CN103179058B - The method of estimation of channel impulse response length and device - Google Patents

Abstract

The embodiment of the invention discloses a kind of method of estimation and device of channel impulse response length, the frequency domain data after synchronous and compensate of frequency deviation through the least square criterion LS channel estimating at pilot tone place, is obtained the channel frequency domain response at pilot tone place by the present embodiment; The channel frequency domain response at described pilot tone place is obtained the channel frequency domain response at data place through interpolation; According to the channel frequency domain response at described pilot tone place and time domain channel response corresponding to the channel frequency domain response at described data place, estimate the channel impulse response length of described channel frequency domain response.Be not suitable for problem in channel circumstance aviation system of broadband wireless communication complicated and changeable in order to solve existing channel estimation methods, can channel estimating performance be improved.

Description

Method and device for estimating channel impulse response length

Technical Field

The embodiment of the invention relates to the field of aviation broadband wireless communication systems, in particular to a method and a device for estimating the length of a channel impulse response.

Background

The rapid development of civil aviation communication services, particularly the rapid growth of ground-air communication services, promotes the development of civil aviation communication networks from narrow bands to wide bands. Orthogonal Frequency Division Multiplexing (OFDM) technology has become a key technology of an aviation broadband wireless communication system due to its characteristics of high transmission rate, high spectrum utilization rate, strong anti-multipath interference capability, and the like. The OFDM technology is particularly sensitive to frequency offset, and the accuracy of channel estimation directly affects the performance of the OFDM technology, so that the channel estimation method is a key for realizing the application of the OFDM technology in an aviation broadband wireless communication system.

Generally, channel estimation methods include a Least Square (LS) channel estimation method, which does not consider noise in a received signal and is therefore greatly affected by white gaussian noise, and an improved LS channel estimation method, which has poor channel estimation performance particularly when the signal-to-noise ratio is low; for the improved LS channel estimation method, when the moving speed of the airplane is increased and the Doppler frequency shift is increased, the channel estimation performance is rapidly reduced; in addition, in an aviation broadband wireless communication system, the channel environment is complex and changeable, including the scenes of takeoff, landing, runway taxiing, parking and the like of an airplane, the channel impulse response length under different scenes is continuously changed, and the channel impulse response length in the improved LS channel estimation method is fixed, so the channel estimation performance is poor.

Disclosure of Invention

The embodiment of the invention provides a method and a device for estimating channel impulse response length, which are applied to a channel estimation scene of a civil aviation air-ground broadband wireless communication system, are used for solving the problem that the existing channel estimation method is not suitable for the aviation broadband wireless communication system with complex and changeable channel environment, and can improve the channel estimation performance.

In a first aspect, the present invention provides a method for estimating a channel impulse response length, which is applied in a channel estimation scenario of a civil aviation air-ground broadband wireless communication system, and includes:

performing Least Square (LS) channel estimation on the frequency domain data after synchronization and frequency offset compensation through a pilot frequency position to obtain channel frequency domain response of the pilot frequency position;

interpolating the channel frequency domain response at the pilot frequency position to obtain the channel frequency domain response at the data position;

and estimating the channel impulse response length of the channel frequency domain response according to the channel frequency domain response at the pilot frequency and the time domain channel response corresponding to the channel frequency domain response at the data.

In a second aspect, the present invention provides a channel impulse response length estimation apparatus, applied to an aviation broadband wireless communication system, including:

the first processing module is used for carrying out Least Square (LS) channel estimation on the frequency domain data after synchronization and frequency offset compensation through a pilot frequency position to obtain channel frequency domain response of the pilot frequency position;

the second processing module is used for interpolating the channel frequency domain response at the pilot frequency position to obtain the channel frequency domain response at the data position;

and the third processing module is used for estimating the channel impulse response length of the channel frequency domain response according to the channel frequency domain response at the pilot frequency and the time domain channel response corresponding to the channel frequency domain response at the data.

