KR20070021917A - Virtual multiple antenna method for OFDM system and OFDM-based cellular system - Google Patents

Abstract

The present invention relates to an orthogonal frequency division multiplexing (OFDM) system and a virtual multiple antenna method in an OFDM based cellular system. The method includes grouping subchannels in the frequency domain of an OFDM symbol to generate at least one group of G subchannels; And considering the G subchannels belonging to the group as multiple channels of a multiple antenna scheme, and virtually applying the multiple antenna scheme to transmission and reception of the OFDM symbol. According to the present invention, an interference signal due to interference between cells and interference by multiple users in a cell can be effectively reduced without physically using multiple antennas, and the effects of spatial division multiple access (SDMA) can be reduced. You can get it.

OFDM, cellular, virtual smart antenna, virtual SDMA, in-cell multi-user interference cancellation, inter-cell interference cancellation, VSR

Description

Virtual multiple antenna method for OFDM system / OFDM-based cellular system {Virtual multiple antenna method for OFDM system and OFDM-based cellular system}

1 is a block diagram illustrating a general OFDM system.

2A and 2B are diagrams for explaining the concept of the conventional MIMO scheme and the virtual MIMO scheme of the present invention, respectively.

3 is a block diagram illustrating an OFDM system to which a virtual multiple antenna scheme is applied according to an embodiment of the present invention.

4A-4C illustrate a grouping scheme used in the present invention.

5 is a flowchart illustrating a virtual multiple antenna scheme of an OFDM system according to an embodiment of the present invention.

6A and 6B are flowcharts illustrating a specific example of the step of applying the virtual multiple antenna scheme of FIG. 5.

7A and 7B illustrate a virtual multiple antenna method in the uplink of an OFDM based cellular system according to an embodiment of the present invention.

8A and 8B illustrate a virtual multiple antenna method in an uplink of an OFDM based cellular system according to an embodiment of the present invention.

9A to 9C are diagrams for describing a virtual multiple antenna method in an uplink of an OFDM based cellular system according to another embodiment of the present invention.

10A and 10B illustrate a virtual multiple antenna method in the uplink of an OFDM based cellular system according to another embodiment of the present invention.

11A to 11C are diagrams for describing a virtual multiple antenna method in downlink of an OFDM based cellular system according to an embodiment of the present invention.

12A and 12B illustrate a virtual multiple antenna method in downlink of an OFDM based cellular system according to an embodiment of the present invention.

13A and 13B illustrate a virtual multiple antenna method in downlink of an OFDM-based cellular system according to an embodiment of the present invention.

14A and 14B are block diagrams illustrating an OFDM cellular system using a VSR scheme and a virtual multiple antenna scheme according to an embodiment of the present invention.

The present invention relates to an OFDM system and an OFDM-based cellular system. More particularly, the present invention relates to a method for virtually applying a multi-antenna technique to an OFDM system and an OFDM-based cellular system to remove an interference signal or to obtain an effect of SDMA. .

The OFDM scheme removes intersymbol interference by inserting a cyclic prefix (CP) longer than an impulse response of a channel into a guard interval existing between adjacent OFDM symbols.A single tap equalizer (single tap) is received at a receiving end. Easily compensate for fading distortion using an equalizer, and use the Inverse Fast Fourier Transform (IFFT) and Fast Fourier Transform (FFT) to perform the fast transform / demodulation process It has the advantage of being able to.

These advantages make it possible for high-speed data transmission systems such as Digital Audio Broadcasting (DAB), Digital Video Broadcasting (DVB), Digital Terrestrial Television Broadcasting (DTTB), and Wireless LAN (Local). OFDM wireless communication systems such as Area Network (LAN), Broadband Wireless Access (IEEE 802.16), etc. are being developed. Recently, such OFDM wireless communication systems are considered as the core technology of the next generation of mobile communication. have.

Meanwhile, a multi-antenna technique, which is a reliable high-speed data transmission and a transmission scheme capable of increasing the capacity of a system, has been actively studied. The multi-antenna technique is a method of using multiple antennas in a transceiver, and examples thereof include an SDMA technique and a smart antenna technique. The SDMA technique is a multiple antenna technique that allows multiple people to simultaneously use the same frequency channel in the same cell in a cellular system. Similarly, the smart antenna technique is a multiple antenna technique that can effectively remove interference and increase signal reliability by forming a beam only in a desired direction using an antenna array structure. However, since the multi-antenna technique is an applicable method when the number of transmit / receive antennas is physically increased, there is a disadvantage in that hardware complexity increases due to the increase in the number of transmit / receive antennas.

In addition, another technique for removing interference is a multi-carrier code division scheme. This method is a technique for eliminating interference of multiple users or neighbor cells by using a single antenna at a transceiver, and eliminating interference using an orthogonal code. However, the multi-carrier code division scheme has a disadvantage in that orthogonality is broken due to channel characteristics and synchronization errors of a band in which an orthogonal code is carried.

An object of the present invention is to provide a virtual multi-antenna method in an OFDM system and an OFDM-based cellular system that can effectively reduce an interference signal without physically using a multi-antenna and obtain an SDMA effect. .

According to an aspect of the present invention, there is provided a virtual multiple antenna method in an OFDM system according to the present invention, including: (a) generating at least one group consisting of G subchannels by grouping subchannels in a frequency domain of an OFDM symbol; And (b) considering the G subchannels belonging to the group as multiple channels of a multiple antenna scheme, and virtually applying the multiple antenna scheme to transmission and reception of the OFDM symbol.

Preferably, the multi-antenna technique applied in step (b) includes a spatial multiplexed division access (SDMA) technique, a MIMO detection technique, and a smart antenna technique.

Preferably, the step (b) comprises: (b1) the transmitting device estimating a channel response for the G subchannels between the Nc receiving devices and the transmitting device; And (b2) the transmitting device precoding an Nc symbol to be transmitted to the G subchannels based on the channel matrix composed of the estimated channel response to present an effect on the channel; And (b3) the transmitting device transmitting the OFDM symbol including the precoded Nc symbols to the Nc receiving devices.

