KR20090061564A - Apparatus and method for channel estimation in mimo systems - Google Patents

Abstract

A transmitter, a receiver and a communication method applied to the transmitter for channel estimation in a wireless communications system in the timing offset are provided to estimate a channel without additional performance deterioration by timing offset regardless of the number of antennas. An MIMO(Multiple Input Multiple Output) frequency area signal generator(110) encodes a binary source received from a MAC(Medium Access Control) layer. A MIMO frequency area signal generating unit is the encoded bit in a subcarrier. A pilot inserting unit inserts a pilot into the signal of the branch. A pilot inserting unit inserts the pilot into the signal of each transmit branch. An IFFT(Inverse Fast Fourier Transform) units convert signal into the waveform of the time domain.

Description

Transmitter and receiver for channel estimation in multi-antenna wireless communication system and communication method applied thereto {APPARATUS AND METHOD FOR CHANNEL ESTIMATION IN MIMO SYSTEMS}

The present invention relates to channel estimation in a multi-antenna wireless communication system, and more particularly, to a preamble structure and a channel estimation algorithm designed to be robust to timing offset in channel estimation in a multi-antenna wireless communication system.

The present invention is derived from the research conducted as part of the IT growth engine technology development project of the Ministry of Information and Communication and the Ministry of Information and Telecommunications Research and Development. [Task Management Number: 2006-S-002-02, Title: Development of 3Gbps 4G Wireless LAN System] (IMT-Advanced Radio Transmission Technology with Low Mobility).

In order to overcome the limitations of IMT-2000 and build 4G mobile communication network, much research is being done all over the world. The 4G mobile communication network is not a single communication network but a complex communication network, in which various communication networks such as satellite communication, wireless LAN, and digital broadcasting are integrated.

The International Telecommunication Union (ITU) has determined that fourth-generation wireless communications can deliver data rates of 100 Mbps on the go and 1 Gbps on the go.

Wireless LAN standards currently used include IEEE 802.11b, IEEE 802.11a, IEEE 802.11g, IEEE 802.11n, and the like. Recently, many wireless LANs satisfying the IEEE 802.11b or IEEE 802.11g standard are used. IEEE 802.11g supports a transmission speed of up to 54 Mbps in the 2.4 GHz band. IEEE 802.11n, which is currently under development, will support up to 540Mbps, although it does not meet the 1Gbps rate required by 4G wireless. Key technologies for realizing such high-speed wireless communication include orthogonal frequency division multiplexing (OFDM) and multiple input / output antenna technology (MIMO) (hereinafter referred to as "multi-antennas"). have.

Entity DoCoMo predicts that Orthogonal Frequency Code Division Multiplexing (OFCDM) and OFDM will be used as the technology for 4G wireless communication. OFCDM and OFDM show better performance than wireless communication access based on Direct Sequence-Code Division Multiple Access (DS-CDMA). In the case of using a wideband channel, the performance degradation caused by the multi-path interference generally can be significantly reduced in OFCDM or OFDM.

On the other hand, in order to increase the transmission speed in the limited frequency resources, the frequency efficiency (Spectral Efficiency) should be increased, one method for this is to use a multi-antenna technology. For the use of multiple antenna technology, a transmission scheme such as Bell Laboratories Layered Space Time (BLAST) is used. Maximum Likelihood Detection (MLD) is V-BLAST (Mertical-BLAST) or Minimum Mean Squared Error in terms of Bit Error Rate (BER) or Block Error Rate (BLER). It is already known that it performs better than MMSE. However, since computational complexity increases exponentially according to the modulation scheme and the number of antennas, it is not suitable for implementation in a real communication system. As a solution to reduce the complexity, which is a problem of MLD, QRM-MLD (Maximum Likelihood Detection with QR Decomposition and M-algorithm), an MLD using M-algorithm and QR decomposition, has been proposed. There is still room for improvement.

As described in the above-described ultra-high speed wireless communication technologies, multi-antenna technology and multiple frequency bands are essential for high-speed wireless communication beyond the transmission rate required by the fourth generation wireless communication. Accordingly, there is a need for an efficient channel estimation method suitable for a wireless communication system using multiple antennas and multiple frequency bands. In addition, in the multi-antenna wireless communication system, the influence of the timing offset may not be small. In this case, a channel estimation algorithm is required to improve the channel offset process because it causes a performance degradation in the channel estimation process itself.

Accordingly, an object of the present invention is to provide efficient channel estimation in a multi-antenna wireless communication system by providing a preamble structure suitable for a multi-antenna wireless communication system.

It is another object of the present invention to prevent further performance degradation due to timing offset in a multi-antenna wireless communication system.

The objects of the present invention are not limited to the above-mentioned objects, and other objects and advantages of the present invention which are not mentioned above can be understood by the following description, and will be more clearly understood by the embodiments of the present invention. It will also be appreciated that the objects and advantages of the present invention may be realized by the means and combinations thereof indicated in the claims.

