CN101414986A - Channel estimation method and apparatus - Google Patents

Abstract

The embodiment of the invention provides a channel estimation device and a method thereof. The channel estimation device comprises a receiving pilot signal gaining unit which is used for gaining the receiving pilot signal at the original pilot position from the received signals, a reference pilot signal gaining unit which is used for gaining reference pilot signals, a pilot channel estimation unit which is used for estimating a first channel response according to the receiving pilot signal and the reference pilot signal, an expanding element which is used for expanding the first channel response and gaining a second channel response, and a channel estimation unit which is used for carrying out the channel estimation by the second channel response and the first channel response. The device and the method carry out the channel estimation by the second channel response and the first channel response, increase the density of the initial channel estimation, and improve the performance of channel estimation, thus improving the demodulation performance of the whole system.

Description

A kind of channel estimation methods and device

Technical field

The present invention relates to a kind of communication technology, relate in particular to a kind of channel estimation methods and device.

Background technology

Since nineteen nineties, multi-transceiver technology becomes the hot spot technology of broadband wireless communications.Wherein use the most representative system of this technology for adopting OFDM (Othogonal Frequency DivisionMultiplex, OFDM) WiMAX of technology (Worldwide Interoperability forMicrowave Access, micro-wave access to global intercommunication) system.In multi-transceiver technology, a bandwidth carrier is divided into a plurality of subcarriers, and on a plurality of subcarriers, transmits data simultaneously.In the middle of the system applies of majority, because the width of subcarrier is less than the coherence bandwidth of channel, and the decline on each subcarrier on the frequency-selective channel is flat fading, thereby just reduced intersymbol interference, and do not need complicated channel equalization, be fit to the transmission of high-speed data.At receiving terminal, before the coherent demodulation of carrying out OFDM, at first need channel is estimated and followed the tracks of, generally can adopt channel estimation methods based on supplementary.

Channel estimation methods based on supplementary is to insert some known pilot signal or training sequences in some fixed position of transmitting terminal signal, utilizes these pilot signals or training sequence to carry out channel estimating according to certain algorithm at receiving terminal.

In the WiMax system, can adopt pectination to the design of pilot signal at present, Fig. 1 shows the pilot symbol structure (2 antenna) under PUSC in the WiMAX system (mapping of the parton carrier wave) pattern.Under similar this pilot configuration, as shown in Figure 2, the channel estimating apparatus of receiving terminal comprises: receive pilot signal and obtain unit, reference pilot signal acquisition unit, pilot signal estimation unit and channel estimating unit, received signal is through behind the channel estimating apparatus, can obtain channel response, the channel response that obtains according to described channel estimating apparatus carries out equilibrium to received signal then, with the received signal that obtains to estimate.Describe each component units of channel estimating apparatus below in detail.

Receive pilot signal and obtain the unit, be used for obtaining to receive pilot signal from received signal R (K).

Reference pilot signal obtains the unit, is used to obtain reference pilot signal.

The pilot channel estimation unit is used to estimate the channel response of pilot (pilot tone) position, and the channel response of pilot frequency locations is:

L ', k ' are the frequency domain and the time-domain position of pilot sub-carrier, R _{L ', k '}The received signal of representing the individual symbol of l ', the individual sub-carrier positions of k ', S _{L ', k '}Represent the individual symbol of l ', the known reference pilot signal of the individual sub-carrier positions of k ', H _{L ', k '}The ideal frequency domain channel response of representing the individual symbol of l ', the individual sub-carrier positions of k ', N _{L ', k '}The Gaussian noise signal of representing the individual symbol of l ', the individual sub-carrier positions of k ', pilot are represented the set of all pilot signals.

Channel estimating unit: the pilot channel response of utilizing the pilot channel estimation unit to obtain is carried out channel estimating, as, by the time-frequency two-dimensional interpolation, can estimate any time-frequency position (l, the domain channel response of k) locating:

Data represents the data symbol set.

