KR100877742B1 - Channel Estimation Apparatus and Method in a Wireless Communication System Supporting OPDM or OPDMA - Google Patents

Abstract

An apparatus and method for estimating a channel in a wireless communication system supporting OFDM or OFDMA are disclosed. The method includes extracting a plurality of pilots from a received signal to compensate for an error according to a time offset and / or a carrier frequency offset, channel estimates of the pilots for which the error is compensated, and weights precomputed for each of the pilots. Estimating a channel for the received signal by using a multiplication of the signals, and the apparatus comprises: a time offset estimator for estimating a time offset by receiving the received signal; and using the estimated time offset to compensate for a phase error. A time offset compensation unit, a weight storage unit configured to store precomputed weights, and a channel of the received signal in the form of a product of pilot channel estimates and the weights of the received signal whose time offset is compensated. Interpolation (int) used for channel estimation, including at least one channel estimator By reducing non-parallel processing such as erpolation and / or averaging, the implementation of the receiving system can be facilitated and the amount of computation can be reduced.

Description

Apparatus and Method for Estimating Channel in OFDM / OFDMA System for Wireless Communication System Supporting OPDM or OPDMA

1 is a diagram illustrating an example of a frame structure used in an IEEE 802.16d / e based portable Internet system.

FIG. 2 is a diagram illustrating a part of a subcarrier structure allocated in a downlink PUSC subchannel section of FIG. 1; FIG.

3 is a diagram illustrating an outline of a SISO system and a MIMO system.

4 is a diagram illustrating a signal transmission method between a transmitting antenna and a receiving antenna in a 2x2 MIMO system.

FIG. 5 is a diagram illustrating a pilot pattern in downlink PUSC mode transmitted by a first transmit antenna and a second transmit antenna in a 2x2 MIMO system, respectively. FIG.

6 illustrates pilot and data patterns received at a receive antenna.

7 is a block diagram illustrating a channel estimating apparatus according to an embodiment of the present invention.

FIG. 8 is a configuration diagram illustrating a time offset estimator of FIG. 7. FIG.

9 is a view for explaining a time offset estimation method according to the present invention.

FIG. 10 is a configuration diagram illustrating a carrier frequency offset estimator of FIG. 7. FIG.

11 is a flow chart of a method for calculating weights in accordance with the present invention.

FIG. 12 is a diagram illustrating a part of a frame for explaining a weight calculation method of FIG. 11; FIG.

13 is a flowchart illustrating a channel estimation method according to an embodiment of the present invention.

Explanation of symbols on the main parts of the drawings

100: FFT unit 200: offset estimation unit

300: offset compensation unit 400: channel estimation unit

500: weight storage unit

The present invention relates to channel estimation in an orthogonal frequency division multiplexing transmission system, and more particularly, to an apparatus and method for channel estimation according to a pilot pattern in a wireless communication system supporting OFDM or OFDMA.

Orthogonal Frequency Division Multiplexing (OFDM) or Orthogonal Frequency Division Multiple Access (OFDMA) based on OFDM is a transmission scheme that transmits data in parallel using multiple subcarriers having orthogonality to each other instead of a single wide carrier. Even in a frequency selective fading channel having a large ISI (Inter-Symbol Interference), the method is based on the fact that each subchannel of a narrow band has a flat fading characteristic.

Such an OFDM / OFDMA system has higher frequency efficiency and transmission rate than a communication system using a single carrier. However, even in such an OFDM / OFDMA system, the reception side requires distortion compensation according to a channel environment for the received OFDM / OFDMA symbol (hereinafter, referred to as a 'symbol'). In particular, when an OFDM / OFDMA system is a system that guarantees mobility, such as a portable Internet service, the wireless channel environment is time-varying. Accordingly, channel estimation should also be designed to keep track of changing channels. For channel estimation for the time-varying channel, the transmitting side transmits a pilot signal known to the receiving side to the pilot subcarrier allocated to some subcarriers in the symbol. Then, the receiver performs channel estimation on a subcarrier through which data is actually transmitted using a pilot.

However, in the case of estimating a channel using the above-described pilot, a complicated operation occurs due to non-parallel operation such as interpolation on the time axis and / or interpolation on the frequency axis, and thus there is a difficulty in hardware implementation. In particular, in a multiple input multiple output (MIMO) system that performs multiple input / output transmissions using a plurality of transmit antennas and a plurality of receive antennas, a plurality of channels exist between a transmitting side and a receiving side. It is necessary to estimate and compensate. Accordingly, in the case of the MIMO system, there is a problem in that the above-mentioned complex operation is further increased according to the number of antennas.

Therefore, research and development on this is ongoing, and in particular, when the MIMO algorithm is applied to the Partial Usage of Subchannels (PUSC) mode, which is preferentially adopted in the OFDM / OFDMA system, the optimal channel estimation for solving the above-mentioned problems. Technology is required.

The present invention was devised to meet the above requirements, and an object of the present invention is to provide an apparatus and method for channel estimation according to a pilot pattern in a wireless communication system supporting OFDM or OFDMA.

In addition, another object of the present invention is to provide a channel estimation apparatus and method for reducing the complexity in channel estimation of the receiving system by offline writing the weight according to the pilot pattern in advance in a multiple input and output wireless communication system supporting OFDM or OFDMA. have.

For this purpose, an apparatus for estimating a channel in a wireless communication system supporting OFDM or OFDMA of one embodiment of the present invention includes a time offset estimating unit for receiving a received signal and estimating a time offset; A time offset compensator for compensating for a phase error using the estimated time offset; A weight storage unit configured to store a pre-calculated weight; And at least one channel estimator estimating the channel using a product of a pilot channel estimate and a weight of the received signal whose time offset is compensated.