The present invention proposed to achieve the above object is a communication method applied to a transmitter for channel estimation in a multi-antenna wireless communication system having N receive antennas, wherein a predetermined pattern is repeated N times on a time axis. Generating a preamble section including a short training field arranged and configured to compensate for a carrier frequency offset; and transmitting a packet having the preamble section. It is characterized by the thing.

In addition, the present invention provides a communication method applied to a transmitter for channel estimation in a multi-antenna wireless communication system, the method comprising: generating a preamble section for compensating for a phase error due to a sampling frequency offset; And transmitting a packet having a section and a data section.

In addition, the present invention provides a communication method applied to a transmitter for channel estimation in a multi-antenna wireless communication system, comprising: a preamble section including two or more identical long training signal sequences contiguous on a time axis for compensating a carrier frequency offset; And generating a packet having the preamble section.

In addition, the present invention provides a communication method applied to a transmitter for channel estimation in a multi-antenna wireless communication system having M transmit antennas, the method comprising: generating a preamble section including a long training signal sequence and having the preamble section And carrying a packet on a plurality of orthogonal subcarriers, wherein the long training signal sequence corresponding to 1 / M or more of the plurality of subcarriers is assigned a pilot for channel estimation. .

In addition, the present invention is a communication method applied to a receiver for channel estimation in a multi-antenna wireless communication system having N receive antennas, the preamble including a short training signal sequence in which a predetermined pattern is repeatedly arranged N times on a time axis. Receiving a packet having a section, and characterized in that it comprises the step of compensating the carrier frequency offset using at least two of the predetermined pattern.

In addition, the present invention provides a communication method applied to a receiver for channel estimation in a multi-antenna wireless communication system, the method comprising: receiving a packet having a preamble section for compensating for a phase error due to a sampling frequency offset; Compensating for the phase error caused by the sampling frequency offset by using the preamble section of another feature.

In addition, the present invention provides a communication method applied to a receiver for channel estimation in a multi-antenna wireless communication system, the method comprising: receiving a packet having a preamble section including two or more identical long training signal sequences contiguous on a time axis; Compensating the carrier frequency offset using the long training signal sequence is another feature.

In addition, the present invention is a communication method applied to a receiver for channel estimation in a multi-antenna wireless communication system having M transmit antennas and N receive antennas, and includes a plurality of orthogonal intervals having a preamble section including a long training signal sequence. Receiving a packet transmitted on a subcarrier, and estimating information on M × N channels using a pilot assigned to the long training signal sequence corresponding to a subcarrier of 1 / M or more of the plurality of subcarriers. It is another feature to do.

According to the present invention as described above, in a multi-antenna wireless communication system having M transmit antennas and N receive antennas, channel estimation can be efficiently performed on all M × N paths simultaneously. In particular, channel estimation can be performed regardless of the number of antennas without additional performance degradation due to timing offset.

The above objects, features, and advantages will be described in detail with reference to the accompanying drawings, whereby those skilled in the art to which the present invention pertains may easily implement the technical idea of the present invention. In describing the present invention, when it is determined that the detailed description of the known technology related to the present invention may unnecessarily obscure the gist of the present invention, the detailed description will be omitted. Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings.

The new Nomadic Local Area Wireless Access (NoLA) system, recently developed by the Korea Electronics and Telecommunications Research Institute (ETRI), uses multiple antenna technology and OFDM to meet the transmission rates required for 4G wireless communications. The NoLA system employs multi-antenna technology using four or eight transmit and eight receive antennas, with a transmission rate of 3.6 Gbps using three frequency bands of 40 MHz and a total of 120 MHz frequency bands at a carrier frequency of 5 GHz. To ensure.

There are four main features of the NoLA system.

First, the use of multiple antenna technology is mentioned. Multi-antenna technology is a method of obtaining high frequency utilization efficiency through antenna diversity, and is used in systems for high-speed wireless communication such as IEEE 802.11g and IEEE 802.11n. IEEE 802.11n uses four transmit / receive antennas each, but NoLA systems use up to eight transmit / receive antennas each to provide higher transmission rates. If the antenna is used a lot, the transmission speed can be increased, but the complexity of the receiver such as a detector is too high.

The second feature of the NoLA system is a method to solve this problem, using Model Driven Development (MDD) designed by the Korea Electronics and Telecommunications Research Institute's next-generation wireless LAN team. The reception rate is increased by obtaining data and removing interference of data received from multiple antennas through successive interference cancellation (SIC).

A third feature of the NoLA system is the use of Low Density Parity Check (LDPC) codes as the channel coding scheme. LDPC codes can be processed in parallel, making them suitable for high-speed data transmissions of more than 3Gbps.