Preferably, in the step (b), (b1) each of the Nc transmitting devices multiplies the symbols carried on the G subchannels by G weights for randomizing channels among the transmitting devices, and includes the multiplied symbols. Transmitting an OFDM symbol; (b2) a receiving device estimating channel responses for the G subchannels between the Nc transmitting devices and the receiving devices, and multiplying the estimated channel response values by a weight used by a target transmitter; And (b3) detecting, by the receiving device, a signal transmitted from the target transmitter by applying a virtual multiple antenna scheme based on the multiplied channel response value.

Preferably, step (a) includes generating the group using any one of a comb type, a cluster type, and a random type.

In order to achieve the above technical problem, the virtual multi-antenna method in the uplink of the OFDM-based cellular system according to the present invention includes (a) Nc terminals grouping subchannels in the frequency domain of the OFDM symbol by the same grouping method. Generating at least one group of G subchannels; (b) the Nc terminal generating an OFDM symbol by mapping a symbol to a subchannel belonging to the group, and transmitting the generated OFDM symbol to a base station; And (c) detecting, by the base station, a signal by applying the most multi-antenna technique that considers the received signals of the G subchannels as signals received from the G virtual antennas.

Preferably, the step (b) includes the Nc terminal multiplying the symbols with G weights for randomizing a downlink channel, and mapping the multiplied symbols to the G subchannels. In the step (c), the base station estimates the channel responses of the G subchannels, and then multiplies the G weights by the estimated channel response values, respectively, based on the multiplied channel response values. It includes the step of applying.

Preferably, each of the weights has the same magnitude and has a value of M-PSK series.

Preferably, step (a) includes generating the group using any one of a comb type, a cluster type, and a random type.

Advantageously, said virtual multiple antenna technique is a virtual SDMA technique. Preferably, step (c) includes removing interference signals and detecting signals of a desired user using signal detection techniques including ZF, MMSE, SIC, PIC, and ML. Advantageously, step (c) comprises simultaneously detecting signals of multiple users using signal detection techniques including ZF, MMSE, SIC, PIC, and ML.

Preferably, the Nc terminals include a terminal located in a cell boundary region and a terminal located in an adjacent cell boundary region.

Advantageously, said virtual multiple antenna technique is a virtual smart antenna technique. Preferably, step (c) comprises: (c1) estimating an autocorrelation matrix of a vector consisting of the received signals of the G subchannels; (c2) estimating a symbol timing offset between terminals; And (c2) detecting the signal using the weight of the virtual smart antenna calculated based on the estimated autocorrelation matrix and the symbol timing offset. Preferably, the step (c1) includes estimating the autocorrelation matrix using a characteristic in which the influence of the symbol timing offset on the received signal is represented by phase rotation between adjacent subcarriers. Preferably, the step (c2) includes the step of estimating the symbol timing offset by virtually applying a smart antenna technique including MUSIC and ESPRIT to the estimated autocorrelation matrix. Preferably, step (c) includes calculating a weight of the virtual smart antenna using a training signal based technique including LMS, RLS and SMI techniques. Preferably, step (c) includes calculating a weight of the virtual smart antenna using a symbol timing offset based technique including null-steering and MVDR. Advantageously, step (c) includes removing interference signals and detecting a signal of a desired user using a virtual smart antenna technique. Advantageously, step (c) comprises simultaneously detecting signals of multiple users using a virtual smart antenna technique.

Preferably, the Nc terminals include a terminal located in a cell boundary region and a terminal located in an adjacent cell boundary region.

In order to achieve the above technical problem, the virtual multi-antenna method in the downlink of the OFDM-based cellular system according to the present invention includes (a) a base station grouping at least one group of G subchannels by grouping subchannels of an OFDM symbol Generating; (b) the base station calculating a channel response matrix for the G subchannels between Nc terminals and the base station; (c) the base station presenting an effect on a downlink channel by precoding Nc symbols based on the channel response matrix; And (d) the base station mapping the precoded G symbols to respective subchannels and transmitting the OFDM symbols as OFDM symbols.

Preferably, the method further comprises the step of (e) each terminal detecting the signal by summing the received signals of the G sub-channels.

Preferably, step (a) includes generating the group using any one of a comb type, a cluster type, and a random type.

Preferably, the precoding comprises ZF, orthogonalization, DPC and THP.

Preferably, the step (b) includes the step of calculating, by the base station, the channel response matrix by multiplying the estimated channel response values of the G subchannels of each terminal by G weights for randomizing the downlink channel. do.

Preferably, each of the weights has the same magnitude and has a value of M-PSK series.

In order to achieve the above technical problem, the virtual multi-antenna method in the downlink of the OFDM-based cellular system according to the present invention includes (a) N _c base stations in the same grouping method, grouping the sub-channels of the OFDM symbol G group Generating at least one group consisting of subchannels; (b) generating, by each base station, an OFDM symbol by mapping a transmission symbol to the G subchannels, and transmitting the generated OFDM symbol; And (c) detecting a desired signal by applying a virtual multi-antenna technique in which the UE regards the received signals of the G subchannels as signals received from the G virtual antennas.

Preferably, step (a) includes generating the group using any one of a comb type, a cluster type, and a random type.

Advantageously, said virtual multiple antenna technique is a virtual SDMA technique. Preferably, step (c) includes removing interference signals using a signal detection technique including ZF, MMSE, SIC, PIC, and ML, and detecting a signal transmitted from a cell to which the terminal belongs. . Preferably, step (c) includes simultaneously detecting a signal of an adjacent cell together with a signal of a target cell using a signal detection technique including ZF, MMSE, SIC, PIC, and ML.

 Preferably, the step (b) includes the step of each base station to multiply each symbol by G weights for randomizing a downlink channel to map the multiplied symbols to the G sub-channels, and (c In step), each UE estimates channel responses of the G subchannels, and then multiplies the estimated channel response values by the G weights, respectively, to perform a virtual multiple antenna scheme based on the multiplied channel response values. Applying steps. Preferably, each of the weights has the same magnitude and has a value of M-PSK series.

