos ( 2 πq ( p - ϵ ^ ) N ) - N u cos ( - 2 πq ϵ ^ N ) ]
In formula, what N was FFT counts, f maxfor maximum doppler frequency, T sfor the sampling interval, N ufor the number of useful subcarrier in OFDM/OFDMA symbol, J 0() is first kind zero Bessel function, ε is true normalization frequency deviation, P lfor the power of Rice channel LOS component, P l=| c| 2, c is channel the first footpath coefficient, P sfor the power of all scattering components of Rice channel and, for channel l pthe coefficient in footpath, var[] for getting variance symbol.
Step 7, according to average interference power P iCIafter calculating compensate of frequency deviation, receiving signal signal interference ratio SIR is:
SIR = [ P L | sin π ϵ ‾ sin π ϵ ‾ / N | 2 + P S ( N + 2 Σ r = 1 N - 1 J 0 ( 2 π f max r T s ) cos ( 2 πρ ϵ ^ r / N ) ) ] / P ICI .
Step 8, determines best compensate of frequency deviation multiple ρ according to the size of the signal interference ratio SIR of signal after compensation, and compensate of frequency deviation multiple ρ corresponding when soon SIR gets maximum is as best compensation multiple.
Embodiment 2, from time domain angle, obtain time domain channel and estimate:
Steps A, builds frequency domain sending metrix: X p=diag (X (m p, k 1), X (m p, k 2) ..., X (m p, k p)), in formula, m pfor the sign with frequency pilot sign, k pfor the sign of pilot sub-carrier position, X (m p, k p) be position (m p, k p) the frequency domain transmitted signal located, diag () represents diagonal matrix symbol.
Step B, by frequency domain sending metrix X pbuild intermediate data matrix: A p=X pbF km, in formula, F kfor N * N unit discrete time Fourier transform matrix, the length that N is Fourier transform, M is N * Q mapping matrix, M = E F , E is Q * Q unit matrix, the maximum multipath time delay length that Q is time domain channel response, and F is (N-Q) * Q null matrix, and B is P * N matrix, and its every a line only has a nonzero value 1, and the capable nonzero value 1 of p is positioned at k prow, P is the number of each symbol pilot sub-carrier.
Step C, builds frequency domain receiving matrix: Y p=[Y (m p, k 1), Y (m p, k 2) ..., Y (m p, k p)] t, in formula, Y (m p, k p) be position (m p, k p) frequency domain located receives signal, [] tfor transposition symbol.
Step D, by described A pand Y pthe time domain channel response that computing obtains h ^ m p = ( A P H A P ) - 1 A P H Y P , In formula, () -1for getting contrary symbol, () hfor conjugate transpose symbol.
Step e, by time domain channel response the form of being write as each footpath separation is: h ^ m p = [ h ^ ( m p , 0 ) , . . . , h ^ ( m p , L - 1 ) ] , In formula, the estimated value of time domain channel the first footpath coefficient, it is the estimated value of time domain channel L-1 footpath coefficient.
Step F, from time domain channel estimated value middle extraction is with the footpath of LOS component, i.e. the first footpath: h ^ los ( m p ) = h ^ ( m p , 0 ) .
Step G, the footpath with LOS component that utilizes estimation to obtain carry out frequency deviation estimation, obtain Doppler frequency deviation estimated value:
ϵ ^ = N 2 π N s DN m Σ m p ∈ κ p arg { h ^ los ( m p + D ) h ^ los * ( m p ) }
In formula, κ pfor the OFDM/OFDMA character position set with pilot tone, m pfor the sign of the OFDM/OFDMA symbol with pilot tone, N s=N+N g, N and N gbe respectively the length of FFT and Cyclic Prefix, arg{} represents to get angle symbol, and D is two symbolic numbers of carrying out the OFDM/OFDMA mark space of relevant treatment, N mfor carrying out the OFDM/OFDMA group number of relevant treatment, for position m p+ D place is with the footpath coefficient estimated value of LOS component, represent conjugate.