Represent the domain channel response that the estimation of l symbol, a k sub-carrier positions obtains, (l, k) ∈ data represents position distribution the to be estimated time-frequency position at whole data symbol,

$(l\′,k\′)\∈{\mathrm{\Γ}}_{k,l}\⊆\mathrm{pilot}$ Expression part and the pilot resources for the treatment of estimated position channel response adjacent (relevant), w _{L ', k ', l, k}Be the weight coefficient of two-dimentional Weiner filter,

In realizing process of the present invention, the inventor finds that there are the following problems at least in the prior art: in channel estimating unit, only utilize the locational pilot channel response of original pilots, obtain the locational channel response of all time-frequencies, the channel estimating performance of this channel estimation methods is undesirable, thereby the demodulation performance of whole system is relatively poor.

Summary of the invention

Embodiments of the invention provide a kind of channel estimation methods and device, can improve channel estimated accuracy, thereby improve the demodulation performance of whole system.

Embodiments of the invention provide a kind of channel estimating apparatus, comprise that receiving pilot signal obtains unit, reference pilot signal acquisition unit, pilot channel estimation unit, expanding element, channel estimating unit,

Described reception pilot signal obtains the unit, is used for obtaining from received signal the reception pilot signal of original pilots position;

Described reference pilot signal obtains the unit, is used to obtain reference pilot signal;

Described pilot channel estimation unit is used for estimating first channel response according to described reception pilot signal and reference pilot signal;

Described expanding element is used for first channel response is expanded, and obtains the second channel response;

Described channel estimating unit is used to utilize the described second channel response and first channel response to carry out channel estimating.

A kind of channel estimation methods comprises:

From received signal, obtain the reception pilot signal of original pilots position;

Obtain reference pilot signal;

Estimate first channel response according to described reception pilot signal and reference pilot signal;

First channel response is expanded, obtained the second channel response;

Utilize the described second channel response and first channel response to carry out channel estimating.

A kind of channel estimation methods, before obtaining the needed initial estimation of channel estimating interpolation, the pilot tone that system is sent is expanded earlier.

According to the embodiment of the invention, carry out channel estimating by the second channel response and first channel response, increased the density of initial channel estimation, improve performance for estimating channel, thereby improved the demodulation performance of whole system.

Description of drawings

Fig. 1 shows in the prior art pilot symbol structure under the PUSC pattern in the WiMAX system;

Fig. 2 shows channel estimating apparatus in the prior art;

Fig. 3 shows the pilot symbol structure of the embodiment of the invention one;

Fig. 4 shows the channel estimating apparatus of the embodiment of the invention one;

Fig. 5 shows the pilot configuration of the expansion front and back of the embodiment of the invention one;

Fig. 6 shows the pilot configuration of embodiment of the invention single antenna;

Fig. 7 shows the channel estimation methods of the embodiment of the invention two;

Fig. 8 shows the pilot configuration after the expansion of embodiment of the invention pilot configuration shown in Figure 6;

Fig. 9 shows the simulation result of signal to noise ratio-Block Error Rate of the employing two dimension Wiener filtering interpolation method of the embodiment of the invention two;

The employing that Figure 10 shows the embodiment of the invention two strengthens the simulation result of the signal to noise ratio-Block Error Rate of least square interpolation method.

Embodiment

Understand and realization the present invention the existing embodiments of the invention of describing in conjunction with the accompanying drawings for the ease of persons skilled in the art.

Embodiment one

As shown in Figure 3, suppose that now there are two transmit antennas in the WiMAX system, pilot configuration is a pectination, and is that time domain and/or frequency domain are staggered, and three kinds of signal: Pilot (pilot signal), Zero (spacing wave) and User data (user data signal) are wherein arranged.

Present embodiment provides a kind of channel estimating apparatus.As shown in Figure 4, the channel estimating apparatus of present embodiment comprises that receiving pilot signal obtains unit, reference pilot signal acquisition unit, pilot channel estimation unit, expanding element, channel estimating unit.Describe each unit of channel estimating apparatus below in detail.