Advantageously, said virtual multiple antenna technique is a virtual smart antenna technique. Preferably, step (c) comprises: (c1) estimating an autocorrelation matrix of a vector consisting of the received signals of the G subchannels; (c2) estimating a symbol offset between adjacent cells; And (c2) detecting the signal using the weight of the virtual smart antenna calculated based on the estimated autocorrelation matrix and the symbol timing offset. Preferably, the step (c1) includes estimating the autocorrelation matrix using a characteristic in which the influence of the symbol timing offset on the received signal is represented by phase rotation between adjacent subcarriers. Preferably, the step (c2) includes the step of estimating the symbol timing offset by virtually applying a smart antenna technique including MUSIC and ESPRIT to the estimated autocorrelation matrix. Preferably, step (c) includes calculating a weight of the virtual smart antenna using a training signal based technique including LMS, RLS and SMI techniques. Preferably, step (c) includes calculating a weight of the virtual smart antenna using a symbol timing offset based technique including null-steering and MVDR.

Advantageously, step (c) includes removing interference signals of adjacent cells using a virtual smart antenna technique and detecting a desired signal. Advantageously, step (c) includes detecting a signal of a neighbor cell and a signal of a target cell simultaneously using a virtual smart antenna technique.

Hereinafter, a method and an apparatus according to the present invention will be described in detail with reference to the accompanying drawings.

1 is a block diagram illustrating a general OFDM system. Referring to FIG. 1, a transmitting end of an OFDM system includes a transmission symbol generator 100, an OFDM symbol generator 110, and an OFDM symbol transmitter 120, and a receiver includes an OFDM symbol receiver 130 and an OFDM symbol recovery. It consists of a grandfather 140 and a data detector 150. However, the structure of FIG. 1 and the name of FIG. 1 are labeled for convenience, and the present invention is applicable to all OFDM systems, and the OFDM system to which the present invention is applied is shown in FIG. It is not limited.

The transmission symbol generator 100 generates transmission symbols to be carried on a channel (subchannel) according to each subcarrier. The data source 101 generates data, and the error correction code (ECC) encoder 102 performs encoding to make the generated data robust to the wireless channel. The interleaver 103 interleaves the encoded data to increase the effect of the ECC. The modulator 104 generates N transmission symbols by performing modulation to convert the interleaved data into an M-ary phase shift keying (M-PSK) signal, an M-ary quadrature amplitude modulation (M-QAM) signal, or the like. .

The OFDM symbol generator 110 generates an OFDM symbol composed of subchannels containing N transmission symbols. The S / P 111 bundles the serially input transmission symbols into N units and outputs them in parallel. The IFFT 112 performs IFFT conversion on the N transmitted symbols, and the P / S 112 provides the conversion result to the guard interval inserting unit 114 in series. The guard interval inserting section 114 inserts a guard interval including a cyclic prefix into the output signal of the P / S 112. Meanwhile, the OFDM symbol generator 110 may further include a pre-compensator 115, and the presenter 115 may influence the output of the guard interval inserter 114. Performs signal processing to compensate in advance.

The OFDM symbol transmitter 120 transmits the generated OFDM symbol on a wireless channel. The D / A 121 converts the OFDM symbol in the digital form into an analog signal for transmission, and the LPF 122 performs low pass filtering on the analog signal to remove the influence on the adjacent band. The HPA 123 amplifies the output signal of the LPF 122, and although the OFDM symbol transmitter 120 is not shown in FIG. 1, the HPA 123 is provided with at least one transmission antenna to provide the amplified output signal with a wireless channel. Discharge into the phase.

The OFDM symbol receiver 130 receives a signal corresponding to an OFDM symbol that has undergone a radio channel through at least one reception antenna (not shown) and provides the signal to the OFDM symbol demodulator 140. The LPF 131 performs low pass filtering on the received signal to remove the influence of the adjacent band, and the A / D 132 digitizes the analog output signal of the LPF 131.

The OFDM symbol demodulator 140 detects the N transmission symbols carried on each subchannel of the OFDM symbol from the output signal of the OFDM symbol receiver 132 and provides them to the data detector 150. The synchronization & channel estimation 146 obtains synchronization and frequency synchronization of the OFDM symbol based on the signal of each step and estimates the channel response of each subchannel. The guard interval removing unit 141 removes the guard interval from the output signal of the OFDM symbol receiver 130 based on the obtained synchronization. The S / P 142 provides the FFT 143 in parallel with the OFDM symbols having the guard intervals removed in series. The equalizer 144 performs equalizing processing on the received signal carried on each subchannel based on the estimated channel response value. The P / S 145 provides the equalized N signals in series to the data detector 150.

The data detector 150 detects data transmitted from the transmitter based on the output signal of the OFDM symbol demodulator 140. The demodulator 151 demodulates the output signal of the OFDM symbol demodulator 140 and provides demodulated data to the deinterleaver 152. The deinterleaver 152 performs deinterleaving, which is an inverse process of the interleaver 103, on the demodulated data. The ECC decoder 153 decodes the deinterleaved data, and the data sink 154 stores / consumes the data detected through the decoding.

That is, the transmitting end of the OFDM system multiplexes the parallelized N transmission symbols with different subcarrier frequencies, and adds each multiplexed transmission symbol and transmits the sum. Here, since N parallelized transmission symbols form one OFDM symbol and each of the N subcarriers of the OFDM symbol are orthogonal to each other, the subcarrier channels (subchannels) do not affect each other. Therefore, compared with the conventional single carrier transmission scheme, since the symbol period can be increased by the number of subchannels N while maintaining the same symbol rate, inter-symbol interference due to multipath fading can be reduced.

2A and 2B are diagrams for explaining a concept of a conventional Multi Input Multi Output (MIMO) technique and a virtual MIMO technique of the present invention, respectively.

Referring to FIG. 2A, the conventional MIMO technique is a technique for physically multiple antennas at a receiver to detect a desired signal or to remove an interference signal. In other words, the multi-antenna device includes G reception antennas, so that h ₁₁ , h ₁₂ ,... , h _1G has multiple channels with channel response in time domain, h ₂₁ , h ₂₂ ,... , h _2G has multiple channels with a time-domain channel response. The multi-antenna device may perform a function of detecting both the signal of the target device and the signal of the interfering device or removing the signal of the interfering device from the received signal by using the characteristic that the correlation between the two multiple channels is not large.

On the other hand, referring to Figure 2b, the virtual MIMO scheme of the present invention regards the G sub-channels in the frequency domain of the OFDM system as a multi-channel used for the application of the multi-antenna technique, even if one antenna is used That's how it works. That is, although the target device, the interfering device, and the receiving device all have one antenna, the same transmission symbol is transmitted on G subchannels included in the OFDM symbol and transmitted. That is, H ₁₁ , H ₁₂ ,... , Multiple channels in the frequency domain H _1G and H ₂₁ , H ₂₂ ,. Multiple channels in the frequency domain, H _2G , are represented by h ₁₁ , h ₁₂ ,... , the multi-channel having a channel response of h _1G and h _21, h _22, ... , h corresponds to multiple channels with a channel response of _2G .