The fourth feature is the use of the OFDM scheme, which is being actively studied recently, as a transmission scheme suitable for high-speed data transmission in wired and wireless channels. The OFDM method uses a plurality of carriers having mutual orthogonality, thereby improving frequency utilization efficiency. In addition, since the process of demodulating a plurality of carriers at the transmitting and receiving end has the same result as performing an Inverse Discrete Fourier Transform (IDFT) and a Discrete Fourier Transform (DFT), the inverse fast Fourier transform is performed. (Inverse Fast Fourier Transform (IFFT) and Fast Fourier Transform (FFT) can be implemented to operate at high speed. Since the OFDM method is suitable for high-speed data transmission, high-speed wireless LAN such as IEEE 802.11a and HIPELAN / 2, broadband wireless access of IEEE 802.16, digital audio broadcasting, digital terrestrial television broadcasting, Asymmetric Digital Subscriber Line (ADSL), and VDSL ( It is adopted as the standard method of Very High-speed Digital Subscriber Line.

1 is a block diagram of a transmitter 100 used in a multi-antenna wireless communication system according to the present invention. In FIG. 1, the transmitter 100 is briefly illustrated as having two transmit antennas 170 and 175, but may include three or more transmit antennas.

The MIMO frequency domain signal generator 110 encodes a binary source received from a medium access control layer (MAC) layer, maps the encoded bits to a sub-carrier, and maps the encoded bits to a subcarrier. Generate a signal. The signal of each transmission branch is transformed into the waveform of the time domain by the inverse fast Fourier transform (IFFT) unit 130, 135 after the pilot is inserted in the pilot insertion unit 120 or 125. do. The guard interval inserting unit 140 or 145 inserts a guard interval (GI) before every OFDM symbol of the time domain signal, and the MIMO preamble generator 180 includes the preamble section to generate a signal packet. . The generated signal packet is converted into an analog signal by a D / A converter (DAC) 150, 155, and transmitted by the transmit antennas 170, 175 via the transmitters 160, 165. .

2 is a block diagram of a receiver 200 used in a multi-antenna wireless communication system according to the present invention. In FIG. 2, the receiver 200 is briefly illustrated as having two receiving antennas 270 and 275, but may include three or more receiving antennas.

The receiver 200 processes the signals received by the two reception antennas 270 and 275 in the corresponding receivers 260 and 265, respectively. This signal is converted into a digital signal by an analog-to-digital converter (ADC) 250, 255, and then used by the synchronization unit 220, 225 to extract frequency and synchronization information. When the guard interval remover 240 or 245 removes the guard interval from the signal, the fast Fourier transform (FFT) unit 230 or 235 converts the signal from which the guard interval is removed to the frequency domain. The MIMO channel estimator 280 performs channel estimation using a signal in the frequency domain. The channel estimation of the MIMO channel estimator 280 includes carrier frequency offset (CFO) compensation, sampling frequency offset (SFO) compensation, channel information extraction, and the like. The receiver 200 detects the MIMO frequency domain 210 by demapping and decoding the frequency domain signal using the channel information of the MIMO channel estimator 280, and then recovers the binary data.

3 is a flowchart of a communication method applied to a transmitter for channel estimation in a multi-antenna wireless communication system according to the present invention. A communication method applied to the transmitter 100 according to the present invention will be described below with reference to FIG. 1.

The transmitter 100 first generates 310 a preamble section including a short training signal sequence and a long training signal sequence. In this case, the signals and pilots included in the short training signal sequence and the long training signal sequence are configured to be suitable for a multi-antenna wireless communication system as described with reference to FIG. 4. Thereafter, the transmitter 100 transmits 320 a packet having the generated preamble section to a plurality of orthogonal subcarriers. Transmission of the packet 320 is performed through a plurality of transmit antennas, and there may be 256 subcarriers used therein.

4 is a flowchart of a communication method applied to a receiver for channel estimation in a multi-antenna wireless communication system according to the present invention. A communication method applied to the receiver 200 according to the present invention will be described below with reference to FIG. 2.

The packet transmitted by the transmitter 100 is received by the receiver 200 via a wireless channel, and the receiver 200 obtains original data from the received packet. Since the wireless channel changes irregularly in time in a time and frequency domain, the receiver 200 undergoes a channel estimation process of finding out information (attenuation, delay, phase shift, etc.) of the wireless channel. Prior to the channel estimation process, the receiver 200 compensates for the frequency offset of the received packet. The frequency offset may include a carrier frequency offset and a sampling frequency offset.

The carrier frequency offset refers to a phenomenon in which a received signal is distorted due to a receiver incorrectly predicting a carrier frequency. Due to the carrier frequency offset, the phase change and inter-carrier interference (ICI) phenomenon occur. In particular, it is important to compensate for the distorted phase because the phase change can cause severe performance degradation.

The carrier frequency offset has a characteristic of increasing in time. Therefore, when the transmitter sends data as shown in Equation 1, the data received at the receiver due to the carrier frequency offset includes a phase error as shown in Equation 2.

은 반송파 주파수 오프셋을 나타내는 값으로 ppm의 단위를 갖는다. 예를 들어, 반송파 주파수가 5GHz일 때 10ppm의 반송파 주파수 오프셋은 40kHz가 된다. 시뮬레이션 상에서는 시간 축의 데이터에

을 곱해주는 것으로 반송파 주파수 오프셋을 간단히 모델링 할 수 있다.In equation (2)

Is a value representing a carrier frequency offset and has a unit of ppm. For example, when the carrier frequency is 5 GHz, the carrier frequency offset of 10 ppm is 40 kHz. In the simulation, the data on the time axis

By multiplying the carrier frequency offset can be easily modeled.

