KR101098760B1 - Channel estimator through comprising the same pilot detector, OFDM receiving apparatus comprising the same channel estimator, and method for estimating channel through restoring distorted pilot - Google Patents

Abstract

The present invention provides a channel estimator for estimating a channel through distortion pilot recovery, which can improve channel estimation performance by recovering distortion of a pilot caused by DC rejection or impulse noise, and estimating a channel using the recovered pilot, An OFDM receiver including the channel estimator and a channel estimation method through distortion pilot compensation are provided. The channel estimator includes: a delay delaying an Orthogonal Frequency Division Multiplexing (OFDM) signal on which a Fast Fourier Transform (FFT) is performed for a predetermined time; A pilot recovery unit that detects and recovers distortion of a pilot extracted from the OFDM signal; An interpolation unit for performing interpolation using a pilot recovered from the pilot recovery unit; And an equalizer (Equlaizer) for compensating for signal distortion by using the channel estimate extracted through the interpolator with respect to the signal output from the delayer.

Orthogonal Frequency Division Modulation (OFDM), Digital Video Broadcasting-Handheld (DVB-H), Fast Fourier Transform (FFT), Guard Interval (GI), Digital Video Broadcasting-Terrestrial (DVB-T)

Description

Channel estimator through comprising the same pilot detector, OFDM receiving apparatus including the same channel estimator, and a channel estimator for estimating a channel through distortion pilot recovery, an OFM receiver including the channel estimator, and a channel estimation method through distortion pilot compensation method for estimating channel through restoring distorted pilot}

The present invention relates to a receiving apparatus and method thereof in an orthogonal frequency division multiplexing (OFDM) system, and more particularly, to a channel estimator and a method for improving channel estimation performance in an OFDM system.

In an OFDM system, the transmitted signals are affected by multi-paths with different amplitudes and delays. This multipath causes fading, which distorts the received signal received at the receiver. In addition, in order to ensure the mobility of the communication terminal, a means for compensating for distortion of the received signal caused by the above-described problem, estimating an accurate channel estimate, and equalizing the received signal according to the estimation. Is required.

In order to estimate a channel of a signal transmitted and received in an OFDM system, a predefined training symbol is required between a transmitter and a receiver. In particular, for signals transmitted over a wireless channel, channel estimation is performed using pilot symbols included in the signal to accurately decode the received signal in a noisy environment that causes multi-echo. .

Conventional methods for channel estimation using pilot include linear interpolation and cubic interpolation, and statistical characteristics of pilot values of previous symbols are used to improve channel estimation performance. .

The conventional channel estimation technique performs channel estimation after all pilots have been received. Since the interpolation technique of one method is used, channel estimation performance is greatly affected by fading.

1A and 1B are diagrams for describing channel estimation performance deteriorated due to fading.

FIG. 1A illustrates a case of accurately estimating a channel when there is no influence due to fading. FIG. 1B illustrates an error range generated when a pilot signal is distorted due to fading by multiple paths, and the channel is estimated by the distorted pilot. That is, since the channel is estimated within the error range indicated by the circle A, the exact channel cannot be estimated, and thus the reception performance of the signal is reduced by that amount.

FIG. 2 is a structural diagram of an OFDM symbol using a pilot. In general, a channel is estimated through a linear interpolation or a lattice interpolation method for the OFDM symbol structure.

Linear interpolation is performed twice in the time axis and frequency axis. There are two methods of interpolating the time axis first, interpolating the frequency axis, and interpolating the frequency axis first and then interpolating the time axis. The interpolation performance between the two is known to be almost the same.

On the other hand, in contrast to linear interpolation in which the time axis and the frequency axis are distinguished sequentially, lattice interpolation simultaneously interpolates the time axis and the frequency axis in an oblique direction forming a lattice. This lattice interpolation requires fewer pilot symbols than the linear interpolation method, and the H / W (hardware) size for interpolation is small and the channel estimation time is fast, so that the linear interpolation method is used in the place where the channel change is severe. It is known to show better channel estimation performance.

However, the conventional channel estimation method using linear or lattice interpolation is vulnerable to DC remove or impulse noise in the frequency band, so that a burst error occurs when DC removal or impulse noise occurs. In the case where the burst error is added, a very serious problem is that recovery is impossible even by channel coding.

The problem to be solved by the present invention is to recover the distortion of the pilot caused by DC rejection or impulse noise, and to estimate the channel by using the recovered pilot, the channel through the distortion pilot recovery that can improve the channel estimation performance A channel estimator for estimating, an OFDM receiver including the channel estimator, and a channel estimation method through distortion pilot compensation are provided.