The embodiment of the invention obtains the channel frequency domain response of the pilot frequency position by carrying out LS channel estimation on the frequency domain data after synchronization and frequency offset compensation through the least square criterion of the pilot frequency position; interpolating the channel frequency domain response at the pilot frequency position to obtain the channel frequency domain response at the data position; and estimating the channel impulse response length of the channel frequency domain response according to the channel frequency domain response at the pilot frequency and the time domain channel response corresponding to the channel frequency domain response at the data. Thereby enabling: in a scene that the channel environment is changeable in the airborne mobile communication system, the embodiment can estimate the channel impulse response length according to the channel frequency domain response at the pilot frequency and the time domain channel response corresponding to the channel frequency domain response at the data, and the channel impulse response length changes in real time along with the change of the channel environment, so that the method can be applied to the airborne broadband wireless communication system with complicated and changeable channel environment and can improve the channel estimation performance.

Drawings

Fig. 1 is a flowchart illustrating a method for estimating a channel impulse response length according to an embodiment of the present invention;

fig. 2 is a schematic flow chart of a channel estimation method of a civil aviation air-ground broadband wireless communication system applied in the embodiment of the invention;

fig. 3 is a schematic structural diagram of an apparatus for estimating a channel impulse response length according to another embodiment of the present invention.

Detailed Description

In order to ensure that the aviation broadband OFDM mobile communication system can well suppress noise in different channel environments, it is necessary to dynamically track changes in the channel impulse response length. The invention provides a method for estimating the channel impulse response length, which can adaptively estimate the size of the channel impulse response length in real time, thereby ensuring the channel estimation performance under the complicated and changeable aviation broadband channel condition.

Fig. 1 is a flowchart illustrating a method for estimating a channel impulse response length according to an embodiment of the present invention, as shown in fig. 1:

101. and performing Least Square (LS) channel estimation on the frequency domain data after synchronization and frequency offset compensation through a Least Square (LS) criterion at the pilot frequency position to obtain channel frequency domain response at the pilot frequency position.

After receiving data, a receiver needs to synchronize first, and after synchronization is successful, further frequency offset estimation and compensation are generally needed to be performed on signals. For example, when the base station transmits data to the user terminal, the receiver is a receiver on the user terminal side, and when the user terminal transmits data to the base station, the receiver is a receiver on the base station side.

Assuming that the frequency domain data matrix at the pilot frequency position after synchronization and frequency offset compensation is Y (diagonal matrix), the frequency domain data matrix at the pilot frequency position of the transmitting end stored locally is X (diagonal matrix), H_LSUsing the least squares criterion for the channel frequency domain response truth matrix at the pilot under the LS criterion, we obtain:

minf(H_LS)=min{(Y-XH_LS)^H(Y-XH_LS)}

from the calculus theorem, the function minf (H) is_LS) Taking the minimum value, it is necessary to make the function pair H_LSHas a partial derivative of 0, i.e.Obtaining an estimated value matrix of channel frequency domain response at the pilot frequency under the LS criterion:

102. and interpolating the channel frequency domain response at the pilot frequency to obtain the channel frequency domain response at the data.

It should be noted that the interpolation algorithm may be the prior art, and generally, the algorithm used for interpolation may select different interpolation algorithms according to the system requirements and the pilot pattern used by the system, and it is assumed that the present embodiment uses a first-order linear interpolation algorithm, where the first-order linear interpolation algorithm uses two adjacent pilot symbols to perform linear interpolation.

Assuming that the interpolation in the time axis direction of the two-dimensional pilot pattern is performed first, the interpolation formula is:

$H(k,i)=(1-\frac{p}{{\mathrm{\Δ}}_t})H(k,q{\mathrm{\Δ}}_t)+\frac{p}{{\mathrm{\Δ}}_t}H(k,(q+1){\mathrm{\Δ}}_t);$

wherein k represents the kth subcarrier, i represents the ith OFDM symbol, p and q are intermediate variables, have no specific physical significance and are convenient for expression of a formula, and p is more than or equal to 1 and is less than or equal to delta_t-1，qΔ_t<i<(q+1)Δ_t,q=0,1,2....；Δ_tThe distance between two adjacent pilots on the time axis.