Since the receiver 200 estimates a channel using a long training signal sequence of the preamble, channel information cannot be used in the preamble section. Therefore, in order to predict and compensate the carrier frequency offset in the preamble section, the preamble section should include data repeated on the time axis. In order to compensate for the distortion caused by the carrier frequency offset, the receiver divides the coarse carrier frequency offset compensation and the fine carrier frequency offset compensation into two offset compensations. The repetition pattern included in the short training signal sequence and the repetitively arranged long training signal sequence are used for two carrier frequency offset compensations.

As shown in FIG. 4, the receiver 200 receives 410 a packet transmitted from the transmitter 100 to perform channel estimation. The receiver 200 compensates 420 the phase error due to the approximate carrier frequency offset using the short training signal sequence of the received packet. The receiver 200 then compensates 430 for the phase error due to the precise carrier frequency offset using the long training signal sequence of the received packet using the long training signal sequence. The phase error due to the sampling frequency offset is compensated for using the preamble section of the received packet (440). After the frequency offsets are compensated (420, 430, 440), the receiver 220 performs channel estimation 450 to find out the information of the wireless channel and continues receiving the signal based on the channel information.

5 is a structural diagram of a packet for channel estimation in a multi-antenna wireless communication system according to the present invention. As shown, a packet according to the present invention can be largely divided into a preamble section and a data section. The preamble interval is used for synchronization, offset compensation, and channel estimation, and the data interval includes data to be transmitted. The preamble section may be divided into a short training field (STF), a long training field (LTF), and a signal field.

The short training signal sequence 510 is used for synchronization and approximate carrier frequency offset compensation. The short training signal sequence 510 includes a repeating pattern, the length of which may be 0.8us each 1/8 of the length of the OFDM symbol.

The long training signal sequence 520, 530, 550, 560 is used for accurate carrier frequency offset compensation and channel estimation. The long training signal strings 520, 530, 550, and 560 are different types by the number of transmitting antennas, and the numbers of LTF-0, LTF-1, ..., LTF- ( N _LTF -1). For example, if the number of transmit antennas is eight, the number N _LTF of the long training signal string type is 8, so the long training signal string includes LTF-0 to LTF-7.

The preamble section according to the present invention includes a short training signal sequence 510, a long training signal sequence repeating two or more times (shown as being repeated twice in FIG. 5), a signal field 540, And the remaining types of long training signal sequences 550 to 560 in order. At this time, every symbol 510, 520, 530, 540, 560, 570, 580 has a predetermined length, it may be configured to have a length of 7.2us as shown. This is the length of the OFDM symbol length 6.4us plus the guard interval 0.8us. Each symbol may include a guard interval of 0.8us at the front of the symbol, and a portion of the symbol is repeatedly arranged by a cyclic prefix.

Meanwhile, the wireless communication system according to the present invention may use a plurality of subcarriers. In this case, 256 subcarriers may be used and 228 out of 256 subcarriers to which data is allocated in the data interval. The remaining subcarriers are assigned pilots or zeros to eliminate interference and change in signal strength due to direct current (DC).

6 and 7 are detailed structural diagrams of the short training signal sequence 510 shown in FIG. 5. As shown in FIG. 6, predetermined patterns D0 to D7 are disposed on the time axis in the short training signal sequence 510. In this case, the number of patterns is equal to the number of receiving antennas or transmitting antennas. FIG. 6 illustrates eight patterns in the case where the number of receive antennas or transmit antennas is eight.

The same pattern is repeatedly arranged in D0 to D7 for the approximate carrier frequency offset compensation. That is, since a signal that is repeated on the time axis is required for the carrier frequency offset compensation as described above, D0 to D7 are all configured with the same bit string. The lengths of D0 to D7 may be 0.8us, which is 1/8 of the OFDM symbol length.

를 시간에 비례하여 증가시키며 수학식 4와 같이 위상 오류를 보상한다. 짧은 훈련 신호열(510)에서 반복되는 패턴의 길이는 0.8us가 될 수 있고, 이는 32개의 데이터에 해당한다.The receiver 200 compensates for the carrier frequency offset by using the last two patterns D6 and D7 among the repeated patterns of the short training signal sequence 510. The receiver 200 is obtained through Equation 3

Increases in proportion to time and compensates for the phase error as shown in Equation 4. The length of the repeated pattern in the short training signal sequence 510 may be 0.8us, which corresponds to 32 data.