Step H, estimates frequency deviation ρ doubly as compensate of frequency deviation value be used for compensating reception signal, ρ ∈ [01] wherein, calculates the average interference power P that receives signal after compensate of frequency deviation iCI:
P ICI = P L N u N 2 Σ p = - N u + 1 N u - 1 ( N u - | p | ) | sin π ( p + ϵ ‾ ) sin [ π ( p + ϵ ‾ ) / N ] | 2 - P L N 2 | sin π ϵ ‾ sin [ π ϵ ‾ / N ] | 2 + N u - 1 N P S
+ 2 P S N u N 2 Σ q = 1 N - 1 ( N - q ) J 0 ( 2 π f max T s q ) [ Σ p = - N u + 1 N u - 1 ( N u - | p | ) cos ( 2 πq ( p - ϵ ^ ) N ) - N u cos ( - 2 πq ϵ ^ N ) ]
In formula, what N was FFT counts, f maxfor maximum doppler frequency, T sfor the sampling interval, N ufor the number of useful subcarrier in OFDM/OFDMA symbol, J 0() is first kind zero Bessel function, ε is true normalization frequency deviation, P lfor the power of Rice channel LOS component, P l=| c| 2, c is channel the first footpath coefficient, P sfor the power of all scattering components of Rice channel and, for channel l pthe coefficient in footpath, var[] for getting variance symbol.
Step I, according to average interference power P iCIafter calculating compensate of frequency deviation, receiving signal signal interference ratio SIR is:
SIR = [ P L | sin π ϵ ‾ sin π ϵ ‾ / N | 2 + P S ( N + 2 Σ r = 1 N - 1 J 0 ( 2 π f max r T s ) cos ( 2 πρ ϵ ^ r / N ) ) ] / P ICI
Step J, determines best compensate of frequency deviation multiple ρ according to the size of the signal interference ratio SIR of signal after compensation, and compensate of frequency deviation multiple ρ corresponding when soon SIR gets maximum is as best compensation multiple.
Effect of the present invention can further illustrate by following emulation:
1. simulated conditions
The performance that has adopted the analogue system based on LTE-OFDMA transmission standard to test frequency deviation estimation with the compensation method of the present invention's proposition in emulation, its design parameter is: systematic sampling frequency is 15.36MHz, carrier frequency is considered 2.3GHZ, train speed is considered 350km/h, subcarrier number N=1024, the useful subcarrier number of system N used=600, circulating prefix-length is N cp=N/8, data acquisition 16QAM modulation system.In emulation, adopt ITU-VAx.y channel model, wherein x represents vehicular speeds, and y represents Rice factor, Rice factor K=5 in emulation, 10.
2. emulation content and result
Emulation 1: according to simulated conditions, contrast VAx channel is frequency deviation estimating method of the present invention and existing methodical performance difference in different Rice factor situations, and simulation result as shown in Figure 5.From the simulation curve of Fig. 5, can find out: estimated accuracy of the present invention is higher than the estimated accuracy of prior art, and along with the increase of Rice factor, frequency offset estimation accuracy further improves.
Emulation 2: according to simulated conditions, contrast the difference that receives SIR in signal in different compensate of frequency deviation multiple ρ situations, simulation result as shown in Figure 6.From the simulation curve of Fig. 6, can find out: in ρ≤0.5 o'clock, along with the increase of compensation multiple ρ, the SIR that receives signal also increases, but when ρ > 0.5, the SIR that receives signal will reduce along with compensating the increase of multiple ρ.The reason of this tendency of analysis chart 6 curve is because inherent spurious frequency deviation value larger on scattering component will cause the variation of scattering component more violent, thereby make to receive interference power in signal, becomes large, causes SIR to decline.Therefore, in this system, best compensate of frequency deviation multiple is 0.5.