Meanwhile, in a wireless communication system supporting OFDM or OFDMA according to one embodiment of the present invention, a method of estimating a channel includes (a) extracting a plurality of pilots from a received signal to correct an error according to a time offset and / or a carrier frequency offset. Compensating; And (b) estimating a channel for the received signal using a product of channel estimates of the error compensated pilots and weights precomputed for each of the pilots.

In addition, a method for estimating a channel in a wireless communication system supporting OFDM or OFDMA according to another aspect of the present invention includes: (a) obtaining a pilot channel estimate for pilots obtained in each cluster; And (b) estimating the channel using a weight previously calculated based on at least one of time-axis interpolation, frequency-axis interpolation, and a moving average, and the pilot channel estimate.

On the other hand, a method for estimating a channel in a wireless communication system supporting OFDM or OFDMA of one embodiment of the present invention includes the steps of: (a) obtaining a pilot channel estimate for a pilot obtained in a subcarrier frequency allocation unit; And (b) estimating the channel using a weight previously calculated based on at least one of time-axis interpolation, frequency-axis interpolation, and a moving average, and the pilot channel estimate.

Hereinafter, the present invention will be described in detail with reference to the accompanying drawings and preferred embodiments. For reference, in the following description, detailed descriptions of well-known functions and configurations that may unnecessarily obscure the subject matter of the present invention are omitted.

FIG. 1 is a diagram illustrating an example of a frame structure used in an IEEE 802.16d / e based portable Internet system. A portable Internet system using the TDD method divides one frame in time and uses it for transmission and reception. .

Referring to FIG. 1, one frame is divided into a downlink frame for transmitting data from a base station to a terminal and an uplink frame for transmitting data from a terminal to a base station, with TTG ( Transmit / receive transition gap and RTG (receive / transmit transition gap) are inserted. In the illustrated example, the downlink frame includes at least a preamble section, a partial usage of subchannels (PUSC) subchannel section, a full usage of subchannels (FUSC) subchannel section, an adaptive modulation & coding (AMC) subchannel section, and the like. One uplink frame includes at least one uplink symbol interval, a PUSC subchannel interval, an AMC subchannel interval, and the like.

In particular, in connection with the present invention, when using 1024 fast fourier transform (FFT) of the subcarrier allocation method for the downlink PUSC mode, it can be allocated as shown in Table 1 below, Figure 2 is a subcarrier allocation structure according to Table 1 Some are shown.

 TABLE 1

Referring to Table 1 and FIG. 2, in downlink PUSC mode using 1024 FFT, 92 and 91 subcarriers, respectively, are used as protection intervals to alleviate interference between adjacent channels among the 1024 subcarriers. Used as a subcarrier. In addition, 840 subcarriers except these are used as effective subcarriers, 120 subcarriers are used as pilots, and the remaining 720 subcarriers are used for data transmission.

The downlink PUSC subchannel period is defined over two consecutive symbol periods on the time axis, and includes a frame control header (FCH) for transmitting frame configuration information. In addition, the downlink PUSC subchannel section includes subcarriers distributed on a frequency axis, and one downlink PUSC subchannel includes 4 pilot subcarriers and 48 data subcarriers. The basic component unit of the downlink PUSC subchannel is a cluster, and the cluster blocks all subcarriers except for the null subcarrier and the DC subcarrier into 14 adjacent subcarriers.

Meanwhile, the present invention is also applied to a MIMO system that performs multiple input / output transmissions using a plurality of transmit antennas and a plurality of receive antennas. Hereinafter, the MIMO system will be described with reference to FIGS. 3 to 5.

First, FIG. 3 is a diagram for explaining an outline of a SISO system and a MIMO system.

As shown in FIG. 3A, a single input single output (SISO) system performs single input / output transmission through one channel H formed between one transmit antenna TxAnt and one receive antenna RxAnt. Perform.

In contrast, a multiple input multiple output (MIMO) system performs multiple input / output transmissions through a plurality of channels formed between a plurality of transmit antennas and a plurality of receive antennas. FIG. 3 (b) illustrates a 2 × 2 MIMO system using two transmit antennas and two receive antennas, of which the first and second transmit antennas TxAnt0 and TxAnt1 and the first are shown. Four channels, that is, a first channel H00, a second channel H01, a third channel H10, and a fourth channel H11 are formed between the second reception antennas RxAnt0 and RxAnt1. For reference, in channel notation H00, the first index 0 is related to the index of the receiving antenna, and the second index 0 is related to the index of the transmitting antenna.

Hereinafter, a method of transmitting a signal of a 2 × 2 MIMO system will be described in detail with reference to FIG. 4.

In the downlink period, a base station (Base Station / Radio Access Station) transmits signals to two transmit antennas (TxAnt0 and TxAnt1), and a terminal (Mobile Station / Portable Subscriber Station) to two receive antennas (RxAnt0 and RxAnt1). Receive the signal. In this case, the preamble is transmitted from one antenna TxAnt0 of two transmit antennas, and the first receiving antenna RxAnt0 and the second receiving antenna RxAnt1 are respectively the first channel H00 and the third channel. The preamble is received through the H10 (see FIG. 4A). The pilot is transmitted in different patterns at the first and second transmit antennas TxAnt0 and TxAnt1, respectively, and the first receive antenna RxAnt0 is transmitted through the first channel H00 and the second channel H01, and The two receiving antennas RxAnt1 receive both pilots transmitted from the first and second transmitting antennas through the third channel H10 and the fourth channel H11, respectively (see FIG. 4 (b)).

FIG. 5 illustrates a pilot pattern in downlink PUSC mode transmitted by the first transmit antenna TxAnt0 and the second transmit antenna TxAnt1, respectively, and is a pilot pattern to which a space time code (STC) is applied. In order to apply spatial multiplexing (SM), different data is transmitted for each antenna.