3 is a block diagram illustrating an OFDM system to which a virtual multiple antenna scheme is applied according to an embodiment of the present invention. Referring to FIG. 3, one device A 300 and N _c (= 2) devices B 310 and 320 are included. Here, the device A 300 and the device B (310, 320) is a device capable of OFDM-based transmission or reception, each having one antenna. The device A 300 and the device B 310 and 320 generate M (= 2) groups by grouping subchannels consisting of N (= 6) subcarriers of the OFDM system in order to apply the virtual multiple antenna scheme. That is, each group consists of G (= 3) subchannels. A group with group index m = 0 consists of three subchannels corresponding to subcarrier indexes 0, 2 and 4, and a group with group index m = 1 consists of three subchannels corresponding to subcarrier indexes 1, 3 and 5. do. This grouping process is generally performed before starting communication in an OFDM system including device A 300 and device B 310, 320.

4A to 4C are diagrams illustrating a grouping method used in the present invention, FIG. 4A is a grouping method according to resource allocation of a cluster type, and FIG. 4B is a grouping method according to resource allocation of a comb type. 4c illustrates a grouping method according to random resource allocation. In the cluster type, each group is composed of G adjacent subcarriers, and in the comb type, each group is composed of G subcarriers separated by M. In the case of a random type, each group is G having random spacing of subcarriers. It consists of two subcarriers. 4A to 4C are merely examples for describing the grouping method, the grouping method applied to the present invention is not necessarily limited thereto.

5 is a flowchart illustrating a virtual multiple antenna scheme of an OFDM system according to an embodiment of the present invention. A virtual multiple antenna scheme according to the present embodiment will be described with reference to FIG. 3.

In step S500, all transmitting / receiving devices to which the present invention is applied generate at least one group consisting of G subchannels by grouping subchannels in a frequency domain of an OFDM symbol in the same grouping method. According to the embodiment of FIG. 3, device A 300 and device B 310, 320 perform this grouping process. However, as described above, the grouping process is conceptual. When the OFDM system according to the present invention is constructed, all the devices may have completed the grouping process. Alternatively, after one device completes the grouping process, the other device may provide information on the grouping method to the remaining devices so that the other device may perform the grouping process based on this information.

In step S510, the transmitting / receiving device to which the present invention is applied regards the G subchannels belonging to the group as a multichannel of the multi-antenna technique, and virtually applies the multi-antenna technique to the transmission and reception of the OFDM symbol. That is, when the device B (310, 320) is a transmitting device, the device A (300) as the receiving device detects a desired signal by applying a multi-antenna technique virtually. Here, as an example of the desired signal, referring to FIG. 1, a transmission symbol that is an output signal of the transmission symbol generator 100 of the device B 310 or the device B 320 may be used. In addition, when the device A 300 is a transmitting device, a multi-antenna technique is virtually applied to the signal transmission so that the receiving device, the device B 310 or 320, can easily detect a desired signal. Here, examples of the multi-antenna technique applied may include, but are not limited to, an SDMA technique, a MIMO detection technique, a smart antenna technique, and the like.

FIG. 6A is a flowchart illustrating a specific example of step S510 of FIG. 5, and illustrates a process of detecting a desired signal from a signal mixed with signals transmitted from N _c transmitting devices by applying a virtual multiple antenna scheme. That is, the device B (310, 320) of Fig. 3 is a transmitting device, the device A (300) is a receiving device. Here, the applied multi-antenna technique may include a virtual SDMA technique and a virtual smart antenna technique. Referring to FIGS. 6A and 3, operation S510 includes the following steps.

In operation S600, the device B 310 or 320 generates an OFDM symbol by mapping transmission symbols to the G subchannels belonging to the group, and transmits the generated OFDM symbol to the device A 300. Device B 310 has transmit symbols X _{i = 1, m = 0} , X _{i = 1, m = 1} and device B 320 has transmit symbols X _{i = 2, m = 0} , X _{i = 2 The} description is based on the assumption that _{m = 1} . The device B 310 transmits OFDM symbols arranged as X ₁₀ , X ₁₁ , X ₁₀ , X ₁₁ , X ₁₀ , X ₁₁ in the frequency domain, and the device B 320 is X ₂₀ , X in the frequency domain. _{It is} to transmit OFDM symbols arranged as ₂₁ , X ₂₀ , X ₂₁ , X ₂₀ , X ₂₁ .

In operation S602, the device A 300 receives a signal given by Equation 1 and considers the received signals of the G subchannels included in the received signals as signals received from the G virtual antennas. Is applied to detect the desired signal. Here, the received signals of the G subchannels are represented by equation (2). In addition, examples of the desired signal may include a transmission symbol of X _10, X ₁₁ of the machine B (310) transmitting symbols of X _10, X ₁₁ or the device B (320) of the two devices B (310, 320) All transmission symbols may be detected simultaneously. Meanwhile, examples of the applied multi-antenna technique may include a virtual SDMA technique and a virtual smart antenna technique, which will be described in detail later.

Here, k denotes a subcarrier index or subchannel index, N denotes the number of subcarriers used for transmission of a transmission symbol in an OFDM system, and Y (k) denotes a received signal of a kth subchannel. i is an index of a transmitting device having a value of 1 to N _c , and referring to FIG. 3, a transmitting device corresponding to i = 1 is a device B 310, and a transmitting device corresponding to i = 2 is a device B 320. to be. X _i (k) denotes a transmission symbol of the kth subchannel of the i th transmission device. In addition, H _i (k) represents a channel frequency response of the k-th subchannel formed between the i-th transmitting device and the base station. N _w (k) represents Additive White Gaussian Noise (AWGN) with a mean of zero and a variance of σ ² .

Representation of the m-th group of received signals in a vector format using Equation 1 is as shown in Equation 2.