If linear interpolation is needed in the direction of the frequency axis, the interpolation formula is as follows:

$H(k,i)=(1-\frac{p}{{\mathrm{\&Delta;}}_f})H(q{\mathrm{\&Delta;}}_f,i)+\frac{p}{{\mathrm{\&Delta;}}_f}H((q+1){\mathrm{\&Delta;}}_f,i)$

wherein k represents the kth subcarrier, i represents the ith OFDM symbol, p and q are intermediate variables, have no specific physical significance and are convenient for expression of a formula, and q is delta_f<k<(q+1)Δ_f,1≤p≤Δ_f-1,Δ_fIs the spacing between two adjacent pilots on the frequency axis.

103. And estimating the channel impulse response length of the channel frequency domain response according to the channel frequency domain response at the pilot frequency and the time domain channel response corresponding to the channel frequency domain response at the data.

Assuming that, under an ideal condition, the channel impulse response coefficient is within the channel impulse response length, the gaussian white noise is uniformly distributed in all the channel samples, and the true value of the channel time domain impulse response of the ith OFDM symbol on the kth subcarrier is:

$h_{k,i}=\left\{\begin{array}{c}{h}_{k,i}^L,0\&le;k<L\\ 0,L\&le;k<N\end{array}\right.-----\left(A\right)$

step 103 may specifically be:

1031. the cumulative mean square error function of the channel impulse response is derived from the definition of the cumulative mean square error.

Wherein the cumulative mean square error function is:

wherein,is the estimated value of the channel time domain impulse response of the ith OFDM symbol on the kth subcarrier, and contains noise components.

The cumulative mean square error function can further be derived from equation (a): $e\left(L\right)=E\left[\underset{k=0}{\overset{N-1}{\mathrm{\&Sigma;}}}{|{\hat{h}}_{k,i}-h_{k,i}|}^2\right]=\underset{k=L}{\overset{N-1}{\mathrm{\&Sigma;}}}E\left[{\left|{\hat{h}}_{k,i}\right|}^2\right]+L\&CenterDot;{\mathrm{\&sigma;}}_{\mathrm{noise}}^2,$ the function is a concave function and has a minimum value, and the value L corresponding to the minimum value of the cumulative mean square error function e (L) is determined as the channel impulse response length of the ith OFDM symbol;

wherein,for estimated noise error, h_k,iIs the channel time domain response true value at the data,the estimated value of the time domain response of the channel at the data position is L, the length of the impulse response of the channel is expressed, N represents the total number of subcarriers, k is more than or equal to 0 and less than or equal to N, M represents the number of OFDM symbols of the orthogonal frequency division multiplexing technology, i represents the ith OFDM symbol of the orthogonal frequency division multiplexing technology, and i is more than or equal to 1 and less than or equal to M;

therefore, in a scene that the channel environment is changeable in the airborne mobile communication system, even if the channel frequency domain response at the pilot frequency and the channel frequency domain response at the data change with the change of the channel environment, the embodiment can change the estimated channel impulse response length in real time according to the time domain channel response corresponding to the channel frequency domain response at the pilot frequency and the channel frequency domain response at the data along with the change of the channel environment.

1032. And taking the maximum value of the channel impulse response length of the M OFDM symbols as the channel impulse response length of the channel frequency domain response.

In this embodiment, the length of the channel impulse response of M OFDM symbols is maximized, so that the more the number of zero forcing elements included in the time domain channel response is, the more the noise is removed, and the more accurate the channel estimation is.