Referring to FIG. 5, the first transmit antenna TxAnt0 transmits pilot and data in the pattern shown in FIG. 5A, and the second transmit antenna TxAnt1 pilots in the pattern shown in FIG. 5B. And transmit the data. Then, the first reception antenna RxAnt0 receives the first and second reception signals (that is, the reception signals of the first channel and the second channel) through the first and second channels H00 and H01, respectively. 2 The reception antenna RxAnt1 receives the third and fourth reception signals (that is, the reception signals of the third channel and the fourth channel) through the third and fourth channels H10 and H11, respectively, and transmits two transmission signals. Receive all signals (downlink frame) transmitted from the antenna.

At this time, the pilot and data patterns received at each receive antenna are as shown in FIG.

Referring to FIG. 6, since the pilot pattern is repeated in a period of 4 symbols in the downlink PUSC mode according to the present invention, the entire pilot pattern for the downlink PUSC mode may be expressed by Equation 1 below. Here, m represents a receive antenna index and l ₀ represents a symbol index. In addition, equations 1- (1) and 1- (3) represent pilot patterns received through the first and second channels, respectively. Represents a pilot pattern received over four channels.

[Equation 1]

Hereinafter, a channel estimation apparatus and method according to an embodiment of the present invention will be described with reference to FIGS. 7 to 13. For reference, the present embodiment relates to a channel estimation apparatus and method applied to a 2x2 MIMO system, and the downlink PUSC mode used in the present embodiment is a pilot pattern to which a space time code (STC) is applied, and a space In order to apply Spatial Multiplexing (SM), different data is transmitted for each transmit antenna.

FIG. 7 is a block diagram illustrating a channel estimating apparatus according to an embodiment of the present invention, and includes a first channel H00, a second channel H01, a third channel H10, and a fourth channel H11. A channel estimation apparatus for the received signal is shown.

As shown in FIG. 7, the channel estimating apparatus includes an FFT (Fast Fourier Transform) unit 100, an offset estimator 200, an offset compensator 300, a weight storage unit 500, and a channel weight. Government 400.

The FFT unit 100 converts a signal in the baseband time domain received through the first channel and the second channel into a signal in the frequency domain. Here, the FFT unit 100 converts the received signals of the first channel and the second channel of the time domain received through the first receiving antenna into signals of the frequency domain, and although not shown, the third channel and the fourth channel. It can be easily inferred that there is a separate FFT unit (not shown) for converting a signal in the time domain into a signal in the frequency domain through. Of course, one FFT unit may be implemented to convert signals in all time domains into signals in frequency domain.

The offset estimator 200 estimates a time offset (TO) and / or a carrier frequency offset (CFO) using the converted frequency domain signal. The signal in the frequency domain converted by the FFT unit 100 includes a preamble, a pilot, data, etc., the preamble is a preamble extractor (not shown), and the pilot is a pilot extractor based on Equation 1 described above. Extracted from each other (not shown) and input to the offset estimator 200. Then, the offset estimator 210 estimates the time offset and the carrier frequency offset using the extracted preamble and pilot. The offset estimator 200 is divided into a time offset estimator 210 estimating a time offset and a carrier frequency offset estimator 220 estimating a carrier frequency offset, which will be described later with reference to FIGS. 8 to 10. do.

The offset compensator 300 compensates for errors generated when the channel passes by using the time offset or the carrier frequency offset estimated by the offset estimator 200. The offset compensator 300 is divided into a time offset compensator 310 for compensating a time offset and a carrier frequency offset compensator 320 for compensating a carrier frequency offset, which will be described later.

On the other hand, the weight storage unit 500 is configured to store the weight calculated through the channel response for each subcluster using the pilot offline. This is based on the fact that the channel characteristics in each cluster are a combination of different ratios of constant pilots in each subcluster unit. Accordingly, in consideration of the subcarrier spacing according to the pattern of each subcluster and adjacent pilots, the weight storage unit 500 obtains a weight reflecting a high ratio in the pilot close to the subcluster and a low ratio in the distant pilot in advance. By storing, it is possible to reduce the amount of computation and its computation time during channel estimation. In the present invention, a weighted pilot pattern is used for each subcluster unit, but a weight may be calculated by combining the pilot pattern and the preamble pattern. Detailed description thereof will also be described later.

The channel estimator 400 estimates a channel based on the weights stored in the weight storage 500. By doing so, the channel estimator 400 can simply obtain a channel estimate for each subcluster in the form of the product of each pilot's channel response and the calculated weight. The channel estimator 400 estimates the channels for the first channel H00 and the second channel H01, and although not shown, the channels for the third channel H10 and the fourth channel H11. Another channel estimator for estimating a can also be easily inferred, and a detailed description of the channel estimator 400 will be described later.

On the other hand, although not shown in Figure 8, the channel estimating apparatus according to an embodiment of the present invention, the space-time code decoder (not shown) for decoding the space-time code encoded on the transmitting side and / or data transmitted differently for each antenna It may further include a spatial multiplexing decoder to decode a next stage of the channel estimator 400.

The configuration of the channel estimation apparatus according to an embodiment of the present invention configured as described above will be described in more detail.

8 is a block diagram illustrating a time offset estimator in accordance with an embodiment of the present invention.

As shown in FIG. 8, the time offset estimator 210 includes a first phase difference calculator 211, a first phase difference accumulator 212, a first linear phase calculator 213, and a time offset calculator 214. ).

The first phase difference calculator 211 receives received signals (first and second received signals) of the first and second channels received through the first receiving antenna and the third and fourth received through the second receiving antenna. The phase difference according to the time offset is calculated using at least two preambles included in at least one of the received signals (the third and fourth received signals) of the channel. In this case, a combination of at least two preambles and at least two pilots may be used. The first phase difference operator 211 may be implemented in the form of a multiplier that performs conjugate multiplication with respect to two complex numbers, for example.