Meanwhile, in a communication system using an OFDM scheme, cyclic prefix is used in a guard interval to prevent inter-carrier interference and inter-symbol interference (ISI) caused by delay spread. Thus, as shown in FIG. 7, the short training signal sequence 510 may include a guard interval in front of the guard interval, and a part of the short training signal sequence 510 is repeated in the guard interval. In general, since the last part of the short training signal sequence 510 is repeated in the guard interval, one or more of the repetition patterns for the approximate carrier frequency compensation may be included in the guard interval. In the example shown in FIG. 7, the short training signal sequence 510 is for a system having 8 receiving antennas or transmitting antennas, and includes one repetition pattern in the guard interval, so that all the patterns included in the guard interval are added together. 9 repeating patterns. In this case, the length of the guard interval may be 1/8 of 6.4us, which is an OFDM symbol period.

In addition, the last repetition pattern of the short training signal sequence 510 shown in FIG. 6 or 7 may be arranged in the reverse order of other repetition patterns to facilitate synchronization.

On the frequency axis, 256 subcarriers are allocated to the short training signal sequence 510 as shown in Equation 5.

For example, in a wireless communication system using three frequency bands, the sequence of Equation 5 is assigned to an intermediate frequency band among three frequency bands represented as upper, middle, and lower. The upper frequency band is given a phase difference of +90 degrees to the sequence of Equation 5, and the lower frequency band is assigned a subcarrier of -90 degrees to 256 subcarriers, respectively. This phase difference serves to reduce the peak to average power ratio (PAPR).

The signal allocated to the short training signal sequence 510 may be represented by Equation 6 on the time axis.

는 고속 푸리에 변환의 사이즈를 나타내고,

는 시간 축에서 짧은 훈련 신호열(510)의 윈도우 함수(windowing function)를 의미한다.

와

는 각각 부반송파 간의 간격과 스트림의 개수를 나타낸다. 또,

스트림의 순환 전치에 의해

가 정의된다.In equation (6)

Represents the size of the fast Fourier transform,

Denotes a windowing function of the short training signal sequence 510 on the time axis.

Wow

Denotes the interval between subcarriers and the number of streams, respectively. In addition,

By the circular transpose of the stream.

Is defined.

8 is a waveform diagram of the time response of the short training signal sequence 510 according to the present invention. When the short training signal sequence 510 is configured as shown in Figs. 5 to 7, the same pattern is repeated every 0.8us. Therefore, the time response of the short training signal sequence 510 shows the same waveform repeated every 0.8us as shown in FIG.

9 is a detailed structural diagram of a preamble section shown in FIG. 5.

For accurate carrier frequency offset estimation, a signal that is repeated on the time axis in the preamble section is required, and two or more identical long training signal sequences 520 and 530 are continuously arranged after the short training signal sequence 510 as shown. . The long training signal sequence 520, 530, 540, 550 to 560 is also used for channel estimation for decoding the signal.

The signal field 540 is positioned after the repeated long training signal sequences 520 and 530, and the remaining long training signal sequences LTF1 to LTF7 are disposed thereafter. In the example illustrated in FIG. 9, since the number of receive antennas is 8, the total length of the long training signal sequence is LTF0 to LTF7 as described above.

를 시간에 비례하여 증가시키며 반송파 주파수 오프셋에 의한 위상 오류를 보상한다. 반복되는 긴 훈련 신호열의 길이는 OFDM 심볼 길이인 6.4us가 될 수 있고, 이는 256개의 데이터에 해당한다.The receiver uses the short training signal sequence to approximate the carrier frequency offset compensation. However, since the repeated patterns included in the short training signal sequence have a short length, the offset compensation through the receiver may be vulnerable to noise. Therefore, accurate carrier frequency offset compensation is performed using long training signal sequences 520 and 530 having a long length of repeated signals. As in the approximate carrier frequency offset compensation method,

Is increased in time and compensates for the phase error caused by the carrier frequency offset. The length of the repeated long training signal sequence may be 6.4us, which is an OFDM symbol length, corresponding to 256 data.

On the frequency axis, 256 subcarriers are allocated to the long training signal sequence as shown in Equation (8).

A signal allocated to a long training signal sequence on the time axis may be represented by Equation 9. Equation 9 corresponds to a signal of a wireless communication system using three frequency bands as in Equation 6, and a phase difference of +90 degrees is given to the sequence of Equation 9 in the upper frequency band, and -90 degrees in the lower frequency band. 256 subcarriers are allocated to each other with a phase difference. This phase difference serves to reduce the maximum power to average power ratio.

는 시간 축에서의 긴 훈련 신호열에 대한 윈도우 함수를 나타내고,

는 순환 전치의 길이를 나타낸다. 또한,

는 수학식 10과 같이 나타낼 수 있다.In Equation 9,

Represents the window function for the long training signal sequence on the time axis,

Represents the length of the cyclic prefix. Also,

May be expressed as in Equation 10.

10 is a layout view of a long training signal sequence according to time and eight transmit antennas in a multi-antenna wireless communication system according to an embodiment of the present invention. In FIG. 10, one Mod represents one long training signal string symbol. As shown in FIG. 10, the first transmission stream (Tx Stream) 1010 transmitted through the first transmit antenna has a long training signal sequence Mod0 repeated twice, a signal field SIG Mod0, and other than Mod0. It contains a series of long training signal sequences (Mod1 to Mod7). The second transmission stream 1020 transmitted through the second transmission antenna to the eighth transmission stream 1080 transmitted through the eighth transmission antenna have the same structure as the first transmission stream 1010, but each increasing the index by one. Contains a signal that is repeated.