Claims (6)

1. under high-speed mobile environment, the Doppler frequency deviation of Rice channel is estimated and a compensation method, comprising:
(1) frequency deviation estimating step:
(1a) extract the footpath with direct projection component LOS
First, extract frequency domain and receive the OFDM/OFDM OFDM/OFDMA frequency pilot sign in signal, and utilize frequency pilot sign and the frequency pilot sign in transmitted signal in this reception signal to carry out channel estimating, obtain pilot frequency locations domain channel response: m wherein pfor the sign with frequency pilot sign, k pfor the sign of pilot sub-carrier position, Y (m p, k p) be position (m p, k p) frequency domain located receives signal, X (m p, k p) be position (m p, k p) the frequency domain transmitted signal located;
Secondly, to pilot frequency locations domain channel response carry out interpolation, obtain the channel frequency domain response on other data subcarriers on whole frequency pilot sign and the current upper domain channel response estimated value with frequency pilot sign is designated as k wherein dfor the sign of data subcarrier position, k is the sign of the sub-carrier positions that takies of system;
Again, to domain channel response carry out IFFT conversion process, obtain time domain channel estimated value in formula, the estimated value of time domain channel the first footpath coefficient, the estimated value of time domain channel L-1 footpath coefficient, for the mean value of whole OFDM/OFDMA symbol upper signal channel response, and be positioned at the centre position of this OFDM/OFDMA symbol;
Finally, from time domain channel estimated value middle extraction is with the footpath of LOS component, i.e. the first footpath: h ^ los ( m p ) = h ^ ( m p , 0 ) ;
(1b) footpath with LOS component that utilizes estimation to obtain carry out frequency deviation estimation, obtain Doppler frequency deviation estimated value:
ϵ ^ = N 2 π N s DN m Σ m p ∈ κ p arg { h ^ los ( m p + D ) h ^ los * ( m p ) }
In formula, κ pfor the OFDM/OFDMA character position set with pilot tone, m pfor the sign of the OFDM/OFDMA symbol with pilot tone, N s=N+N g, N and N gbe respectively the length of FFT and Cyclic Prefix, arg{} represents to get angle symbol, and D is two symbolic numbers of carrying out the OFDM/OFDMA mark space of relevant treatment, N mfor carrying out the OFDM/OFDMA group number of relevant treatment, for position m p+ D place is with the footpath coefficient estimated value of LOS component, represent conjugate;
(2) best Doppler frequency deviation compensation process:
(2a) the signal interference ratio SIR of signal after calculating compensate of frequency deviation
Suppose for the frequency deviation value of compensate of frequency deviation to be wherein ρ ∈ [0 1] is compensation multiple, for the estimated value of frequency deviation, after compensate of frequency deviation, the signal interference ratio SIR of signal is:
SIR = [ P L | sin π ϵ ‾ sin π ϵ ‾ / N | 2 + P S ( N + 2 Σ r = 1 N - 1 J 0 ( 2 π f max r T s ) cos ( 2 πρ ϵ ^ r / N ) ) ] / P ICI
In formula, f maxfor maximum doppler frequency, T sfor the sampling interval, J 0() is first kind zero Bessel function, ε is true normalization frequency deviation, P lfor the power of Rice channel LOS component, P l=| c| 2, c is channel the first footpath LOS component coefficient, P sfor the power of all scattering components of Rice channel and, for channel l pthe coefficient in footpath, var[] for getting variance symbol, P iCIfor average interference power:
P ICI = P L N u N 2 Σ p = - N u + 1 N u - 1 ( N u - | p | ) | sin π ( p + ϵ ‾ ) sin [ π ( p + ϵ ‾ ) / N ] | 2 - P L N 2 | sin π ϵ ‾ sin [ π ϵ ‾ / N ] | 2 + N u - 1 N P S + 2 P S N u N 2 Σ q = 1 N - 1 ( N - q ) J 0 ( 2 π f max T s q ) [ Σ p = - N u + 1 N u - 1 ( N u - | p | ) cos ( 2 πq ( p - ϵ ^ ) N ) - N u cos ( - 2 πq ϵ ^ N ) ]
In formula, N unumber for useful subcarrier in OFDM/OFDMA symbol;
(2b) according to the size of the signal interference ratio SIR of signal after compensation, determine best compensate of frequency deviation multiple ρ, compensate of frequency deviation multiple ρ corresponding when soon SIR gets maximum is as best compensation multiple.