The first phase difference accumulator 212 accumulates the phase difference according to each time offset calculated by the first phase difference calculator 211 to generate a phase difference accumulation value. The first phase difference accumulator 212 may estimate a more accurate time offset by accumulating the phase difference calculated for a greater number of preambles. For reference, the first phase difference accumulator 212 may be implemented in the form of an adder.

The first linear phase calculator 213 converts the phase difference accumulator accumulated in the first phase difference accumulator 212 into a linear phase Ф _TO according to a time offset. Since the phase difference cumulative values exist in a complex form, the linear phase calculator 213 converts the phase difference cumulative values into a real denominator and an imaginary part as a numerator, and performs an arc tangent operation on the converted fractions. Then, by dividing this by the subcarrier index difference (that is, the preamble position difference used in the phase difference calculation), the linear phase according to the time offset can be obtained. Here, the arc tangent operation may be performed by using a look-up table that takes a ratio of a complex real part and an imaginary part as an input and outputs an arc tangent calculated value, and uses another known calculation method. You can also find the linear phase. The linear phase? _TO for the time offset obtained as described above represents an average phase difference according to a time offset occurring between adjacent subcarriers (that is, subcarriers having a difference of subcarrier indices of 1).

The time offset calculator 214 converts the linear phase? _TO according to the time offset calculated by the first linear phase calculator 213 into a time offset TO. For example, when using the 1024 FFT as in the present embodiment, the time offset (TO) can be calculated by the following equation (2).

[Equation 2]

Hereinafter, a method of estimating a time offset will be described with reference to FIG. 9.

Referring to FIG. 9, the first symbol of the downlink frame is used as a preamble. Since the preambles have a high signal level and the same symbol index, it is easy to estimate a phase difference according to a time offset. 9 shows a preamble transmission structure divided into three segments (Segment 0, Segment 1, and Segment 2). Accordingly, the base station transmits the preamble subcarrier in a pattern corresponding to one of the three segments. In addition, left and right guard bands for reducing interference of adjacent frequency bands are formed on the left and right sides of the preamble subcarrier, and the first segment 0 includes a DC subcarrier (preamble subcarrier index = 142). . In addition, a phase difference corresponding to three times the linear phase according to a time offset occurs between adjacent preamble subcarriers in one segment (when the difference in the preamble index is 1), and when the difference in the preamble index is 2, the time offset occurs. It can be seen that a phase difference corresponding to six times the linear phase occurs. For reference, Equation 3 below shows an example of a calculation result of the linear phase according to the time offset calculated by the first linear phase calculator 213 through the first phase difference calculator 211 and the first phase difference accumulator 212. It is shown. In Equation 3, P denotes a preamble subcarrier, k denotes a preamble subcarrier index, and m denotes a reception antenna index.

[Equation 3]

The linear phase θ _TO according to the time offset thus obtained is converted into a time offset TO value in the time offset calculator 214, which is used later to compensate for the time offset in the time offset compensator 310.

Here, an embodiment according to the present invention uses a method of calculating a phase difference according to a time offset using a pilot pair having the same symbol index in addition to the above-described method for estimating a time offset, and using two pairs of pilots having the same symbol index difference. A method of calculating a phase difference according to a time offset may also be applied.

은 시간 오프셋이 보상된 하향링크 PUSC 모드의 파일럿 및 데이터를 나타내며, l(l=0, 1, ..., 23)은 OFDMA 심볼 인덱스를 나타낸다.Meanwhile, referring back to FIG. 7, the time offset compensator 310 compensates an error according to the time offset by compensating for the phase of the received signal using the time offset estimated by the time offset estimator 210. In this case, although the above-described time offset estimation is performed using the preamble, the time offset compensator 310 compensates the time offset in symbol units. Therefore, the time offset compensation for the pilot and data in the downlink PUSC mode is expressed by Equation 4 below. Where k (k = 0, 1, ..., 1023) represents the subcarrier index, Ф _TO represents the linear phase of the time offset in radians, and r _m represents the pilot and data of the received downlink PUSC mode. Indicate,

Denotes pilot and data of a downlink PUSC mode with time offset compensation, and l ( l = 0, 1, ..., 23) denotes an OFDMA symbol index.

[Equation 4]

In this case, the exponential function for the linear phase (k Ф _TO ) of the time offset may be represented in the form of a trigonometric function, and can be expressed in a complex form as shown in Equation 6 below by using Equation 5. Therefore, Equation 6 can compensate for the phase of the time offset on the complex plane.

[Equation 5]

 [Equation 6]

10 is according to an embodiment of the present invention. It is a block diagram which shows a carrier frequency offset estimation part.

As shown in FIG. 10, the carrier frequency offset estimator 220 includes a second phase difference calculator 221, a second phase difference accumulator 222, a second linear phase calculator 223, and a carrier frequency offset calculator. 224 and parameter converter 225.

The second phase difference calculator 221 receives at least one of the received signals of the first and second channels and the received signals of the third and fourth channels received through the second receiving antenna. At least two pilots included in the signal are extracted to calculate a phase difference according to a carrier frequency offset.

The second phase difference accumulator 222 accumulates the phase difference according to each carrier frequency offset calculated by the second phase difference calculator 221 to generate a phase difference accumulation value. The second phase difference accumulator 222 may estimate a more accurate carrier frequency offset by accumulating the phase difference calculated for a larger number of pilots.

The second linear phase calculator 223 converts the phase difference accumulated value accumulated in the second phase difference accumulator 222 into a linear phase? _CFO according to the carrier frequency offset. The linear phase (Ф _CFO ) with respect to the carrier frequency offset transformed as described above represents an average phase difference with respect to the carrier frequency offset occurring between adjacent symbols having the same subcarrier index (that is, subcarriers having a difference of symbol indices of 1).