Fig. 2 is a schematic flow chart of a channel estimation method of a civil aviation air-ground broadband wireless communication system applied in the embodiment of the present invention, as shown in fig. 2:

and performing Least Square (LS) channel estimation on the frequency domain data after synchronization and frequency offset compensation through a Least Square (LS) criterion at the pilot frequency position to obtain channel frequency domain response at the pilot frequency position. And interpolating the channel frequency domain response at the pilot frequency to obtain the channel frequency domain response at the data.

Assuming that, under an ideal condition, the channel impulse response coefficient is within the channel impulse response length, and the gaussian white noise is uniformly distributed in all the channel samples, the channel frequency domain impulse response of the ith OFDM symbol on the kth subcarrier is:

$H\left(k\right)=\frac{1}{\sqrt{N}}\underset{n=0}{\overset{L-1}{\mathrm{\&Sigma;}}}h\left(n\right)\&CenterDot;e^{-j\frac{2\mathrm{\&pi;nk}}{N}},$ j mathematically represents the imaginary part of the complex number;

rewriting the formula into $H(k,i)=\frac{1}{\sqrt{N}}\underset{n=0}{\overset{L-1}{\mathrm{\&Sigma;}}}h\left(n\right)\&CenterDot;e^{-j\frac{2\mathrm{\&pi;nk}}{N}};$

Wherein N represents the total number of subcarriers, and k is more than or equal to 0 and less than or equal to N; l denotes a channel impulse response length.

The frequency domain response vector of each OFDM symbol isWherein the matrix F is a Fourier transform matrix, N_DFor the number of useful sub-carriers, N is not less than N_DWhen N includes useful subcarriers and virtual subcarriers, N is greater than N_D,. When only useful subcarriers are contained in N, N_DAnd N are both equal.

Carrying out QR decomposition on the matrix F to obtain a matrix:

wherein,is a normalized matrix of the number of bits in the matrix,is an upper triangular matrix, where N_DThe elements of the L rows are zero.

Carrying out QR decomposition on the channel frequency domain response H of the data obtained by interpolation in the step 102 to obtainTo: further written as:wherein, F is a Fourier transform matrix, and H is a time domain response matrix corresponding to the frequency domain response H.

In mathematical theory, there is H = Fh, but in practical applications, there is noise in the channel estimation, and therefore, it is necessary to superimpose the noise(ii) a Where F = QR is a mathematical matrix decomposition method, called QR decomposition of a matrix.

The transposition of channel frequency domain response H multiplied by Q, namely the left and right sides of the expression are respectively multiplied by Q^HTo obtain a vector $\hat{g}:Q^H{\hat{H}}_{\mathrm{LS}}=Q^H(\mathrm{QRh}+\hat{W})=\mathrm{Rh}+Q^H\hat{W};$

Since R is an upper triangular matrix, there is N_D-L lines of zero elements, it can be seen that the first item on the right side has only the first L elements, the last N_DThe L elements are zero and the second term on the right of the equation is the noise term.

Suppose thatWherein the first L elements mainly comprise channel impulse response and the last N elements_D-L elements contain mainlyNoise. Therefore, the first L elements are useful data needed by the user, and in order to further reduce the influence of noise on the estimation performance, the last N elements can be used_DL noise elements zero forcing yields:

the final channel frequency response is:

the embodiment of the invention obtains the channel frequency domain response of the pilot frequency position by carrying out LS channel estimation on the frequency domain data after synchronization and frequency offset compensation through the least square criterion of the pilot frequency position; interpolating the channel frequency domain response at the pilot frequency position to obtain the channel frequency domain response at the data position; and estimating the channel impulse response length of the channel frequency domain response according to the channel frequency domain response at the pilot frequency and the time domain channel response corresponding to the channel frequency domain response at the data. Thereby enabling: in a scene that the channel environment is changeable in the airborne mobile communication system, the embodiment can estimate the channel impulse response length according to the channel frequency domain response at the pilot frequency and the time domain channel response corresponding to the channel frequency domain response at the data, and the channel impulse response length changes in real time along with the change of the channel environment, so that the method can be applied to the airborne broadband wireless communication system with complicated and changeable channel environment and can improve the channel estimation performance.