는 m번째 그룹의 채널 계수 행렬이다. X(m)은 송신 신호 벡터이며, N _w(m)는 잡음 벡터이다. Y(m),

, X(m), N _w(m)은 각각 수학식 3 내지 수학식 6으로 표현된다.Here, m represents an index of a resource composed of G subchannels, that is, a group index, and has a relationship of M = N / G. Y (m) is the received signal vector of the m-th group.

Is the channel coefficient matrix of the m-th group. X (m) is a transmission signal vector and N _w (m) is a noise vector. Y (m),

, X (m), N _w (m) is represented by the equations (3) to (6), respectively.

Here, J _mg represents a subchannel index corresponding to the g subchannel of the m th group, and is determined according to the grouping method as shown in FIGS. 4A to 4C. Equations 7, 8 and 9 represent subchannel indexes according to a clustering type, a comb type, and a random type grouping method, respectively.

Here, rand (m, g) has a value between 0 and M-1 and represents a randomly selected value.

의 i번째 열 벡터인

는 i 번째 송신 기기와 수신 기기 사이에 형성된 m번째 그룹의 채널 응답을 나타내며, 수학식 10으로 주어진다.In equation (4)

I vector of columns

Denotes a channel response of the m-th group formed between the i-th transmitting device and the receiving device, and is given by Equation (10).

에 해당되는 채널과

에 해당되는 채널이 서로 독립적이라고 가정하면, 가상 SDMA 기법을 적용하여 기기 B(210)의 전송 심볼 및 기기 B(212)의 전송 심볼을 검출할 수 있다.In equation (4)

Which channel corresponds to

Assuming that channels corresponding to are independent of each other, a transmission symbol of the device B 210 and a transmission symbol of the device B 212 may be detected by applying a virtual SDMA technique.

에 해당되는 채널과

에 해당되는 채널이 서로 상관도 (correlation)가 클수록 가상 다중 안테나 기법으로 얻을 수 있는 이득이 적을 수 있기 때문에, 본 발명의 다른 일 실시예에 따르면, 채널을 랜덤화하는 과정을 도 6a에 더 부가할 수 있다. 이 경우, S600 단계는 상기 기기 B(310, 320)가 각각 전 송 심볼에 채널을 랜덤화하는 가중치를 각각 곱하고, 상기 곱하여진 심볼을 상기 G개의 부채널에 맵핑하는 과정을 더 포함한다. 즉, 기기 B(210)는 c₁(0)X₁₀, c₁(0)X₁₁, c₁(1)X₁₀, c₁(1)X₁₁, c₁(2)X₁₀, c₁(2)X₁₁이 주파수 영역상에서 배치된 OFDM 심볼을 송신하고, 기기 B(212)는 c₂(0)X₂₀, c₂(0)X₂₁, c₂(1)X₂₀, c₂(1)X₂₁, c₂(2)X₂₀, c₂(2)X₂₁이 주파수 영역 상에서 배치된 OFDM 심볼을 송신한다. 여기서, c_i(0), c_i(1),…,c_i(G-1)은 랜덤화하는 가중치로서 자세한 내용은 후술한다. 그리고, S602 단계에서는 기기 A(300)가

,

을 추정한 후, 상기 추정된

,

의 각 엘리먼트에 상기 가중치를 각각 곱하고, 상기 곱하여진 채널 응답 값을 기초로 가상 다중 안테나 기법을 적용하는 것이다. 즉, m=0인 경우, c₁(0)H₁(0), c₁(1)H₁(2),c₁(2)H₁(4), c₂(0)H₂(0), c₂(1)H₂(2),c₂(2)H₂(4)로 이루어진 행렬을 가상 다중 안테나 기법에 사용되는 채널 응답 행렬로 간주하여 신호를 검출한다.Meanwhile,

Which channel corresponds to

According to another embodiment of the present invention, a process of randomizing a channel is further added to FIG. 6A because the channels corresponding to each other have a higher correlation with each other. can do. In this case, step S600 may further include multiplying each transmission symbol by a weight for randomizing a channel and mapping the multiplied symbols to the G subchannels. That is, device B (210) is c ₁ (0) X ₁₀ , c ₁ (0) X ₁₁ , c ₁ (1) X ₁₀ , c ₁ (1) X ₁₁ , c ₁ (2) X ₁₀ , c ₁ (2) X ₁₁ transmits OFDM symbols arranged in the frequency domain, and device B 212 transmits c ₂ (0) X ₂₀ , c ₂ (0) X ₂₁ , c ₂ (1) X ₂₀ , c ₂ ( 1) X ₂₁ , c ₂ (2) X ₂₀ , c ₂ (2) X ₂₁ transmits OFDM symbols arranged on the frequency domain. Where c _i (0), c _i (1),... , c _i (G-1) is a weight to be randomized, and will be described later in detail. In operation S602, the device A 300

,

After estimating, the estimated

,

Multiply each of the elements by the weights and apply a virtual multiple antenna scheme based on the multiplied channel response values. That is, when m = 0, c ₁ (0) H ₁ (0), c ₁ (1) H ₁ (2), c ₁ (2) H ₁ (4), c ₂ (0) H ₂ (0 ), a signal consisting of c ₂ (1) H ₂ (2) and c ₂ (2) H ₂ (4) is regarded as a channel response matrix used in the virtual multi-antenna technique.

FIG. 6B is a flowchart illustrating another specific example of step S510 of FIG. 5, in which one transmitting device transmits a signal by applying a multi-antenna technique to N _c receiving devices, and each receiving device detects the signal. . That is, the device A 300 of FIG. 3 is a transmitting device, and the devices B 310 and 320 are receiving devices. Here, an example of a multi-antenna technique applied may be a pre-coding technique for pre-compensating a channel. Referring to FIGS. 6B and 3, operation S510 includes the following steps.

,

으로 이루어진 채널 응답 행렬

을 산출한다. 산출 과정은 다양한 채널 추정 알고리듬이 있으나, 이에 대한 설명은 생략한다.In operation S610, the device A 300 determines that the transmitting device

,

Channel response matrix

To calculate. The calculation process has various channel estimation algorithms, but a description thereof will be omitted.

을 기초로, N_c(=2)개의 심볼에 대해 채널에 의한 영향을 선보상하는 프리코딩을 수행하여 (G=3)개의 심볼을 생성한다. 프리코딩 과정은 수학식 11로 표현될 수 있다.In operation S612, the device A 300 performs a channel response matrix.