The carrier frequency offset calculator 224 converts the linear phase? _CFO according to the carrier frequency offset calculated by the second linear phase calculator 223 into a carrier frequency offset CFO. For example, when an OFDMA symbol interval has 115.2 μs in a downlink frame, a carrier frequency offset (CFO) may be calculated by the following equation.

[Equation 7]

이며,

는 갱신된 계수이다.The parameter converter 225 converts a carrier frequency offset measured in radians into a Hz (Hertz) value and may be calculated by Equation 8 below. Here, f _current is an output of the carrier frequency offset estimator 220 in a frame of a current downlink PUSC mode, f _pre is an output of a carrier frequency offset estimator in a frame of a previous downlink PUSC mode, and Gain is

Is,

Is the updated coefficient.

[Equation 8]

Hereinafter, a method of estimating a carrier frequency offset will be described as a specific example.

First, the pilot extractor (not shown) extracts pilots whose time offset is compensated in the form of Equation 1 described above. This pilot extractor differs from the aforementioned pilot extractor in that it extracts pilots whose time offset is compensated. The second phase difference calculator 221 calculates the phase difference according to the carrier frequency offset using these pilots. Equation (9) below generalizes this and shows a cumulative result of the linear phase according to the carrier frequency offset calculated by the second phase difference calculator 221 using the positional relationship of two pilots whose time offset is compensated. That is, the second phase difference calculator 221 performs a complex products operation on the extracted pilot pair, and the second phase difference accumulator 222 performs the complex product operation on the pilot. Accumulate pairs. The accumulated pilot pairs may be expressed as in Equation 9 below. In this case, P _m (4, l , v) is P _m (0, v), P _m (8, l , v) is P _m (1, v), and P _m (4, l +1, v) is P _m (2, v), P _m (8, l +1, v) is P _m (3, v), and P _m (0, l +2, v) is P _m (0, v), P _m (12, l +2, v) is P _m (1, v), P _m (0, l +3, v) is P _m (2, v), And P _m (12, l + 3, v) is set to P _m (3, v), v represents the cluster index, N _C represents the cumulative number of clusters.

[Equation 9]

Subsequently, the linear phase calculator 223 calculates the linear phase? _CFO according to the carrier frequency offset using Equation 10 below. That is, the linear phase calculator 223 converts the phase difference accumulated value expressed by Equation 9 into a real part as a denominator and an imaginary part as a numerator, and an arc tangent operation is performed on the converted fraction form. After performing the operation, the linear phase (θ _CFO ) according to the carrier frequency offset is calculated by dividing this by the subcarrier index difference (that is, the preamble position difference used in the phase difference calculation).

[Equation 10]

The linear phase (θ _CFO ) according to the carrier frequency offset calculated in this way is converted into a carrier frequency offset (CFO) value in radians in the carrier frequency offset calculator 224, which in turn is Hz in the parameter converter 225. The carrier frequency offset compensation unit 320 is used to compensate for the carrier frequency offset. For example, the carrier frequency offset compensator 320 compensates for an error of an oscillator (eg, a voltage controlled crystal oscillator (VCXO)) using an AFC (Automatic Frequency Controller) or the like based on the estimated carrier frequency offset to adjust the carrier frequency offset. To compensate.

In the present embodiment, in addition to the method of estimating the carrier frequency offset described above, a method of calculating a phase difference according to the carrier frequency offset using a preamble and a pilot transmitted from the same transmitting antenna, and two pairs having the same subcarrier index difference A method of calculating a phase difference according to a carrier frequency offset using a pilot may be used. Of course, in addition to the above-described method, the linear phase according to the carrier frequency offset can be obtained using another combination of the preamble and the pilot.

Hereinafter, a method of calculating weights previously stored in the weight storage unit will be described in detail with reference to FIGS. 11 and 12.

FIG. 11 is a flowchart illustrating a weight calculation method according to an embodiment of the present invention, and FIG. 12 is a diagram illustrating a part of a frame for explaining the weight calculation method of FIG. 11, wherein 14 subcarriers and 16 symbols are used as examples. do. Here, p denotes a pilot, d denotes data, a thick dotted subcluster is a subcluster selected to obtain a channel response, and the channel response of the corresponding subcluster is represented by H.

First, in the downlink PUSC mode, as shown in FIG. 12 for the pilots having the same pattern as in Equation 1, two sub-continuous symbols on the time axis and four subcarriers consecutive on the frequency axis are sub-clusters. partition into clusters). That is, it is assumed that three subclusters are virtually divided in the subcarrier index direction (the subcarrier frequency axis) for each cluster, and one subcluster has the same channel response. Here, l ₀ in Equation 1 represents a symbol index at which a space time code (STC) applied to the present invention starts, and Equation 1- (2) and Equation 1- (4) represent a third index. Note that the pilot patterns for the channel and the fourth channel are shown.

Subsequently, the channel response for the partitioned subcluster is calculated and expressed as a channel response for at least one adjacent subcluster or pilots having the same channel spacing (S510-S520). That is, the channel response of the subcluster selected in FIG. 12 is represented by H. Since H corresponds to an average of adjacent channels, it may be represented by H = 1/4 (h0 + h1 + h2 + h3).