In order to achieve the above object, the present invention is a delay (Delay) for delaying the Orthogonal Frequency Division Multiplexing (OFDM) signal subjected to the Fast Fourier Transform (FFT) by a predetermined time; A pilot recovery unit that detects and recovers distortion of a pilot extracted from the OFDM signal; An interpolation unit for performing interpolation using a pilot recovered from the pilot recovery unit; And an equalizer (EQula) for compensating for signal distortion by using the channel estimate extracted through the interpolation unit with respect to the signal output from the delayer.

In the present invention, the pilot recovery unit, Correlator for obtaining a correlation value (Correlation Value) between the adjacent pilot; A distortion pilot detector for detecting a distorted pilot using the correlation value; And a pilot recoverer for recovering the detected distortion pilot. In addition, the distortion pilot detector may obtain a difference between successive correlation values, and detect that distortion occurs in the corresponding pilot when the difference between the correlation values has a predetermined threshold or more.

Meanwhile, the pilot recoverer may recover the distortion pilot by using an interpolation method by using at least one pilot adjacent to each of the distortion pilots.

The present invention also provides a CP remover for removing a cyclic prefix (CP) for an OFDM signal converted into a baseband digital signal through a receiver; An FFT unit performing an FFT on the output signal of the CP removing unit; A pilot extractor extracting a pilot from the output signal of the FFT unit; A channel estimator for detecting and recovering distortion of the pilot extracted from the OFDM signal and estimating a channel using the recovered pilot; A demapper which demaps an output signal of the channel estimator; And a channel decoder for performing channel decoding on the output signal of the demapper.

Furthermore, in order to achieve the above object, the present invention includes the steps of extracting a pilot for the signal on which the FFT is performed; Obtaining a correlation value between the adjacent pilots; Detecting a distorted pilot using the correlation value; Recovering a distortion pilot detected as being distorted; And estimating a channel using the recovered pilot.

In the present invention, the channel estimating step may include extracting a channel estimate value through an interpolation method, and compensating for signal distortion using the channel estimate value. Here, the interpolation method may be any one of linear interpolation, lattice interpolation, linear and lattice complex interpolation, and modified lattice interpolation methods. Here, the linear and lattice complex interpolation may be performed by performing and averaging linear interpolation and lattice interpolation together, and the modified lattice interpolation may be performed using fewer pilots than the lattice interpolation.

According to an embodiment of the present invention, a channel estimator for estimating a channel through distortion pilot recovery, an OFDM receiver including the channel estimator, and a channel estimation method using distortion pilot compensation recover a distorted pilot in advance through a pilot recovery unit and restore the pilot. By estimating the channel using, the channel estimation performance can be significantly improved.

That is, the pilot recovery unit detects a distorted pilot using a correlation value between pilots, and also recovers the detected distorted pilot through an adjacent pilot, thereby fundamentally solving the problem of channel estimation impossible caused by the conventional pilot itself being distorted. Can be.

The channel estimator, the OFDM receiver, and the channel estimation method of the present invention can be usefully used for channel estimation in an OFDM system such as DVB-H and DVB-T using Busan pilot, and have an advantage of simple implementation.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. In the following description, when a component is described as being connected to another component, it may be directly connected to another component, but a third component may be interposed therebetween. In addition, in the drawings, the structure or size of each component is exaggerated for convenience and clarity of explanation, and parts irrelevant to the description are omitted. Like numbers refer to like elements in the figures. It is to be understood that the terminology used is for the purpose of describing the present invention only and is not used to limit the scope of the present invention.

3 is a block diagram of an OFDM receiver according to an embodiment of the present invention.

Referring to FIG. 3, referring to FIG. 2, the OFDM receiver of the present embodiment includes an RF tuner 100 (RF Rx), an analog digital converter 200 (ADC), and a cyclic prefix remover 300 (CP remover). Pre-FFT unit 400, FFT unit 500, pilot extractor 600, pilot detector 600, channel estimator 700, demapper 800 and channel decode The unit 900 includes a channel decoder.

The RF tuner 100 converts the received signal into a baseband signal, and the ADC 200 converts an analog signal into a digital signal. The CP remover 300 removes the CP attached to the received signal, and the pre-FFT unit 400 obtains the synchronization for the symbol. That is, the pre-FFT unit 400 estimates mode detection, guard detection, and symbol timing offset (STO) / fractional carrier frequency offset (FCFO).

The FFT unit 500 performs an FFT on signals from which the CP has been removed, and the pilot extractor 600 extracts a pilot from the signals on which the FFT has been performed. The extracted pilot may be a scattered pilot. Using the extracted pilot, the channel estimator 700 performs linear interpolation, lattice interpoation, linear and lattice complex interpolation, or modified lattice interpolation. Estimate the channel through.