11 is a layout view of a long training signal sequence according to time and four transmit antennas in a multi-antenna wireless communication system according to an embodiment of the present invention. In the case of using four transmitting antennas, there are four types of long training signal strings, and accordingly, Mod0 to Mod3 are arranged in a structure as described above. That is, the first transport stream 1110 includes two repeated long training signal sequences Mod0, a signal field SIG Mod0, and a series of long training signal sequences Mod1 to Mod3 other than Mod0. The remaining transport stream contains a signal that is repeated by increasing the index by one.

12 is an explanatory diagram for channel estimation in a multi-antenna wireless communication system according to an embodiment of the present invention.

For example, in a wireless communication system using eight transmit antennas and eight receive antennas, only one eighth of a plurality of subcarriers should be allocated to a pilot and 0 should be allocated to the remaining subcarriers for accurate channel estimation without interference between subcarriers. . As shown, when the signal received by the first receiving antenna 1210 is received, the pilots transmitted by the first transmitting antenna 1211 to the eighth transmitting antenna 1281 are sequentially arranged, and then again the first transmitting antenna The pilot sent by 1211 is placed. As shown in FIG. 12, when a pilot is transmitted through eight transmit antennas 1211, 1221, 1231, 1241, 1251, 1261, 1271, and 1281, the receiver receives a pilot and uses the same to obtain an impulse response of a channel. Only information on channels can be inferred. This is because the receiver already knows the transmitted pilot. In this case, a total of 64 channels may exist by multiplying the number of transmit antennas and the number of receive antennas. If multiple subcarriers are used, 64 channel information can be obtained for each subcarrier by channel estimation in this manner. As another example, in a wireless communication system using four transmit antennas, only one quarter of a plurality of subcarriers are allocated a pilot and zeros are assigned to the remaining subcarriers.

Hereinafter, the sampling frequency offset will be described with reference to FIGS. 13 and 14.

In FIG. 13, the solid line represents the sampling period in the transmitter, and the dotted line represents the sampling period in the receiver. The sampling frequency offset refers to the difference when the sampling period at the transmitter and the sampling period at the receiver are different as shown in FIG. Sampling frequency offset reduces the signal strength, changes the phase in the frequency domain, and causes interference between carriers.

The sampling frequency offset is generated using a sink function as shown in FIG. If there is no sampling frequency offset, the data is multiplied by the center value of one sync function, and the data is positioned at the zero point at the remaining sampling frequency offset, so that signal distortion does not occur. However, if a sampling frequency offset occurs and slightly deviates from the center of the sink function, it is affected by other data and distortion occurs.

FIG. 15 is a constellation diagram of a phase error caused by a sampling frequency offset according to an embodiment of the present invention. FIG.

When the phase error in the frequency domain due to the sampling frequency offset is calculated according to Equation 11, the maximum phase error of the data symbol in the frequency domain due to the sampling frequency offset is expressed by Equation 12. Where n = 128, m = 2880, and N = 256.

As shown in Equation 12, when the sampling frequency offset is 30ppm, the phase error is about 15.5 degrees, and thus the performance degradation of the wireless communication system due to the phase error may be severe. Therefore, compensation must be made properly. Since the phase error is already present when the receiver estimates the channel using the preamble section, the phase error obtained using the pilot of the data section cannot be an accurate value. Therefore, it is necessary to track and compensate the sampling frequency offset from the preamble section. That is, as illustrated in FIG. 4, there are three methods for compensating for an error due to a sampling frequency offset using a preamble section, as follows.

Firstly, a repeating long training signal sequence disposed in the preamble section for the carrier frequency offset compensation is further used at the end of the preamble section to calculate a phase error. Secondly, a pilot is allocated to an unused subcarrier section and a phase error is obtained through this. Third, a method of allocating a pilot for phase error of a preamble section at the same position as the pilot used in the data section.

Each of the three methods described above has advantages and disadvantages, but the third method has an advantage of obtaining a phase error without changing the structure of the preamble. The first method uses less than 256 subcarriers to calculate phase error, but the third method is relatively vulnerable to noise because only four pilots are used. The receiver can use all the information obtained from the signals received through the eight receive antennas, and reduce the influence of noise by adjusting the bandwidth of the loop filter.

16 is an explanatory diagram for phase error calculation using a pilot according to an embodiment of the present invention.

According to the third method described above, the receiver calculates a phase error due to a sampling frequency offset using a pilot arranged in common in the preamble section and the data section. As with the method of compensating the carrier frequency offset, the same pilot is allocated to all symbol intervals because there is no information on the channel in the preamble interval. However, a symbol repeated on the time axis is used to compensate for the carrier frequency offset, but a symbol repeated on the frequency axis is required to compensate for the sampling frequency offset.

As shown in Equation 13, the receiver converts the received data into a fast Fourier transform into a frequency domain and calculates a phase error using a pilot repeated on the frequency axis.