Next, the channel response of the at least one neighboring subcluster or pilot is expressed as a combination of the surrounding pilots and the relative symbol distance ratio between the neighboring pilots based on the subcluster (S530). For example, in FIG. 12, if the channel estimate for the pilot is p (k, j), h0 is p (0,2), p (0, 6), p (4, 4), and p (4, 8). ), H1 can be expressed using p (4, 4) and p (4, 8), h2 can be expressed by p (0, 6), p (0, 10), p (4, 8) and p (4, 12), and h3 can be expressed using p (4, 8) and p (4, 12). Here, k corresponds to a symbol index and j corresponds to a subcarrier index. In addition, the combination of relative symbol distance ratios means that a distance ratio for a pilot close to the subcluster is set to be larger than a distance ratio for a pilot located far from the subcluster. As such, when the channel estimates of the at least one adjacent subcluster or pilot are expressed as a combination of relative symbol distance ratios, h0 is p (0,2), p (0, 6), and p (4, 4). ) And p (4, 8), h0 = 1/16 {1.5 × 3.0 × p (0,2) + 2.5 × 3.0 × p (0, 6)} + 3.5 × p (4 , 4) + 0.5 × p (4, 8)}. In this way, for h0, h1, h2, and h3, respectively, h0 = (1.5 × 3.0) / 16 × p (0, 2) + (2.5 × 3.0) / 16 × p (0, 6) + ( 3.5 × 1.0) / 16 × p (4, 4) + (0.5 × 1.0) / 16 × p (4, 8)}, h1 = 3.5 / 4 × p (4, 4) + 0.5 / 4 × Expressed as p (4, 8), h2 = (1.5 × 3.0) / 16 × p (0, 6) + (2.5 × 3.0) / 16 × p (0, 10) + (3.5 × 1.0) / 16 × Expressed as p (4, 8) + (0.5 × 1.0) / 16 × p (4, 12), expressed as h3 = 3.5 / 4 × p (4, 8) + 0.5 / 4 × p (4, 12) Can be.

Subsequently, the channel response of the subcluster is expressed as a combination of the relative pilot distance relative to the subcluster and the relative symbol distance ratios of the peripheral clusters (S540). For example, in FIG. 12, since H = 1/4 (h0 + h1 + h2 + h3), H = 1/4 {(1.5 × 3.0) / 16} × p (0, 2) + 1/4 {(2.5 × 3.0) / 16 + (1.5 × 3.0) / 16} × p (0, 6) + 1/4 {(3.5 × 1.0) / 16 + 3.5 / 4} × p (4, 4) + 1/4 {( 0.5 × 1.0) / 16 + 0.5 / 4 + (3.5 × 1.0) / 16 + 3.5 / 4} × p (4, 8) + 1/4 {(0.5 × 1.0) / 16 + 0.5 / 4} × p ( 4, 12) +1/4 {(2.5 × 3.0) / 16} × p (0, 10). In this case, the ratio (i.e. weight) multiplied by p (0, 2) before w0, p (0, 6) for w1, p (4, 4) for w2, p (4, 8) for w3, If p (4, 12) is set to w4 and p (0, 10) is set to w5, respectively, then H = w0 × p (0, 2) + w1 × p (0, 6) + w2 × p (4, 4 H is expressed as a combination of the peripheral pilots and the relative symbol distance ratios of the neighboring pilots, such as) + w3 × p (4, 8) + w4 × p (4, 12) + w5 × p (0, 10).

Finally, a relative symbol distance ratio for each of the peripheral pilots is calculated and a coefficient for each of the pilots is stored in the weight storage unit (S550). For example, in Fig. 12, w0 = 1/4 {(1.5 x 3.0) / 16} = 0.0703, w1 = 1/4 {(2.5 x 3.0) / 16 + (3.5 x 1.0) / 16} = 0.1719, w2 = 1/4 {(3.5 × 1.0) / 16 + 3.5 / 4} = 0.2734, w3 = 1/4 {(0.5 × 1.0) / 16 + 0.5 / 4 + (3.5 × 1.0) / 16 + 3.5 / 4} = 0.3125, w4 = 1/4 {(0.5 x 1.0) / 16 + 0.5 / 4} = 0.0391, w5 = 1/4 {(2.5 x 3.0) / 16} = 0.1172. Therefore, the relative symbol distance ratios w0 to w5 for each of the neighboring pilots thus stored are stored in the weight storage unit so that the channel estimator 400 refers to the channel estimation unit.

In the present embodiment, six pilots are used for the channel response (H) of one subcluster, but it is obvious that a plurality of pilots can be represented in the same manner. For example, using four adjacent pilots for the channel response H of one subcluster is as follows. That is, for the channel response H of the subcluster shown in FIG. 12, its neighbor pilots (eg, p (0, 6), p (4, 4), p (4, 8), and p (0, 10). )), H = 1/8 (3.5 × p (0, 6) + 0.5 × p (0, 10) + 1.5 × p (4, 4) + 2.5 × p (0.6)}, each pilot The relative symbol distance ratios for the coefficients for p (0, 6) are 0.4375, the coefficients for p (0, 10) are 0.0625, and the coefficients for p (4, 4) are 0.1875, p (4, 8). The coefficient for may be expressed as 0.3125.

In another embodiment, similar to the above-described method, each cluster is partitioned into subcluster units, and weights are calculated in advance based on at least one of time-axis interpolation, frequency-axis interpolation, and a moving average for the pilots of the subcluster. It may be stored in the storage unit 500.

Summarizing by the method described so far, weights for all channels can be obtained as described in the following Tables 2 to 8.

Tables 2 and 3 below summarize the channel weights for each pilot in the symbols located at the left and right ends. Specifically, Tables 2 and 3 illustrate frames received through the second channel H01 or the fourth channel H11 (the downlink frame includes 27 symbols, one symbol for the preamble and two symbols for the FCH). Since 24 data symbols are used for each subcluster, the weights for each subcluster are summarized for the left and right ends ( l ₀ +5> l and l > l _e -5) of the frame. The channel weight is calculated and stored in the weight storage unit 500. Here, l ₀ represents the start symbol index and l _e represents the last symbol index.