On the other hand, the channel estimating apparatus 700 includes a pilot recovery unit for recovering a pilot distorted by DC removal or impulse noise, unlike the conventional art. The interpolation method in the channel estimator 700 and the channel estimator 700 will be described in more detail below with reference to FIG. 4.

After channel estimation, the demapper 800 performs demapping on the data through demodulation in response to the mapped modulation scheme. For example, when a transmission signal is mapped in BPSK (Binary Phase Shift Keying), DBPSK demodulation is performed. In the case where the transmission signal is mapped in QPSK (Quadrature Phase Shift Keying), DQPSK demodulation is performed.

After demapping, the channel decoder 900 performs final channel decoding based on the channel previously estimated by the channel estimating apparatus 700.

The OFDM receiver of the present embodiment includes a pilot recovery unit, so that when DC elimination or impulse noise occurs, the distortion of the pilot is normally restored and the channel is estimated based on the recovered pilot, thereby enabling robust channel estimation. do.

4 is a block diagram illustrating a structure of a channel estimator included in the OFDM receiver of FIG. 3 in more detail.

Referring to FIG. 4, the channel estimator 700 of the present embodiment includes a delay unit 710, an interpolator 720, an equalizer 730, and a channel state information (CSI) extractor 740. And a pilot recovery unit 750. Here, the pilot recovery unit 750 detects and recovers a pilot distorted by the DC removal or the impulse noise generation of the pilot previously extracted from the pilot extraction unit 600. The pilot recovery unit 750 will be described in more detail in the description of FIGS. 5, 7, and 8.

The delay unit 710 delays the signal on which the FFT is performed by the FFT unit 500 by a predetermined time used for interpolation, and interpolates 720 the signal output through the delay unit 710. The signal distortion is compensated for in the equalizer 730 by using the channel estimate extracted at.

The interpolator 720 calculates a channel estimate by performing interpolation through the pilot recovered by the pilot recovery unit 750. Although the interpolation method can be used as an interpolation method that can be used in the interpolator 720, Other lattice interpolations, linear and lattice complex interpolations, or modified lattice interpolations can be used.

Here, linear and lattice complex interpolation is performed by performing and averaging linear interpolation and lattice interpolation together, and modified lattice interpolation modifies the existing lattice interpolation method to estimate a channel of data symbols using fewer pilots. This modified lattice interpolation adds linear interpolation in channel estimation for data symbols to reduce the number of pilots used.

The equalizer 730 compensates for the signal distortion by using the channel estimate extracted through the complex interpolation on the signal output from the delayer 710. Meanwhile, the CSI extractor 740 extracts channel state information by using the output signal from the complex interpolator.

5 is a block diagram illustrating in more detail the pilot recovery unit of FIG. 4.

Referring to FIG. 5, the pilot recoverer 750 includes a correlator 752, a distorted pilot detector 754, and a pilot recoverer 756. The correlator 752 obtains a correlation value between adjacent pilots. The pilot detector 754 obtains the difference between the correlation values thus obtained and detects that the pilot is distorted when the difference between the correlation values exceeds a predetermined threshold value. Meanwhile, the pilot recoverer 756 performs recovery on the detected distorted pilot. In this embodiment, the pilot recoverer 756 performs recovery using interpolation. Therefore, the pilot recoverer is represented as an interpolator in the drawing. Of course, the recovery of the pilot may be performed using other methods. For example, there may be a method of matching the pilot immediately preceding or the pilot immediately following.

A method for detecting and recovering a distorted pilot will be described in more detail with reference to FIGS. 6 to 8 below.

6 are graphs illustrating a frequency interpolation method after time-base interpolation when a normal case, a DC removal occurs, and an impulse noise occurs.

Referring to FIG. 6, (a) shows a state after time-base interpolation when an OFDM symbol is normally received without distortion. These OFDM symbols include pilot (circle) and data symbols (triangle). If normally received, the pilot can properly estimate the channel for data symbols.

However, when the DC cancellation occurs as shown in (b), the pilot is removed, and thus the channel estimation for the data symbols is performed abnormally as shown. That is, in normal cases, the channel is estimated along the dotted line, but in the case of DC elimination, the same channel as the solid line is estimated. Here, the position of the pilot when the X-marked 00 portion is normal.

On the other hand, even when impulse noise occurs as shown in (c), normal channel estimation is impossible. Here again, the position of the pilot when the X-marked part is normal.

7 are graphs illustrating a method of detecting a distorted pilot using a correlation value.