Four pilots are allocated to the data section, and using the four pilots arranged at such a position in the preamble section, both the phase error due to the carrier frequency offset and the phase error due to the sampling frequency offset can be obtained. The phase error due to the carrier frequency offset appears the same in all frequency domains, and the phase error due to the sampling frequency offset appears in proportion to the frequency. The phase error due to the carrier frequency offset is shown in Equation 14, and the phase slope due to the sampling frequency offset is shown in Equation 15.

를 이용하면 두 가지 위상 오류를 모두 구할 수 있다.Therefore, the phase difference obtained from four pilots

Can be used to find both phase errors.

As described above, if the pilot for the phase error is placed in the preamble section at the same position as the pilot used in the data section, since four pilots are used, the phase error may not be accurate due to noise and interference between carriers. have. Therefore, a loop filter as shown in FIG. 17 is used to compensate the phase more accurately. 17 illustrates a second order loop filter 1700 designed as one embodiment of the present invention.

The loop filter 1700 functions as a low pass filter (LPF) using delay units (memory) 1710 and 1720. In this case, the bandwidth may be adjusted by changing the loop coefficients 1 and 2 by adjusting the proportional gain and the integral gain of the loop filter 1700.

18 is a graph of delay power spectrum according to an embodiment of the present invention.

The channel of the wireless communication system according to the present invention may be a multi-path Rayleigh distributed channel. The characteristics of multipath channels are determined by the root mean square delay spread (RMS Delay Spread). The characteristics of multipath channels assume exponential decaying.

Equation 16 is an expression for expressing the delay power spectrum.

은 경로에 따른 평균 전력을 나타낸다.

는 지수 감쇠의 상수값으로 제곱 평균 제곱근 지연 확산과 샘플링 주파수에 의한 함수로 나타낼 수 있다.

은 정규화(normalization)를 위한 상수 값이다.

은 경로의 개수로서, 코히런스 대역폭(Coherence Band Width)를 제곱 평균 제곱근 지연 확산의 5배에 대한 역수로 정의하여 표현한 것이다. 도 18은 50ns의 제곱 평균 제곱근 지연 확산에 샘플링 주파수가 40MHz, 경로가 11개 일 때의 지연 전력 스펙트럼을 도시한 것이다. In equation (16)

Represents the average power along the path.

Is a constant value of exponential decay and can be expressed as a function of the root mean square delay spread and the sampling frequency.

Is a constant value for normalization.

Is the number of paths, which is defined as the coherence bandwidth (Coherence Band Width) is defined as the inverse of 5 times the root mean square delay spread. FIG. 18 shows a delay power spectrum when a 50 nm square root mean square delay spread has a sampling frequency of 40 MHz and 11 paths.

19 to 21 are graphs for measuring channel estimation performance according to an embodiment of the present invention, and FIG. 19 is a graph measuring bit error rates with respect to signal to noise ratio (SNR), and FIG. FIG. 21 is a graph measuring a block error rate with respect to a signal-to-noise ratio. FIG. 21 is a graph measuring a packet error rate (PER) with respect to a signal-to-noise ratio.

에 의해 원래의 데이터를 변조한다. The channel estimation method used in this simulation is Modular-8. Scrambler is a generator function for security at the transmitting end.

Modulate the original data.

A scrambler such as Equation 17 may be used, and the same scrambler may be used for the transmitter and the receiver. A non-zero pseudo random code may be used for the initialization of the scrambler, and a bit string of 1011101 may be used as an initial state.

The simulation is implemented by implementing a multi-antenna wireless communication system using four or eight transmit antennas and eight receive antennas under the conditions shown in Table 1 and Equation 18. As a result of the simulation, even when there is a sampling frequency offset of 30 ppm, as shown in FIGS. 19 to 21, the difference is less than 0.3 dB compared to the case where there is no sampling frequency offset, resulting in a remarkable performance improvement.

The apparatus and system described above may be implemented in hardware, software, or a combination thereof. In a hardware implementation, the modules used for channel estimation include one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing units (DSPDs), programmable logic devices (PLDs), field programmable gates. Array (FPGA), processor, controller, micro-controller, microprocessor, other electronic unit designed to perform the functions described herein, or a combination thereof. The software may be implemented through a module that performs the functions described herein. Software code may be stored in memory units and executed by a processor. The memory unit may be implemented inside or outside the processor, in which case the memory may be coupled to the processor via various known means.

On the other hand, the method of the present invention as described above can be written in a computer program. And the code and code segments constituting the program can be easily inferred by a computer programmer in the art. In addition, the written program is stored in a computer-readable recording medium (information storage medium), and read and executed by a computer to implement the method of the present invention. The recording medium includes all types of computer-readable recording media (tangible media such as CD and DVD as well as intangible media such as carrier waves).

The present invention described above is capable of various substitutions, modifications, and changes without departing from the technical spirit of the present invention for those skilled in the art to which the present invention pertains. It is not limited by the drawings.