Based on the above, precoding that represents the effect of the channel on N _c (= 2) symbols is performed to generate (G = 3) symbols. The precoding process may be represented by Equation 11.

,

으로 이루어진 채널 응답 행렬

을 기초로 산출되며, 자세한 내용은 후술한다. X(m)은, N_c×1인 벡터로서, m번째 그룹에 속하는 기기 B(310, 320)에 전송할 데이터 중 m번째 그룹에 해당되는 전송 심볼 X_1m, X_2m로 이루어진다. 즉, 수학식 11을 통하여 G(=3)개의 프리코딩된 심볼로 이루어진 Z(m)이 산출된다.Where F (m) is a precoding matrix of G × N _c , and basically

,

Channel response matrix

It is calculated based on the details, which will be described later. X (m) is a N _c × 1 vector, and is composed of transmission symbols X _1m and X _2m corresponding to the m th group among data to be transmitted to the device B (310, 320) belonging to the m th group. That is, Z (m) consisting of G (= 3) precoded symbols is calculated through Equation (11).

In operation S614, the device A 300 maps each of the precoded symbols to a subchannel to generate an OFDM symbol, and transmits the generated OFDM symbols to the device B 310 or 320.

In operation S616, the device B 310 detects the transmission symbol X _1m by summing up the received signals of the G subchannels included in the received OFDM symbol, and the device B 320 performs the same signal processing to perform the X _{2m operation.} Is detected.

에 해당되는 채널과

에 해당되는 채널이 서로 상관도가 클수록 다중 안테나 기법에 의해 얻을 수 있는 효과가 적을 수 있기 때문에, 본 발명의 다른 일 실시예에 따르면 채널을 랜덤화 하는 과정을 도 6b에 더 부가할 수 있다. 이 경우, S610 단계는 상기 기기 A(300)가 F(m)을 산출함에 있어서,

을 그대로 이용하는 것이 아니라,

의 각 엘리먼트에 채널을 랜덤화하는 가중치를 곱한 후, 곱한 결과를 기초로 F(m)을 산출한다. 나머지 과정은 상술한 바와 동일하다.Meanwhile,

Which channel corresponds to

Since the corresponding channels have a higher correlation with each other, the effects that can be obtained by the multi-antenna technique may be less. Therefore, according to another embodiment of the present invention, the process of randomizing the channels may be further added to FIG. 6B. In this case, in step S610, when the device A 300 calculates F (m),

Not just use

After multiplying each element of by a weight that randomizes the channel, F (m) is calculated based on the multiplication result. The rest of the procedure is the same as described above.

Such a concept of the present invention can be applied to an OFDM-based cellular system. In this case, a target base station, a neighbor base station, a target terminal, a terminal at a cell boundary, a terminal of a neighbor cell, and the like correspond to the above-described device A or device B. In this specification, seven exemplary embodiments applied to an OFDM-based cellular system will be described in detail to explain the basic concept of the present invention. On the other hand, the present invention can be implemented in various forms in addition to the seven embodiments to be described later using a variety of multi-antenna technique is not limited to the embodiments described herein.

These seven embodiments are roughly classified into a virtual MIMO scheme performed in a base station of uplink / downlink and a virtual MIMO technique performed in a downlink terminal.

The first embodiment is a virtual SDMA technique at the uplink base station, and the second embodiment is a virtual MIMO detection technique that removes interference of multiple users at the cell boundary at the uplink base station. The third embodiment is a virtual smart antenna scheme in an uplink base station. The fourth embodiment is a virtual smart antenna technique for eliminating interference of multiple users at the cell boundary in the uplink base station, and the fifth embodiment is a virtual SDMA technique using an appearance at the downlink base station. All five embodiments are virtual MIMO techniques applicable to a base station and are used to reduce multi-user interference in a cell or multi-user interference between cells in a cell boundary region or to perform SDMA. The sixth embodiment is the virtual MIMO detection technique in the downlink terminal, and the seventh embodiment is the virtual smart antenna technique in the downlink terminal. The sixth embodiment and the seventh embodiment are virtual MIMO techniques applicable to the terminal, and are used to reduce inter-cell interference in a cellular system having a frequency reuse factor of 1.

In addition, in such embodiments, a process of multiplying each element of the channel response matrix by a weight independently modifying the channel response may be obtained to obtain better performance when the virtual SDMA technique or the virtual MIMO detection technique is applied. The embodiment is further described. That is, by applying the multi-antenna technique based on the weighted channel response matrix, a better effect can be obtained.

7A is a flowchart illustrating a virtual multi-antenna method in an uplink of an OFDM-based cellular system according to an embodiment of the present invention, and illustrates an embodiment of detecting a signal of multiple users in a cell by applying a virtual SDMA technique. FIG. 7B shows a frequency domain resource allocation structure for explaining the embodiment of FIG. 7A.

In step S700, N _c terminals in the cell group at least one subchannel on the frequency domain of the OFDM symbol to generate at least one group consisting of G subchannels. Here, Nc and G form a relationship of Nc≤G.

In step S702, each terminal generates an OFDM symbol by mapping the transmission symbols to the G sub-channels belonging to the group, and transmits the generated OFDM symbol to the base station.

In step S704, the base station receives a signal given by the equation (1), by applying a virtual SDMA technique that considers the received signal of the G sub-channel included in the received signal as a signal received from the G virtual antenna, A signal transmitted from a terminal of a user is detected. On the other hand, since the channel response value is required in the process of performing the virtual SDMA technique, step S704 includes a channel estimation process.