Referring to FIG. 7, the method for detecting a distorted pilot according to the present invention obtains a correlation value between adjacent pilots as described above, and calculates a difference between the correlation values and compares the threshold value with a predetermined threshold value. If abnormal, it is detected that distortion has appeared in the pilot.

A simple expression of this method is as follows.

d _i = (p _i-2 × p ^* _i-1 )-(p _i-1 × p ^* _i ) .................. (1)

Where d _i is the threshold difference, and p _i is the current pilot being examined and p _i-1 , p _i-2 are the previous and previous pilots of the current pilot. Specifically, referring to (a) the drawing, when distortion occurs in the fourth pilot p ₄ as shown, the correlation value between the second pilot p ₂ and the third pilot p ₃ is OFDM Due to the coherent bandwidth characteristic of the symbol, it has a certain value. However, in the case of the pilot p _{4 in} which the distortion occurs, the correlation value with the adjacent pilot has a value that is very large or small compared with other correlation values. Accordingly, the difference between successive correlation values also becomes large. In addition, the correlation value for the sixth pilot (p ₆ ) also has a correlation value with the adjacent pilot is very large or small compared to other correlation values due to the distortion (p ₄ ) generated distortion. As a result, the difference in successive correlation values also increases.

As shown in (b), when the distortion occurs in the fourth pilot, it can be seen that the correlation value difference (di) calculated through Equation (1) is much larger than the difference value of the correlation value elsewhere. Also, due to the distortion of the fourth pilot, the correlation value difference di for the sixth pilot may also be confirmed to be much larger than the difference value of the correlation elsewhere. Here, the dotted line is a threshold value for determining whether distortion occurs. This threshold value is preferably selected according to the characteristics of the OFDM symbol. Meanwhile, although the previous pilot symbol is used to detect distortion of the current pilot symbol, the subsequent pilot symbol may be used. On the other hand, the difference value of this correlation value is an absolute value.

(c) and (d) show the case where the pilot distortion is caused by the impulse noise. Also, when the fourth pilot is distorted, it can be seen that the correlation difference is derived from the fourth and sixth.

8 are graphs illustrating a method of recovering a pilot using two or more adjacent pilots.

Referring to FIG. 8, when the distorted pilot is detected through the above method, it is necessary to recover the distorted pilot. In this embodiment, an interpolation method is used for distorted pilot recovery. That is, interpolation is performed using pilots adjacent to both of the distorted pilots to recover the distorted pilots.

(a) shows interpolation using two adjacent pilots, and (b) shows two pilots before and one pilot after the distorted pilot (p ₄ ). . Using multiple pilots for distorted pilot recovery is more accurate. However, in consideration of computational complexity, it is desirable to perform recovery using an appropriate number of pilots. Here, black circles are distorted pilots, circles of lattice hatching are recovered pilots, and circles of solid hatching are pilots used for pilot recovery.

On the other hand, the recovery method may be a method using a pilot value immediately before or after the interpolation method. Therefore, the repairing method of the distorted pilot is not limited to the interpolation method.

The channel estimator of the present embodiment first detects and recovers a distorted pilot and estimates a channel, thereby preventing in advance a serious error in channel estimation that has occurred based on a conventionally distorted pilot, thereby enabling robust channel estimation. .

9 is a flowchart illustrating channel estimation methods according to another embodiment of the present invention.

Referring to FIG. 9, a pilot is first extracted from a received OFDM signal (S100). Next, a correlation value between adjacent pilots is obtained (S200). Thereafter, the difference between the correlation values is obtained to detect the distorted pilot (S300). That is, when the difference between the correlation values is obtained and the value is compared with a predetermined threshold, the corresponding pilot is detected as distorted. do. If a distorted pilot is detected, the distorted pilot is recovered (S400). The recovery method may use an interpolation method using a pilot adjacent to the distorted pilot, but is not limited thereto. After pilot recovery, the channel for the data symbols is estimated through various interpolation methods (S500).

So far, the present invention has been described with reference to the embodiments shown in the drawings, which are merely exemplary, and those skilled in the art will understand that various modifications and equivalent other embodiments are possible therefrom. . Therefore, the true technical protection scope of the present invention will be defined by the technical spirit of the appended claims.

1A and 1B are diagrams for describing channel estimation performance deteriorated due to fading.

2 is a structural diagram of an OFDM symbol using a pilot.

3 is a block diagram of an OFDM receiver according to an embodiment of the present invention.

4 is a block diagram illustrating in more detail the structure of the channel estimator included in the OFDM receiver of FIG. 2.

5 is a block diagram illustrating in more detail the pilot recovery unit of FIG. 4.