2. frequency deviation estimating method according to claim 1, wherein step (1a) described to pilot frequency locations domain channel response carry out interpolation, refer to any one in linear interpolation, Gaussian interpolation and Cubic interpolation.
3. frequency deviation estimating method according to claim 2, wherein said linear interpolation is to be undertaken by following formula:
H ^ ( m p , k d ) = H ^ ( m p , k p + m ) = ( 1 - m N f ) H ^ p ( m p , k p ) + m N f H ^ p ( m p , k p + N f )
In formula, data subcarrier position (the m that interpolation draws p, k d) domain channel response, m pfor the sign with frequency pilot sign, k pfor the sign of pilot sub-carrier position, k dfor the sign of data subcarrier position, N fbe the sub-carrier number at two adjacent pilot sub-carrier intervals, m gets different value makes interpolation result traverse all data subcarrier positions, and m ∈ [1 N f-1], for position (m p, k p) channel estimation value located, position (the m obtaining for interpolation p, k p+ m) channel estimation value of locating, for position (m p, k p+ N f) channel estimation value located.
4. frequency deviation estimating method according to claim 2, wherein said Gaussian interpolation is to be undertaken by following formula:
H ^ ( m p , k d ) = H ^ ( m p , k p + m ) = 1 2 [ ( m N f ) 2 - m N f ] H ^ p ( m p , k p - N f ) + [ 1 - ( m N f ) 2 ] H ^ p ( m p , k p ) + 1 2 [ ( m N f ) 2 + m N f ] H ^ p ( m p , k p + N f )
In formula, data subcarrier position (the m that interpolation draws p, k d) domain channel response, m pfor the sign with frequency pilot sign, k pfor the sign of pilot sub-carrier position, k dfor the sign of data subcarrier position, N ffor the carrier number at two adjacent pilot sub-carrier intervals, m gets different value makes interpolation result travel through all data subcarrier positions, and m ∈ [1 N f-1], position (the m obtaining for interpolation p, k p+ m) channel estimation value of locating, for position (m p, k p) channel estimation value located, for position (m p, k p+ N f) channel estimation value located, for position (m p, k p-N f) channel estimation value located.
5. frequency deviation estimating method according to claim 2, wherein said Cubic interpolation is to be undertaken by following formula:
H ^ ( m p , k d ) = H ^ ( m p , k p + m ) = ( 1 + 2 m N f ) ( m - N f - N f ) 2 H ^ p ( m p , k p ) + ( 1 + 2 ( m - N f ) - N f ) ( m N f ) 2 H ^ p ( m p , k p + N f ) + m ( m - N f - N f ) 2 H ^ p * ( m p , k p ) + ( m - N f ) ( m N f ) 2 H ^ p * ( m p , k p + N f )
In formula, data subcarrier position (the m that interpolation draws p, k d) domain channel response, m pfor the sign with frequency pilot sign, k pfor the sign of pilot sub-carrier position, k dfor the sign of data subcarrier position, N ffor the carrier number at two adjacent pilot sub-carrier intervals, m gets different value makes interpolation result traverse all data subcarrier positions, and m ∈ [1 N f-1], for position (m p, k p) channel estimation value located, position (the m obtaining for interpolation p, k p+ m) channel estimation value of locating, for position (m p, k p+ Nf) channel estimation value of locating, represent conjugation, represent conjugation.
6. under high-speed mobile environment, the Doppler frequency deviation of Rice channel is estimated and a compensation method, comprising:
1) frequency deviation estimating step:
1a) the OFDM/OFDMA time-domain signal receiving is adopted to time domain channel algorithm for estimating, obtains time domain channel estimated value:
1a1) build frequency domain sending metrix: X p=diag (X (m p, k 1), X (m p, k 2) ..., X (m p, k p)), in formula, m pfor the sign with frequency pilot sign, k pfor the sign of pilot sub-carrier position, X (m p, k p) be position (m p, k p) the frequency domain transmitted signal located, diag() represent diagonal matrix symbol;