Fig. 3 is a schematic structural diagram of an apparatus for estimating a channel impulse response length according to another embodiment of the present invention, which is applied to a channel estimation scenario of a civil aviation air-ground broadband wireless communication system, and as shown in fig. 3, the apparatus includes:

the first processing module 31 is configured to perform Least Square (LS) channel estimation on the frequency domain data after synchronization and frequency offset compensation through a pilot frequency to obtain a channel frequency domain response at the pilot frequency;

a second processing module 32, configured to interpolate the channel frequency domain response at the pilot frequency to obtain a channel frequency domain response at the data;

a third processing module 33, configured to estimate a channel impulse response length of the channel frequency domain response according to the channel frequency domain response at the pilot frequency and the time domain channel response corresponding to the channel frequency domain response at the data.

For example, the third processing module 33 is specifically configured to:

using the cumulative mean square error function: $e\left(L\right)=E\left[\underset{k=0}{\overset{N-1}{\mathrm{\&Sigma;}}}{|{\hat{h}}_{k,i}-h_{k,i}|}^2\right]=\underset{k=L}{\overset{N-1}{\mathrm{\&Sigma;}}}E\left[{\left|{\hat{h}}_{k,i}\right|}^2\right]+L\&CenterDot;{\mathrm{\&sigma;}}_{\mathrm{noise}}^2,$ determining the value L corresponding to the minimum value of the cumulative mean square error function e (L) as the channel impulse response length of the ith OFDM symbol;

wherein,for estimated noise error, h_k,iIs the channel time domain response true value at the data,the estimated value of the time domain response of the channel at the data position is L, the length of the impulse response of the channel is expressed, N represents the total number of subcarriers, k is more than or equal to 0 and less than or equal to N, M represents the number of OFDM symbols of the orthogonal frequency division multiplexing technology, i represents the number of OFDM symbols of the ith orthogonal frequency division multiplexing technology, and i is more than or equal to 1 and less than or equal to M;

and taking the maximum value of the channel impulse response length of the M OFDM symbols as the channel impulse response length of the channel frequency domain response.

The embodiment of the invention obtains the channel frequency domain response of the pilot frequency position by carrying out LS channel estimation on the frequency domain data after synchronization and frequency offset compensation through the least square criterion of the pilot frequency position; interpolating the channel frequency domain response at the pilot frequency position to obtain the channel frequency domain response at the data position; and estimating the channel impulse response length of the channel frequency domain response according to the channel frequency domain response at the pilot frequency and the time domain channel response corresponding to the channel frequency domain response at the data. Thereby enabling: in a scenario where the channel environment is changeable in the airborne mobile communication system, even if the channel frequency domain response at the pilot frequency and the channel frequency domain response at the data change with the change of the channel environment, the present embodiment may change the estimated channel impulse response length in real time according to the time domain channel response corresponding to the channel frequency domain response at the pilot frequency and the channel frequency domain response at the data according to the change of the channel environment, and therefore, may be applicable to an airborne broadband wireless communication system where the channel environment is complicated and changeable, and may improve the channel estimation performance in a scenario where the channel environment is changeable in the airborne mobile communication system, the present embodiment may estimate the channel impulse response length according to the channel frequency domain response at the pilot frequency and the time domain channel response corresponding to the channel frequency domain response at the data according to the change of the channel environment, and the channel impulse response length changes in real time with the change of the channel environment, therefore, the method can be applied to aviation broadband wireless communication systems with complex and variable channel environments, and can improve the channel estimation performance.

Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and the modifications or the substitutions do not make the essence of the corresponding technical solutions depart from the scope of the technical solutions of the embodiments of the present invention.