Receive pilot signal and obtain the unit, be used for obtaining the reception pilot signal of original pilots position from received signal

Reference pilot signal obtains the unit, is used to obtain reference pilot signal

The pilot channel estimation unit is used for the estimating pilot frequency channel response, and method of estimation is as follows: by the pilot signal of original pilot frequency locations in the received signal

With this pilot frequency locations corresponding reference pilot signal

Utilize formula (3) to estimate that the channel response of original pilots position (describe for convenient, the channel response of original pilots position is also referred to as first channel response) is:

Wherein, l _o', k _o' be the frequency domain and the time-domain position of original pilots subcarrier,

Represent l _o' individual symbol, k _oThe received signal of ' individual sub-carrier positions,

Represent l _o' individual symbol, k _oThe known reference pilot signal of ' individual sub-carrier positions,

Represent l _o' individual symbol, k _oThe ideal frequency domain channel response of ' individual sub-carrier positions,

Represent l _o' individual symbol, k _oThe Gaussian noise signal of ' individual sub-carrier positions, pilot represents the set of original pilots.

Expanding element is used for first channel response is expanded, and can obtain spread pilot channel response (describe for convenient, the spread pilot channel response is also referred to as the second channel response) as shown in Figure 5.Described expanding element can be specially the time-frequency expanding element, and described time-frequency expanding element is used in time domain and/or frequency domain first channel response being expanded, to get the second channel response.The method of expansion can be to utilize linear interpolation method or non-linear interpolation method, for example, when extended method is the method f of simple time domain interpolation, has

Wherein, l ', k ' is the frequency domain and the time-domain position of expansion original pilots, Epilot is the pilot set of expansion.

Channel estimating unit utilizes the described second channel response and first channel response to carry out channel estimating.Described channel estimating unit can be specially the time-frequency channel estimating unit, described time-frequency channel estimating unit, be used for according to the described second channel response and first channel response, utilize the time-frequency two-dimensional interpolating method to carry out channel estimating, can estimate any time-frequency position (l, the domain channel response of k) locating:

Data represents the data symbol set.

Because expanding element is expanded first channel response, has increased the second channel response, therefore, has increased the density of initial channel response when channel estimating, thereby when estimating the channel response of any time-frequency position, can increase channel estimating performance.

Embodiment two

Present embodiment provides a kind of channel estimation methods, as shown in Figure 3, supposes that the WiMAX system still adopts double antenna, and pilot configuration is a pectination, frequency pilot sign to be distributed as time domain and/or frequency domain staggered.Describe the channel estimation methods of present embodiment below with reference to Fig. 7, described method comprises step:

Step 71, from received signal, obtain the reception pilot signal of original pilots position.

Step 72, acquisition reference pilot signal.

Step 73, estimate first channel response according to described reception pilot signal and reference pilot signal.

Step 74, first channel response is expanded, can be obtained the second channel response; After first channel response was expanded, pilot distribution as shown in Figure 5.

Can expand first channel response in time domain and/or frequency domain, the method for expansion can be to utilize linear interpolation method or non-linear interpolation method.

The method of extended channel response under the double antenna situation has been described in embodiment one and present embodiment.The embodiment of the invention is fit to single antenna equally, reaches other many antennas (more than or equal to 3 antennas) situation the extended method of first channel response.Fig. 6 shows the pilot configuration under the single-antenna case; Fig. 8 shows the pilot configuration after the expansion of pilot configuration shown in Figure 6.

Step 75, utilize the response of described second channel and first channel response, utilize the time-frequency two-dimensional interpolating method to carry out channel estimating.Described time-frequency two-dimensional interpolating method comprises two-dimentional Wei Na (Wiener) filtering interpolation method and strengthens the least square interpolation method.Introduce this two kinds of methods below respectively.

1, TWO DIMENSIONAL WIENER filtering interpolation method

Any time-frequency position (l, the domain channel response of k) locating:

Data represents the data symbol set.

In formula (A-1), (l ', k ') ∈ Γ _{L, k}Be illustrated in actual estimated

The time estimation region in all pilot resources set.In order to reduce complexity, estimation region can only utilize part and the pilot resources for the treatment of that the estimated position channel response is adjacent or relevant, at this moment less than the data area (set of several continuous subchannel time slots) of descending symbol $(l\′,k\′)\∈{\mathrm{\Γ}}_{l,k}\⊆\mathrm{pilot},$ The coefficient number of estimation filter also is ‖ Γ simultaneously _{K, l}‖.w _{L ', k ', l, k}Weight coefficient for two-dimentional Weiner filter.Estimating w _{L ', k ', l, k}The time follow MMSE criterion, i.e. mean square deviation of Gu Jiing

Minimum, then

By orthogonality principle (projection theorem) as can be known:

Formula (A-1) substitution (A-2) is then had

The definition cross-correlation function

The definition auto-correlation function

With (4) substitution (A-5), and can get according to (A-4)

Wherein E (| S _{L ', k '}| ²) the frequency pilot sign S that sends of expression _{L ', k '}Average energy.