In Table 2, for channel estimates P (y, x) for the pilot, y (y = 0,1) represents the pilot index and x (x = 0, 1, ..., 11) represents the slot symbol (downward). Since the channel is estimated in units of two symbols in the link PUSC mode, the two symbol units are called slot symbols. In this case, the pilot index is set to 0 when the corresponding pilot is located at 6 or more on the basis of a total of 14 subcarriers on the frequency axis, and is set to 1 when it is located at the 6 or less.

[Table 2] Weight (boundary symbol) of channel H01 or H11

On the other hand, to the table _{3, mod (l - l 0} , 4) == 0, when means a case in which a symbol assigned to the l corresponds to the multiple of 4 in the initial l _0, and mod (l - l _0, 4) ~ = 0 means the case of multiple of 2.

On the other hand, Table 4 below shows the weight of each subcluster in a frame received through the second channel H01 or the fourth channel H11, and the weight of each subcluster in the center portion of the frame ( l ₀ +5 < l < l _e -5). ), The channel weights for each pilot are calculated and stored in the weight storage unit 500.

[Table 3] Channel (H01 or H11) weight (boundary symbol)

[Table 4] Channel (H01 or H11) weight

When l ₀ +5 < l < l _e -5, l _e corresponds to the end of the subframe.

Tables 5 to 7 below summarize the weights for each subcluster obtained from the frames received through the first channel H00 or the third channel H10. The weight storage unit calculates the channel weight for each pilot. Save to 500. Other than Tables 5 to 7 are similar to those described in Tables 2 to 4 described above, detailed description thereof will be omitted.

[Table 5] Channel (H00 or H10) weight

when l ₀ +5 < l < l _e -5, l _e is the end of the subframe

[Table 6] Channel (H00 or H10) weight (boundary symbol)

[Table 7] Channel (H00 or H10) weight (boundary symbol)

Referring back to FIG. 7, the channel estimator 400 extracts pilots whose time offsets are compensated for the received signal as shown in Equation 1, and is previously calculated offline and stored in the weight storage 500. The channel estimate is calculated by calculating a channel response for each subcluster based on the following Equation 11 with reference to a weight value such as 7. Where u (u = 0, 1, 2) represents the subcluster index, v (v = 0, 1, ..., 59) represents the cluster index and x (x = 0, 1, ... 11 denotes the slot symbol index as described above, and y (y = 0, 1) denotes the pilot index.

[Equation 11]

For example, in the case of the subcluster 1 of Table 2, the channel response of the subcluster 1 with respect to the symbol l is calculated as in Equation 12 below.

[Equation 12]

For example, if the pilot P _{00 corresponds} to {1.4903 + 0.6939i, 1.4674 + 0.7532i, 1.52 + 0.6329i 1.4585 + 0.6959i}, and the weight W ₀₀ corresponds to {0.5313, 0.1875, 0.1875, 0.0938}, the channel response H ₀₀ is (1.4903 + 0.6939i) x 0.5313 + (1.4674 + 0.7532i) x 0.1875 + (1.52 + 0.6329i) x 0.1875 + (1.4585 + 0.6959i) x 0.0938, so finally, the channel of subcluster 1 for symbol l Corresponds to the response H ₀₀ 1.4887 + 0.6938i.

Meanwhile, the first channel estimator 410 for estimating the first channel H00 and the second channel H01 associated with the first receiving antenna has been described. However, the second channel estimator 420 is similarly described. The third channel H10 and the fourth channel H11 related to the second receiving antenna can be estimated through the above.

A channel estimation method according to an embodiment of the present invention configured as described above will be described with reference to FIG. 13. Hereinafter, a brief description will be made to avoid overlapping with the description of the channel estimation apparatus.

First, in step S810, the pilot is extracted in the form as described above in Equation (1). Next, in step S820, time offset estimation is performed using the extracted pilot (see a time offset estimator). Next, in step S830, the phase error according to the estimated time offset is compensated (see the time offset compensator). Next, in step S840, the carrier frequency offset is estimated (see carrier frequency offset). Subsequently, in step S850, a phase error according to the estimated carrier frequency offset is compensated for. Such compensation may compensate for an error of the oscillator through an AFC (Automatic Frequency Controller). In this case, the compensation step of the carrier frequency offset and the compensation step of the time offset may be performed first. Finally, in step S860, a channel response is performed in subcluster units based on the previously stored weights to estimate the total channel (see the channel estimator and the weight storage unit). On the other hand, although not shown in Figure 13, after the step S860 may be decoded for the space-time coding (STC) and spatial multiplexing (SM), respectively.

Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art to which the present invention pertains can implement the present invention in other specific forms without changing the technical spirit or essential features, The embodiments are to be understood in all respects as illustrative and not restrictive.

In addition, the scope of the present invention is specified by the appended claims rather than the detailed description, and all changes or modifications derived from the meaning and scope of the claims and their equivalent concepts are included in the scope of the present invention. Should be interpreted as

According to the present invention, it is easy to implement a receiving system by estimating a channel based on a weight according to a pilot pattern and reducing non-parallel processing such as interpolation and / or averaging used in channel estimation. And the amount of computation can be reduced.