6 are graphs illustrating a frequency interpolation method after time-base interpolation when a normal case, a DC removal occurs, and an impulse noise occurs.

7 are graphs illustrating a method of detecting a distorted pilot using a correlation value.

8 are graphs illustrating a method of recovering a pilot using two or more adjacent pilots.

9 is a flowchart illustrating channel estimation methods according to another embodiment of the present invention.

Claims (15)

A delay (Delay) for delaying an Orthogonal Frequency Division Multiplexing (OFDM) signal subjected to Fast Fourier Transform (FFT) by a predetermined time; A pilot recovery unit that detects and recovers distortion of a pilot extracted from the OFDM signal; An interpolation unit for performing interpolation using a pilot recovered from the pilot recovery unit; And An equalizer (Equlaizer) for compensating for signal distortion by using the channel estimate value extracted through the interpolator with respect to the signal output from the delayer, The pilot recovery unit obtains a correlation value between adjacent pilots, obtains a difference between successive correlation values, and when the difference between the correlation values has a predetermined threshold or more, distortion occurs in the corresponding pilot. And estimating a channel estimator. The method according to claim 1, The pilot recovery unit, A correlator for obtaining the correlation value; A distortion pilot detector for detecting a distortion pilot using the correlation value; And And a pilot recoverer for recovering the detected distortion pilot. delete The method of claim 2, The pilot restorer, And recovering the distortion pilot by an interpolation method using at least one pilot adjacent to both of the distortion pilots. The method according to claim 1, The interpolation unit interpolates using any one of linear interpolation, lattice interpolation, linear and lattice complex interpolation, and enhanced lattice interpolation methods. Characterized by a channel estimator. 6. The method of claim 5, The linear and lattice complex interpolation is performed by performing and averaging linear interpolation and lattice interpolation together, Wherein the modified lattice interpolation is performed using a smaller number of the recovered pilots than the lattice interpolation. A CP remover for removing a CP (Cyclic Prefix) for an OFDM signal converted into a baseband digital signal through a receiver; An FFT unit performing an FFT on the output signal of the CP removing unit; A pilot extractor extracting a pilot from the output signal of the FFT unit; A channel estimator for detecting and recovering distortion of the pilot extracted from the OFDM signal and estimating a channel using the recovered pilot; A demapper for demapping the output signal of the channel estimator; And And a channel decoder configured to perform channel decoding on the output signal of the demapper. The channel estimator includes a pilot recovery unit that detects and recovers a distortion of a pilot extracted from the OFDM signal, The pilot recovery unit obtains a correlation value between adjacent pilots, obtains a difference between successive correlation values, and detects that distortion occurs in the corresponding pilot when the difference between the correlation values has a predetermined threshold value or more. Is an OFDM receiver. 8. The method of claim 7, The channel estimator A delay (Delay) for delaying an Orthogonal Frequency Division Multiplexing (OFDM) signal subjected to Fast Fourier Transform (FFT) by a predetermined time; An interpolation unit for performing interpolation using the pilot recovered by the pilot recovery unit; And And an equalizer (EQula) (EQula) for compensating for signal distortion by using the channel estimate extracted through the interpolator with respect to the signal output from the delayer. The method of claim 8, The pilot recovery unit, A correlator for obtaining the correlation value; A distortion pilot detector for detecting a distortion pilot using the correlation value; And And a pilot recoverer for recovering the detected distortion pilot. In the pilot extractor, Extracting a pilot for the signal on which the FFT is performed; Obtaining a correlation value between the adjacent pilots in a pilot recovery unit; Detecting, by the pilot recovery unit, a distortion pilot using the correlation value; Recovering, by the pilot recovery unit, a distortion pilot detected as distorted; And Estimating a channel using the recovered pilot in the interpolator; Wherein the detection, to obtain the correlation between adjacent pilots, obtain a difference between the correlation value of a row, and wherein the detection in the case where a difference between the correlation values with a predetermined threshold value or more to the distortion pilot Channel estimation method. delete The method of claim 10, The distortion pilot recovery step may be performed by interpolation using at least one pilot adjacent to each of the distortion pilots. The method of claim 10, The channel estimating step Extracting a channel estimate through an interpolation method, and compensating for signal distortion using the channel estimate. The method of claim 13, The interpolation method is any one of linear interpolation, lattice interpolation, linear and lattice complex interpolation, and modified lattice interpolation methods. 15. The method of claim 14, The linear and lattice complex interpolation is performed by performing and averaging linear interpolation and lattice interpolation together, And the modified lattice interpolation is performed using a smaller number of the recovered pilots than the lattice interpolation.

Images