So when we obtain the cross-correlation function of channel and the signal to noise ratio of frequency pilot sign by channel statistic property After, just can be very according to formula (A-3) and (A-6) estimate filter factor w _{L ', k ', l, k},

$w_{l,k}^T={\mathrm{\θ}}_{l,k}^T{\mathrm{\Φ}}^{-1},---(A-7)$

Carry out channel estimating according to the second channel response and first channel response, when channel estimating, increased the density of initial channel response, can improve performance for estimating channel.When moving velocity of terminal is 30 kilometers/hour, when adopting the second channel response shown in Figure 8 and first channel response to carry out channel estimating, can obtain the simulation result of signal to noise ratio-Block Error Rate as shown in Figure 9.Wherein the modulation system of system's employing is 64QAM, and coded system is 1/2CTC.

By analogous diagram shown in Figure 9 as can be seen, when adopting the second channel response and first channel response to carry out channel estimating, can reach about 2dB the performance improvement of modulating.

2, strengthen the least square interpolation method

The interpolation channel estimation methods that strengthens least square (LSE) is described below.Suppose:

OFDM symbol fft_size=N;

The subcarrier number of wherein carrying data is M, M＜N;

The pilot tone number that uniformly-spaced is inserted with in OFDM symbol is S, S＜M;

The time domain extension length of channel is L, L＜＜N;

The channel time domain response can be expressed as vectorial h (L * 1).

Step1: at first calculate following time-frequency conversion matrix:

1. (the FFT transformation matrix W of N * N)

$W=\left[\begin{array}{cccc}{\mathrm{<figure-callout\; id="0"\; label="w"\; filenames="A200710182007E00141.png"\; state="\{\{state\}\}">w</figure-callout>}}_{\mathrm{0,0}}& {\mathrm{<figure-callout\; id="0"\; label="w"\; filenames="A200710182007E00141.png"\; state="\{\{state\}\}">\; <figure-callout\; id="1"\; label="w"\; filenames="A200710182007E00141.png,A200710182007E00182.png"\; state="\{\{state\}\}">w</figure-callout>\; </figure-callout>}}_{\mathrm{0,1}}& \&CenterDot;\&CenterDot;\&CenterDot;& {\mathrm{<figure-callout\; id="0"\; label="w"\; filenames="A200710182007E00141.png"\; state="\{\{state\}\}">w</figure-callout>}}_{0,N-1}\\ {\mathrm{<figure-callout\; id="1"\; label="w"\; filenames="A200710182007E00141.png,A200710182007E00182.png"\; state="\{\{state\}\}">w</figure-callout>}}_{\mathrm{1,0}}& {\mathrm{<figure-callout\; id="1"\; label="w"\; filenames="A200710182007E00141.png,A200710182007E00182.png"\; state="\{\{state\}\}">\; <figure-callout\; id="1"\; label="w"\; filenames="A200710182007E00141.png,A200710182007E00182.png"\; state="\{\{state\}\}">w</figure-callout>\; </figure-callout>}}_{\mathrm{1,1}}& \&CenterDot;\&CenterDot;\&CenterDot;& {\mathrm{<figure-callout\; id="1"\; label="w"\; filenames="A200710182007E00141.png,A200710182007E00182.png"\; state="\{\{state\}\}">w</figure-callout>}}_{1,N-1}\\ \&CenterDot;& & & \&CenterDot;\\ \&CenterDot;& & & \&CenterDot;\\ \&CenterDot;& & & \&CenterDot;\\ w_{N-\mathrm{1,0}}& w_{N-\mathrm{1,1}}& \&CenterDot;\&CenterDot;\&CenterDot;& w_{N-1,N-1}\end{array}\right]---(B-1)$

$w_{n,m}=\exp (-\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="A200710182007E00141.png"\; state="\{\{state\}\}">j</figure-callout>}\frac{2\mathrm{\&pi;}}{N}\mathrm{nm})$

n，m∈[0，1，2...N-1]

2. (the transformation matrix of M * L)

3. (the transformation matrix W of S * L)

Step2: pilot tone is carried out initial estimation:

The method of initial estimation estimates that for time domain LS strengthens this method of estimation can reduce the noise component(s) in the pilot tone initial estimate.