1 is a block diagram of a transmitter used in a multi-antenna wireless communication system according to the present invention.

2 is a block diagram of a receiver used in a multi-antenna wireless communication system according to the present invention.

3 is a flowchart of a communication method applied to a transmitter for channel estimation in a multi-antenna wireless communication system according to the present invention.

4 is a flowchart of a communication method applied to a receiver for channel estimation in a multi-antenna wireless communication system according to the present invention.

5 is a structural diagram of a packet for channel estimation in a multi-antenna wireless communication system according to the present invention.

FIG. 6 is a detailed structural diagram of the short training signal sequence shown in FIG. 5.

FIG. 7 is a detailed structural diagram of a protection section that may be included in the short training signal sequence shown in FIG. 5.

8 is a waveform diagram of a time response of a short training signal sequence according to the present invention.

9 is a detailed structural diagram of a preamble section shown in FIG. 5.

10 is a layout view of eight transmit antennas and a long training signal sequence over time in a multi-antenna wireless communication system according to an embodiment of the present invention.

11 is a layout view of four transmission antennas and a long training signal sequence over time in a multi-antenna wireless communication system according to an embodiment of the present invention.

12 is an explanatory diagram for channel estimation in a multi-antenna wireless communication system according to an embodiment of the present invention.

13 is an explanatory diagram of a sampling frequency offset according to an embodiment of the present invention.

14 is an explanatory diagram of a sampling frequency offset according to an embodiment of the present invention.

15 is a constellation diagram of a phase error caused by a sampling frequency offset according to an embodiment of the present invention.

16 is an explanatory diagram for phase error calculation using a pilot according to an embodiment of the present invention.

17 is a block diagram of a loop filter according to an embodiment of the present invention.

18 is a graph of delay power spectrum according to an embodiment of the present invention.

19 is a graph measuring channel estimation performance according to an embodiment of the present invention.

20 is a graph measuring channel estimation performance according to an embodiment of the present invention.

21 is a graph measuring channel estimation performance according to an embodiment of the present invention.

Claims (12)

A communication method applied to a transmitter for channel estimation in a multi-antenna wireless communication system having N receive antennas, Generating a preamble section including a short training field configured by repeating a predetermined pattern N times on a time axis and used to compensate for a carrier frequency offset; Transmitting a packet having the preamble interval Communication method of the transmitter comprising a. The method of claim 1, The short training signal sequence has a guard interval before or after it, And the predetermined pattern is further repeated one or more times in the guard period. The method according to claim 1 or 2, The predetermined pattern is composed of a bit string arranged on the time axis, The last pattern of the pattern of the short training signal sequence and the guard interval is arranged in the reverse order of the predetermined pattern (reverse). The method according to claim 1 or 2, And the preamble section further comprises at least two identical Long Training Fields continuous in the time axis. The method according to claim 1 or 4, The packet transmission step A packet having the preamble section is transmitted on a plurality of orthogonal subcarriers, And a pilot for channel estimation is allocated to the long training signal sequence corresponding to 1 / M or more of the plurality of subcarriers. A communication method applied to a transmitter for channel estimation in a multi-antenna wireless communication system, Generating a preamble section for compensating for a phase error due to a sampling frequency offset; Transmitting a packet having the preamble section and a data section Communication method of the transmitter comprising a. The method of claim 6, The preamble section includes a first long training signal sequence for compensating for a phase error due to a sampling frequency offset and a second long training signal sequence for compensating a carrier frequency offset, And the first and second long training signal sequences are identical to each other. The method of claim 6, The preamble section includes a long training signal sequence, The transmitting step is carried by transmitting the packet on a plurality of orthogonal subcarriers, A pilot for channel estimation is allocated to the long training signal sequence corresponding to 1 / M or more of the plurality of subcarriers. And a pilot for compensating for a phase error due to a sampling frequency offset is allocated to the long training signal sequence corresponding to the remaining subcarriers. The method of claim 6, The preamble section includes a first pilot for compensating for a phase error due to a sampling frequency offset, The data interval includes a second pilot for channel estimation, And the first pilot is assigned to the same location as the second pilot. A communication method applied to a transmitter for channel estimation in a multi-antenna wireless communication system, Generating a preamble interval comprising two or more identical long training signal sequences contiguous on the time axis for compensating the carrier frequency offset; Transmitting a packet having the preamble interval Communication method of the transmitter comprising a. A communication method applied to a transmitter for channel estimation in a multi-antenna wireless communication system having M transmit antennas, Generating a preamble section including a long training signal sequence; And transmitting the packet having the preamble section on a plurality of orthogonal subcarriers, And a pilot for channel estimation is allocated to the long training signal sequence corresponding to 1 / M or more of the plurality of subcarriers. A communication method applied to a receiver for channel estimation in a multi-antenna wireless communication system having N reception antennas, Receiving a packet having a preamble section including a short training signal sequence in which a predetermined pattern is repeatedly arranged N times on a time axis; Compensating a carrier frequency offset using at least two of the predetermined patterns; Communication method of a receiver comprising a.

Images