Claims (24)

A method of estimating a channel in a wireless communication system supporting OFDM or OFDMA, (a) extracting pilots from the received signal to compensate for a phase error according to at least one of a time offset and a carrier frequency offset; (b) obtaining pilot channel estimates for the pilots that are compensated for the phase error; And (c) Subcluster-The subcluster is a partition of a cluster, which is a basic component unit of a Partial Usage of Subchannels (PUSC) subchannel, in the subcarrier index axis direction-pilots designated for each subcluster according to an index and a slot symbol index. Estimating a channel for the received signal in subcluster units by using a product of the pilot channel estimates corresponding to the pre-computed weights for the pilots designated for each subcluster. Channel estimation method. The method of claim 1, wherein the weights are Sub-clusters corresponding to the first three slot symbols, subclusters corresponding to the last three slot symbols, and subclusters corresponding to the remaining slot symbols except for the first and last three slot symbols. A channel estimation method. The method of claim 1 or 2, wherein the weights are: The channel estimation method characterized in that it is determined according to the relative symbol distance ratio of the pilots arranged based on the subcluster. The method of claim 1 or 2, wherein the weights are: And the weight for the pilot located close to the subcluster is greater than the weight for the pilot located far to the subcluster. The method of claim 1 or 2, wherein the received signal, At least one of a received signal of a first channel and a second channel received through a first receive antenna, and a received signal of a third channel and a fourth channel received through a second receive antenna, among the plurality of receive antennas. Channel estimation method. A method of estimating a channel in a wireless communication system supporting OFDM or OFDMA, (a) obtaining a pilot channel estimate for the pilots obtained in each cluster; And (b) Subcluster-The subcluster is a partition of a cluster, which is a basic configuration unit of a Partial Usage of Subchannels (PUSC) subchannel, in the subcarrier index axis direction-pilots designated for each subcluster according to an index and a slot symbol index. Estimating the channel in subcluster units by using a product of pre-computed weights for the pilot channel estimates corresponding to and the pilots designated for each subcluster. The method of claim 6, wherein the cluster is A channel estimating method comprising three subclusters. 7. The method of claim 6, wherein the weights are Sub-clusters corresponding to the first three slot symbols, subclusters corresponding to the last three slot symbols, and subclusters corresponding to the remaining slot symbols except for the first and last three slot symbols. A channel estimation method. 9. The apparatus of claim 8, wherein the weights are The subclusters corresponding to the remaining slot symbols except for the first and last three slot symbols may be divided into subclusters corresponding to odd-numbered slot symbols and subclusters corresponding to even-numbered slot symbols. Channel estimation method. delete An apparatus for estimating a channel in a wireless communication system supporting OFDM or OFDMA, A time offset estimator which receives a received signal and estimates a time offset; A time offset compensator for compensating a phase error according to a time offset using the estimated time offset; Subcluster-The subcluster is a cluster formed by sub-cluster index axis direction of the cluster, which is a basic component unit of the Partial Usage of Subchannels (PUSC) subchannel. A weight storage unit for storing the calculated weights; And At least one of estimating a channel for the received signal in subcluster units by using a product of pilot channel estimates corresponding to pilots designated for each subcluster and the weights of the received signal whose time offset is compensated; And a channel estimator. The method of claim 11, wherein the cluster, OFDMA-based channel estimation apparatus, characterized in that consisting of three subclusters. 12. The method of claim 11, wherein the weights are Sub-clusters corresponding to the first three slot symbols, subclusters corresponding to the last three slot symbols, and subclusters corresponding to the remaining slot symbols except for the first and last three slot symbols. Channel estimation apparatus. The method of claim 11, wherein the weights are: Channel estimation apparatus, characterized in that determined according to the relative symbol distance ratio of the pilots arranged based on the subcluster. The method of claim 11, wherein the weights are: And a weight for a pilot disposed close to the subcluster is greater than a weight for a pilot disposed far to the subcluster. The method of claim 13, wherein the weights are: The subclusters corresponding to the remaining slot symbols except for the first and the last three slot symbols may be divided into subclusters corresponding to odd-numbered slot symbols and subclusters corresponding to even-numbered slot symbols. Channel estimation device. The method of claim 11, wherein the time offset estimator, A first phase difference calculator for calculating a phase difference according to a time offset using at least two preambles or pilots for the received signal; A first phase difference accumulator for accumulating the phase difference; A first linear phase calculator configured to convert the accumulated value of the phase differences into a linear phase according to the time offset; And And a time offset calculator for converting the linear phase into the time offset. The method of claim 11, A carrier frequency offset estimator estimating a carrier frequency offset using the signal in the frequency domain; And And a carrier frequency offset compensator for compensating a phase error according to a carrier frequency offset using the estimated carrier frequency offset. The method of claim 18, wherein the carrier frequency offset estimator, A second phase difference calculator for extracting at least two pilots with respect to the received signal whose time offset is compensated, and calculating a phase difference according to a carrier frequency offset; A second phase difference accumulator for accumulating the phase difference; A second linear phase calculator configured to convert the accumulated value of the phase difference into a linear phase according to the carrier frequency offset; A carrier frequency offset calculator for converting the linear phase to the carrier frequency offset; And And a parameter converter for converting the carrier frequency offset measured in radians in Hertz (Hz). The method according to any one of claims 11 to 19, wherein the received signal is Among the plurality of receiving antennas, at least one of the received signals of the first and second channels received through the first receiving antenna and the received signals of the third and fourth channels received through the second receiving antenna. Channel estimation apparatus, characterized in that. A method of estimating a channel in a wireless communication system supporting OFDM or OFDMA, (a) obtaining a pilot channel estimate for the pilot obtained in the subcarrier frequency allocation unit; And (b) Subcluster-The subcluster is a partition of a cluster, which is a basic configuration unit of a Partial Usage of Subchannels (PUSC) subchannel, in the direction of a subcarrier index axis. Estimating the channel in subcluster units using a product of weights precomputed with and the pilot channel estimates corresponding to pilots designated for each subcluster. . The method of claim 21, wherein the weights are Sub-clusters corresponding to the first three slot symbols, subclusters corresponding to the last three slot symbols, and subclusters corresponding to the remaining slot symbols except for the first and last three slot symbols. A channel estimation method. The method of claim 22, wherein the weights are: The subclusters corresponding to the remaining slot symbols except for the first and last three slot symbols may be divided into subclusters corresponding to odd-numbered slot symbols and subclusters corresponding to even-numbered slot symbols. Channel estimation method. delete

Images