1a2) by frequency domain sending metrix X pbuild intermediate data matrix: A p=X pbF km, in formula, F kfor N * N unit discrete time Fourier transform matrix, the length that N is Fourier transform, M is N * Q mapping matrix, M = E F , E is Q * Q unit matrix, the maximum multipath time delay length that Q is time domain channel response, and F is (N-Q) * Q null matrix, and B is P * N matrix, and its every a line only has a nonzero value 1, and the capable nonzero value 1 of p is positioned at k prow, P is the number of each symbol pilot sub-carrier;
1a3) build frequency domain receiving matrix: Y p=[Y (m p, k 1), Y (m p, k 2) ..., Y (m p, k p)] t, in formula, Y (m p, k p) be position (m p, k p) frequency domain located receives signal, [] tfor transposition symbol;
1a4) by A pand Y pthe time domain channel response that computing obtains in formula, () -1for getting contrary symbol, () hfor conjugate transpose symbol;
1a5) by time domain channel response the form of being write as each footpath separation is:
h ^ m p = [ h ^ ( m p , 0 ) , . . . , h ^ ( m p , L - 1 ) ]
In formula, the estimated value of time domain channel the first footpath coefficient, the estimated value of time domain channel L-1 footpath coefficient, for the mean value of whole OFDM/ODMA symbol upper signal channel response, and be positioned at the centre position of this OFDM/OFDMA symbol, m psign for the OFDM/OFDMA symbol with pilot tone;
1b) from time domain channel estimated value middle extraction is with the footpath of LOS component, i.e. the first footpath: h ^ los ( m p ) = h ^ ( m p , 0 ) ;
The footpath with LOS component that 1c) utilizes estimation to obtain carry out frequency deviation estimation, obtain Doppler frequency deviation estimated value:
ϵ ^ = N 2 π N s D N m Σ m p ∈ κ p arg { h ^ los ( m p + D ) h ^ los * ( m p ) }
In formula, κ pfor the OFDM/OFDMA character position set with pilot tone, m pfor the sign of the OFDM/OFDMA symbol with pilot tone, N s=N+N g, N and N gbe respectively the length of FFT and Cyclic Prefix, arg{} represents to get angle symbol, and D is two symbolic numbers of carrying out the OFDM/OFDMA mark space of relevant treatment, N mfor carrying out the OFDM/OFDMA group number of relevant treatment, for position m p+ D place is with the footpath coefficient estimated value of LOS component, represent conjugation;
2) best Doppler frequency deviation compensation process:
2a) the signal interference ratio SIR of signal after calculating compensate of frequency deviation
Suppose for the frequency deviation value of compensate of frequency deviation to be wherein ρ ∈ [0 1] is compensation multiple, for the estimated value of frequency deviation, after compensate of frequency deviation, the signal interference ratio SIR of signal is:
SIR = [ P L | sin π ϵ ‾ sin π ϵ ‾ / N | 2 + P S ( N + 2 Σ r = 1 N - 1 J 0 ( 2 π f max r T s ) cos ( 2 πρ ϵ ^ r / N ) ) ] / P ICI
In formula, f maxfor maximum doppler frequency, T sfor the sampling interval, J 0() is first kind zero Bessel function, ε is true normalization frequency deviation, P lfor the power of Rice channel LOS component, P l=| c| 2, c is channel the first footpath LOS component coefficient, P sfor the power of all scattering components of Rice channel and, for channel l pthe coefficient in footpath, var[] for getting variance symbol, P iCIfor average interference power:
P ICI = P L N u N 2 Σ p = - N u + 1 N u - 1 ( N u - | p | ) | sin π ( p + ϵ ‾ ) sin [ π ( p + ϵ ‾ ) / N ] | 2 - P L N 2 | sin π ϵ ‾ sin [ π ϵ ‾ / N ] | 2 + N u - 1 N P S + 2 P S N u N 2 Σ q = 1 N - 1 ( N - q ) J 0 ( 2 π f max T s q ) [ Σ p = - N u + 1 N u - 1 ( N u - | p | ) cos ( 2 πq ( p - ϵ ^ ) N ) - N u cos ( - 2 πq ϵ ^ N ) ]
In formula, N unumber for useful subcarrier in OFDM/OFDMA symbol;
2b) according to the size of the signal interference ratio SIR of signal after compensation, determine best compensate of frequency deviation multiple ρ, compensate of frequency deviation multiple ρ corresponding when soon SIR gets maximum is as best compensation multiple.
CN201110366374.1A 2011-11-17 2011-11-17 Method for estimating and compensating doppler frequency offset in Rician channels in high-speed mobile environment Expired - Fee Related CN102404268B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201110366374.1A CN102404268B (en) 2011-11-17 2011-11-17 Method for estimating and compensating doppler frequency offset in Rician channels in high-speed mobile environment