For the individual OFDM symbol of l ', the initial estimate that is obtained pilot frequency locations by formula (B-2) constitutes vector

Then, have according to the LS algorithm

Step3: frequency domain interpolation

Utilize transformation matrix

Carry out frequency domain interpolation, obtain being inserted with the channel frequency domain response on the subcarrier of all beared informations in the OFDM symbol of pilot tone.

Carry out channel estimating according to the second channel response and first channel response, when channel estimating, increased the density of initial channel response, can improve performance for estimating channel.When moving velocity of terminal is 30 kilometers/hour, when adopting the second channel response shown in Figure 8 and first channel response to carry out channel estimating, can obtain the simulation result of signal to noise ratio one Block Error Rate as shown in figure 10.Wherein the modulation system of system's employing is QPSK, and coded system is 1/2CTC.By simulation result shown in Figure 10 as can be seen, under the QPSK modulation system, this improvement algorithm has the gain about 0.5dB.

Further, by Fig. 9 and Figure 10 as can be seen, the channel estimation methods of the embodiment of the invention all can improve performance for estimating channel at the algorithms of different of channel estimating, the not modulation system of system.

According to the embodiment of the invention, carry out channel estimating by the second channel response and first channel response, when channel estimating, increase the density of initial channel response, thereby improved performance for estimating channel.

Though described the present invention by embodiment, those of ordinary skills know, without departing from the spirit and substance in the present invention, just can make the present invention that many distortion and variation are arranged, and scope of the present invention is limited to the appended claims.

Claims (11)

1. A channel estimation device, characterized in that it comprises a receiving pilot signal obtaining unit, a reference pilot signal obtaining unit, a pilot channel estimation unit, an extension unit, and a channel estimation unit, The received pilot signal obtaining unit is used to obtain the received pilot signal of the original pilot position from the received signal; The reference pilot signal obtaining unit is configured to obtain a reference pilot signal; The pilot channel estimation unit is configured to estimate a first channel response according to the received pilot signal and the reference pilot signal; The extension unit is configured to extend the first channel response to obtain a second channel response; The channel estimation unit is configured to perform channel estimation by using the second channel response and the first channel response. 2. The channel estimation device according to claim 1, wherein the extension unit is specifically a time-frequency extension unit, and the time-frequency extension unit is configured to use a linear interpolation method or The nonlinear interpolation method extends the first channel response to obtain the second channel response. 3. The channel estimation device according to claim 1 or 2, characterized in that the channel estimation unit is specifically a time-frequency channel estimation unit, and the time-frequency channel estimation unit is configured to use the second channel response and the first A channel response, using a time-frequency two-dimensional interpolation method for channel estimation. 4. A channel estimation method, characterized in that, comprising: obtaining a received pilot signal at an original pilot position from the received signal; Obtain a reference pilot signal; estimating a first channel response based on the received pilot signal and the reference pilot signal; Extending the first channel response to obtain a second channel response; Perform channel estimation by using the second channel response and the first channel response. 5. The method according to claim 4, wherein the channel estimation using the second channel response and the first channel response is specifically: using the second channel response and the first channel response, using time-frequency two dimensional interpolation method for channel estimation. 6. The method according to claim 5, wherein the time-frequency two-dimensional interpolation method includes two-dimensional Wiener filter interpolation method and enhanced least squares interpolation method. 7. The method according to claim 4, wherein said extending the first channel response comprises: extending the first channel response by using a linear interpolation method or a nonlinear interpolation method. 8. The method according to claim 4, wherein said extending the first channel response comprises: extending the first channel response in time domain and/or frequency domain. 9. The method according to any one of claims 4 to 8, characterized in that the distribution of the pilot symbols corresponding to the received pilot signal is interleaved in time domain and/or frequency domain. 10. A channel estimation method, characterized in that, before obtaining the initial estimate required for channel estimation interpolation, the pilot frequency sent by the system is extended. 11. The method according to claim 10, characterized in that the spreading of the pilot can be selected in time domain or frequency domain.

Images