, X(m), 및 N _w(m)는 각각 수학식 3 내지 수학식 6과 같이 표현된다. 또한, 수학식 4에서

의 i번째 열 벡 터인

는 i번째 단말과 기지국 사이에 형성된 m번째 그룹의 채널 응답을 나타내며 수학식 8과 같이 주어진다. 수학식 4에서

벡터가 서로 독립적이라고 가정하면, S704 단계에서

행렬을 이용하여, 선형 검출(linear detection) 기법 또는 비선형 검출 기법을 이용하여 원하는 신호를 검출할 수 있다. 여기서, 선형 검출 기법의 예로는 최소 자승(Least Square : LS) 기법, 최소 평균 자승 오차(Minimum Mean Square Error : MMSE) 기법을 들 수 있으며, 비선형 검출 기법의 예로는 병렬 간섭 제거(Parallel Interference Cancellation : PIC) 기법, 순차 간섭 제거(Successive Interference Cancellation : SIC) 기법 및 최우(Maximum Likelihood : ML) 기법을 들 수 있으나, 반드시 이에 한정되는 것은 아니다.In Equation 1, X ₁ (k) and X _i (k) of i ≠ 1 denote transmission symbols transmitted from a terminal of a desired user and transmission symbols transmitted from a terminal of an i-th interference user, respectively. In addition, H _i (k) represents the frequency response of the channel formed between the i-th terminal and the base station, N _w (k) represents AWGN having a mean of 0 and a variance of σ ² . Meanwhile, the received signals of the G subchannels are represented by Equation 2, Y (m),

, X (m), and N _w (m) are each expressed as in Equations 3 to 6. Also, in Equation 4

Is the i th column vector of

Denotes the channel response of the m-th group formed between the i-th terminal and the base station and is given by Equation (8). In equation (4)

Assuming the vectors are independent of each other, in step S704

Using a matrix, a desired signal can be detected using a linear detection technique or a nonlinear detection technique. Here, examples of the linear detection technique may include a least square (LS) technique and a minimum mean square error (MMSE) technique, and an example of the nonlinear detection technique may include parallel interference cancellation (PSA). PIC), Successive Interference Cancellation (SIC), and Maximum Likelihood (ML), but are not limited thereto.

Equation 12 shows a linear detection method.

Here, the weighting matrix W in the m-th Resource (m) in the case of the least squares method to the N _c × G matrix for a minimum mean square error method with equation (13) are expressed respectively by the following equation 14.

'는 의사 역(pseudo inverse) 행렬을 의미하고, 기호 '

'는 켤레 전치(conjugate transpose) 행렬을 나타내며, σ²은 송신전력이 1일 때의 잡음 전력을 나타낸다. LS 기법은 잡음을 고려하지 않고 채널 응답의 의사 역행렬을 구하여 간섭신호를 널링(Nulling)하는 기법이며, MMSE 기법은 잡음까지 고려하여 복조신호가 최대 신호대 잡음비를 갖도록 하는 기법이다. 수학식 13, 14와 같이 구한 가중치 행렬 W(m) 중 i번째 행 벡터는 i번째 사용자의 신호를 남기고 다른 사용자의 신호를 제거하는 정보를 갖는다. 그러므로, i번째 사용자의 신호가 검출 대상인 신호라면, i번째 행벡터와 수신신호 벡터 Y(m)의 곱을 통하여 타겟 사용자 신호의 검출이 가능하며, 간섭 사용자의 신호도 동시에 검출해야 하는 경우에는 수학식 13, 14에서 구해진 가중치 행렬과 수신신호 벡터 Y(m)의 곱을 통하여 N_c개의 사용자 신호 모두를 검출할 수도 있다.Where symbol '

'Means pseudo inverse matrix, symbol'

'Denotes the conjugate transpose matrix, and σ ² denotes the noise power when the transmit power is 1. The LS technique is a technique of nulling an interference signal by obtaining a pseudo inverse of channel response without considering noise, and the MMSE technique is a technique in which a demodulated signal has a maximum signal-to-noise ratio in consideration of noise. The i th row vector of the weight matrices W (m) obtained as shown in Equations 13 and 14 has information for leaving the i th user's signal and removing another user's signal. Therefore, if the signal of the i-th user is a signal to be detected, the target user signal can be detected through the product of the i-th row vector and the received signal vector Y (m), , N _c may be detected by multiplying the weight matrix obtained at 14 and the received signal vector Y (m).

In addition, the signal detection method in the case of the ML technique which is an example of the nonlinear detection technique is given by Equation 15.

은 i번째 사용자가 m번째 그룹의 부채널에 실을 수 있는 전송 심볼을 나타낸다.here,

Denotes a transmission symbol that an i-th user can carry on a subchannel of an m-th group.

On the other hand, the received signal of the base station is composed of the sum of the signals transmitted from the various terminals via each channel, and the symbol timing offset is generated between the signals passing through the base station. Since the phase rotation component is estimated with the channel response in the channel estimation step, no performance degradation due to symbol timing offset occurs.

8A is a flowchart illustrating a virtual multi-antenna method in an uplink of an OFDM-based cellular system according to an embodiment of the present invention, which detects signals of multiple users in a cell boundary region by applying a virtual MIMO detection technique. Indicates. FIG. 8B shows a frequency domain resource allocation structure for explaining the embodiment of FIG. 8A. In the present embodiment, an interference signal of a neighbor cell and a signal of a user in a cell are separated from a signal received by a base station using a virtual MIMO detection technique. Referring to FIG. 8A, the process is as follows.

In step S800, the N _c terminals group at least one subchannel on the frequency domain of the OFDM symbol to generate at least one group consisting of G subchannels.

In step S802, each terminal generates an OFDM symbol by mapping the transmission symbols to the G subchannels belonging to the group, and transmits the generated OFDM symbol.

In step S804, the base station receives a signal given by the equation (1), by applying a virtual SDMA technique that regards the received signal of the G sub-channel included in the received signal as a signal received from the G virtual antenna, Detect the signal transmitted from the terminal. Since the process of applying the virtual SDMA technique requires a channel response value, step S804 includes a channel estimation process. The difference between the embodiment of FIG. 8A and the embodiment of FIG. 7A is that N _c terminals in a group are terminals located in a cell boundary region, up to a channel response of a terminal using resources belonging to the same group allocated to an adjacent cell. It is a point to be estimated, and since other mathematical parts are the same as the embodiment of FIG. 7A, the description is omitted here.

9A is a flowchart illustrating a virtual multi-antenna method in an uplink of an OFDM-based cellular system according to another embodiment of the present invention, and illustrates a method of detecting a signal of multiple users in a cell by applying a virtual smart antenna technique. . FIG. 9B shows a frequency domain resource allocation structure for explaining the embodiment of FIG. 9A. In the present embodiment, a base station eliminates interference by multiple users or simultaneously demodulates signals of multiple users. The resource allocation method for applying the present embodiment is preferably a cluster type resource allocation method. This embodiment is as follows with reference to FIG. 9A.

In step S900, N _c terminals in the cell group at least one subchannel on the frequency domain of the OFDM symbol to generate at least one group consisting of G subchannels.