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201110366374.1A CN102404268B (en) 2011-11-17 2011-11-17 Method for estimating and compensating doppler frequency offset in Rician channels in high-speed mobile environment

Publications (2)

Publication Number Publication Date
CN102404268A CN102404268A (en) 2012-04-04
CN102404268B true CN102404268B (en) 2014-02-12

Family

ID=45886065

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201110366374.1A Expired - Fee Related CN102404268B (en) 2011-11-17 2011-11-17 Method for estimating and compensating doppler frequency offset in Rician channels in high-speed mobile environment

Country Status (1)

Country Link
CN (1) CN102404268B (en)

Families Citing this family (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103428147B (en) * 2012-05-24 2016-12-14 普天信息技术研究院有限公司 A kind of method that TDD-LTE system frequency deviation compensates
CN102739298B (en) * 2012-06-13 2015-01-28 西安电子科技大学 Mobile relay receiving method and device under multi-radio-remote-unit (RRU) scene of high-speed railway
CN104022976A (en) * 2014-06-16 2014-09-03 中国科学院计算技术研究所 Doppler frequency shift estimation method and system for high speed railway LTE system
CN104849546B (en) * 2015-05-05 2017-06-23 大连理工大学 A kind of aircarrier aircraft scattered signal doppler frequency estimation method
CN105007241A (en) * 2015-07-01 2015-10-28 杭州祥声通讯股份有限公司 Doppler frequency offset estimation method and system under high-speed rail environment
CN105119852A (en) * 2015-07-10 2015-12-02 广东顺德中山大学卡内基梅隆大学国际联合研究院 Estimation method based on broadband OFDM system frequency-dependent Doppler shift
CN105007244B (en) * 2015-07-16 2018-07-03 上海交通大学 Inter-carrier interference elimination method under high-speed mobile environment based on geographical location information
CN105306393B (en) * 2015-11-18 2019-04-05 电子科技大学 A kind of Rice channel method in multiple antenna and carrier system
CN106817333B (en) * 2017-01-05 2019-10-11 西安电子科技大学 High dynamic carrier synchronization method based on open-loop acquisition and Closed loop track
CN107465637A (en) * 2017-09-11 2017-12-12 山东财经大学 It is a kind of to be applied to OFDM channel estimation method and device based on pilot tone in smart city
CN109495407B (en) * 2017-09-13 2020-09-11 电信科学技术研究院 Channel estimation method, device, equipment and computer readable storage medium
CN110602015B (en) * 2019-09-12 2020-07-03 北京邮电大学 Doppler frequency offset compensation and signal sending method and device in OFDM system