Claims (2)

1. A method for estimating the length of a channel impulse response is applied to a channel estimation scene of a civil aviation air-ground broadband wireless communication system, and is characterized by comprising the following steps:

performing Least Square (LS) channel estimation on the frequency domain data after synchronization and frequency offset compensation through a pilot frequency position to obtain channel frequency domain response of the pilot frequency position;

interpolating the channel frequency domain response at the pilot frequency position to obtain the channel frequency domain response at the data position;

estimating the channel impulse response length of the channel frequency domain response according to the channel frequency domain response at the pilot frequency and the time domain channel response corresponding to the channel frequency domain response at the data;

wherein the estimating a channel impulse response length of the channel frequency domain response according to the channel frequency domain response at the pilot frequency and the time domain channel response corresponding to the channel frequency domain response at the data comprises:

using the cumulative mean square error function: $e\left(L\right)=E\&lsqb;\underset{k=0}{\overset{N-1}{\&Sigma;}}|{\hat{h}}_{k,i}-h_{k,i}{|}^2\&rsqb;=\underset{k=L}{\overset{N-1}{\&Sigma;}}E\&lsqb;|{\hat{h}}_{k,i}{|}^2\&rsqb;+L\&CenterDot;{\mathrm{\&sigma;}}_{noise}^2,$ determining the value L corresponding to the minimum value of the cumulative mean square error function e (L) as the channel impulse response length of the ith OFDM symbol;

wherein,for estimated noise error, h_k,iIs the channel time domain response true value at the data,the estimated value of the time domain response of the channel at the data position is L, the length of the impulse response of the channel is expressed, N represents the total number of subcarriers, k is more than or equal to 0 and less than or equal to N, M represents the number of OFDM symbols of the orthogonal frequency division multiplexing technology, i represents the number of OFDM symbols of the ith orthogonal frequency division multiplexing technology, and i is more than or equal to 1 and less than or equal to M;

and taking the maximum value of the channel impulse response length of the M OFDM symbols as the channel impulse response length of the channel frequency domain response.

2. An apparatus for estimating channel impulse response length, applied in a channel estimation scenario of a civil aviation air-ground broadband wireless communication system, comprising:

the first processing module is used for carrying out Least Square (LS) channel estimation on the frequency domain data after synchronization and frequency offset compensation through a pilot frequency position to obtain channel frequency domain response of the pilot frequency position;

the second processing module is used for interpolating the channel frequency domain response at the pilot frequency position to obtain the channel frequency domain response at the data position;

a third processing module, configured to estimate a channel impulse response length of the channel frequency domain response according to the channel frequency domain response at the pilot frequency and a time domain channel response corresponding to the channel frequency domain response at the data;

wherein the third processing module is specifically configured to:

using the cumulative mean square error function: $e\left(L\right)=E\&lsqb;\underset{k=0}{\overset{N-1}{\&Sigma;}}|{\hat{h}}_{k,i}-h_{k,i}{|}^2\&rsqb;=\underset{k=L}{\overset{N-1}{\&Sigma;}}E\&lsqb;|{\hat{h}}_{k,i}{|}^2\&rsqb;+L\&CenterDot;{\mathrm{\&sigma;}}_{noise}^2,$ determining the value of L corresponding to the minimum value of the cumulative mean square error function e (L) as the second valueⁱThe channel impulse response length of OFDM symbol;

wherein,for estimated noise error, h_k,iIs the channel time domain response true value at the data,the estimated value of the time domain response of the channel at the data position is L, the length of the impulse response of the channel is expressed, N represents the total number of subcarriers, k is more than or equal to 0 and less than or equal to N, M represents the number of OFDM symbols of the orthogonal frequency division multiplexing technology, i represents the number of OFDM symbols of the ith orthogonal frequency division multiplexing technology, and i is more than or equal to 1 and less than or equal to M;

and taking the maximum value of the channel impulse response length of the M OFDM symbols as the channel impulse response length of the channel frequency domain response.