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101588338A (en) * 2009-04-15 2009-11-25 山东大学 OFDM carrier frequency offset estimation method suitable for packet transmission
CN101814931A (en) * 2009-02-19 2010-08-25 中兴通讯股份有限公司 Doppler frequency shift estimation and compensation method in TD-SCDMA (Time Division-Synchronization Code Division Multiple Access) system
CN102025680A (en) * 2010-12-15 2011-04-20 华为技术有限公司 Speed estimating method and speed estimating device

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR100800692B1 (en) * 2005-04-11 2008-02-01 삼성전자주식회사 Apparatus for frequency offset estimation in orthogonal frequency division multiplexing systems and method of the same

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101814931A (en) * 2009-02-19 2010-08-25 中兴通讯股份有限公司 Doppler frequency shift estimation and compensation method in TD-SCDMA (Time Division-Synchronization Code Division Multiple Access) system
CN101588338A (en) * 2009-04-15 2009-11-25 山东大学 OFDM carrier frequency offset estimation method suitable for packet transmission
CN102025680A (en) * 2010-12-15 2011-04-20 华为技术有限公司 Speed estimating method and speed estimating device

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
《OFDMA测距中一种新的定时偏移估计算法》;倪浩 等;《西安电子科技大学学报》;20101031(第5期);第783-788页 *
倪浩 等.《OFDMA测距中一种新的定时偏移估计算法》.《西安电子科技大学学报》.2010,(第5期),

Also Published As

Publication number Publication date
CN102404268A (en) 2012-04-04

Similar Documents

Publication Publication Date Title
CN102404268B (en) Method for estimating and compensating doppler frequency offset in Rician channels in high-speed mobile environment
US8064328B2 (en) Channel estimation device
CN102387115B (en) OFDM pilot scheme design and channel estimation method
CN103107969B (en) Incremental iterative time-varying channel evaluation and inter carrier interference (ICI) elimination method of fast orthogonal frequency division multiplexing (OFDM) system
CN101371546B (en) Method and device for estimating channel of uplink signal in wireless communication system
US20090279421A1 (en) Apparatus and methods for estimating and compensating sampling clock offset
CN103051578B (en) With the iteration error propagation judgement OFDM channel estimation method that ICI eliminates
US20060017613A1 (en) High doppler channel estimation for OFD multiple antenna systems
CN103873422A (en) Method for eliminating multipath interference in system symbol in underwater sound orthogonal frequency-division multiplexing system
CN102291363A (en) Channel estimation and data detection method for OFDM (Orthogonal Frequency Division Multiplexing) system
CN102006249B (en) Channel estimation method in cooperative orthogonal frequency division multiplexing system
CN110493156B (en) Frequency offset estimation method based on constellation point diversity in 5G mobile communication system
CN105187352A (en) Integer frequency offset estimation method based on OFDM preamble
CN102833194A (en) OFDM (orthogonal frequency division multiplexing) channel estimation method based on symmetrical basis expansion model for quick time-varying channel
CN101741775B (en) Taylor expansion-based single-frequency OFDM time-varying channel estimation method
CN103095638B (en) The blind evaluation method of the sampling frequency deviation of ofdm system under a kind of multidiameter fading channel
CN104735014B (en) A kind of time synchronization method related based on leading symbol difference
CN104836770B (en) It is a kind of based on related average and adding window timing estimation method
CN102113285A (en) A simplified equalizationscheme for distributed resource allocation in multi-carrier systems
US8320481B2 (en) Synchronization method and apparatus for orthogonal frequency division multiplexing system
US10334605B2 (en) Time domain pilot of single-carrier MIMO system and synchronization method thereof
CN104253772A (en) Channel estimation method for orthogonal frequency division multiplexing system
CN102625448B (en) Improved TDOA positioning method for LTE system
CN102801662B (en) Superimposed-pilot-based channel estimation method and device for multi-band ultra-wideband system
CN102238128B (en) Method for self-elimination of inter-subcarrier interference of differential OFDM (orthogonal frequency division multiplexing) system

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant
CF01 Termination of patent right due to non-payment of annual fee
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20140212

Termination date: 20191117