In step S902, each terminal generates an OFDM symbol by mapping the transmission symbols to the G sub-channels belonging to the group, and transmits the generated OFDM symbol to the base station.

를 검출한다.In step S904, the base station receives a signal given by the equation (1), by applying a virtual smart antenna technique that regards the received signal of the G sub-channel included in the received signal as a signal received from the G virtual antenna, Detect the signal transmitted from the desired terminal. The signal detection calculates the weight vector w (m) of the virtual smart antenna scheme to obtain a desired signal as shown in Equation 16.

Detect.

Examples of the weight vector calculation method of step S904 may include a training signal based technique and a symbol timing offset based technique. The training signal-based technique does not require symbol timing offset estimation, but has a disadvantage in that overhead is generated according to the training signal. Examples of such training signal-based techniques include the least mean square (LMS) technique, the recursive least square (RLS) technique, and the sample matrix inversion (SMI) technique. The symbol timing offset-based technique requires estimation of the symbol timing offset. Examples of the symbol timing offset-based technique include a null-steering technique and a minimum variation distortionless response (MVDR) technique. Meanwhile, in the present specification, a description of the training signal-based technique is omitted for convenience and some techniques of the symbol timing offset-based technique will be described.

9C is a flow diagram embodying step S904 in which a symbol timing offset based technique is used.

는 수학식 17로 표현된다.In step S910, the base station estimates the autocorrelation matrix R (m) of the vector consisting of the received signals of the G subchannels. I-th column vector of Equation 4 when using a cluster type resource allocation method

Is expressed by equation (17).

Here, H _i (J _mg ) represents a channel response value when no symbol timing offset occurs, and a (δ _i ) represents a steering vector formed by the symbol timing offset δ _i . Ⓧ also represents a Hadamard product. Assuming a quasi-static fading channel, the autocorrelation matrix R (m) of the m-th group of received signal vectors is given by Equation 18.

Here, R _x (m) represents the autocorrelation matrix of the transmission signal vector. If there is no correlation between the transmission signals of multiple users, and the transmission signals of each user have the same average power σ _x ² , then R _x (m) is σ _x ² I _Nc and R (m) is simplified as

That is, the estimation method according to Equation 19 is a method of estimating the autocorrelation matrix by using a characteristic in which an influence of a symbol timing offset on a received signal is represented by phase rotation between adjacent subcarriers.

In step S912, the base station estimates the symbol timing offset between the terminals. As a symbol timing offset estimation method, there is a method of applying various Direction of Arrival (DoA) estimation techniques applicable to existing smart antenna techniques, such as an Estimation of Signal Parameters via Rotational Invariance Technique (ESPRIT) technique. However, in the present specification, for convenience, an estimation method based on a multiple signal classification (MUSIC) method among subspace based methods will be described, and a description of the remaining estimation method will be omitted.

The MUSIC method estimates the symbol timing offset by finding the position of the peak value by changing the symbol timing offset δ in the MUSIC spectrum as shown in Equation 19 when G is larger than N _c .

Here, in the case where there is no correlation between user signals, the G × (GN _c ) matrix V (m) consisting of the eigenvectors of the noise space is [ q _Nc (m),... , q _{G −1} (m)], and q _j (m) denotes the j th noise subspace corresponding to the eigenvalue of R (m) close to σ ² , which is a variance of noise. The steering vector a (δ _i ) by the symbol timing offset is almost orthogonal to the noise subspace q _j (m). Due to this orthogonality, the denominator of Equation 20 is minimized at the symbol timing where δ = δ _i and consequently P _m (δ) has a peak value. If there is no correlation between user signals and G is greater than N _c , P _m (δ) in Equation 20 has N _c peaks.

In operation S914, a desired signal is detected using a weight of a virtual smart antenna calculated based on the estimated autocorrelation matrix and symbol timing offset.

A null-steering technique will be described as an example of a weight calculation method at step S914. Assuming that the base station estimates symbol timing offsets of OFDM signals received from N _c multiple users, the weight vector of the null-steering scheme is given by Equation 21.

Here, e , which is a G × 1 vector, and P, which is a G × N _c matrix, have e = [1, 0,... , 0] and P = [ a (δ ₁ ),. , a (δ _Nc )]. The symbol timing offset δ _i required in the null-steering technique is a value obtained in step S912.

Another example of the weight calculation method is described in the MVDR technique. The weight vector maximizing the signal-to-interference noise ratio (SINR) under a constraint condition of setting the power of the desired received signal to 1 can be obtained by solving the problem of minimizing the average received power as shown in Equation 22.

Here, the symbol timing offset is a value obtained in step S912, and the autocorrelation matrix is obtained in step S910.

10A is a flowchart illustrating a virtual multi-antenna method in an uplink of an OFDM-based cellular system according to another embodiment of the present invention, and illustrates an embodiment of a virtual smart antenna technique for removing interference of multiple users at a cell boundary. . FIG. 10B illustrates a frequency domain resource allocation structure for explaining the embodiment of FIG. 10A. This embodiment shows that the interference signal of the neighboring cell and the signal of the user in the cell are separated using the virtual smart antenna among the signals received at the base station. This embodiment is as follows with reference to FIG. 10A.

In step S1000, N _c terminals at the cell boundary generate at least one group consisting of G subchannels by grouping subchannels in a frequency domain of an OFDM symbol.

In step S1002, each terminal generates an OFDM symbol by mapping a transmission symbol to G subchannels belonging to the group, and transmits the generated OFDM symbol to a base station.

를 검출한다.In step S1004, the base station receives a signal given by Equation 1, and applies a virtual smart antenna technique that considers the received signals of the G subchannels included in the received signals as signals received from the G virtual antennas, Detect the signal transmitted from the desired terminal. The signal detection calculates the weight vector w (m) of the virtual smart antenna scheme to obtain a desired signal as shown in Equation 16.

Detect.

The difference between the embodiment of FIG. 10A and the embodiment of FIG. 9A is that N _c terminals in a group are terminals located in a cell boundary region, up to a channel response of a terminal using resources belonging to the same group allocated to an adjacent cell. It is to be estimated, and other mathematical parts are the same as in the embodiment of FIG. 9A, and thus description thereof is omitted.

In the above, the virtual SDMA technique and the virtual smart antenna technique have been described as virtual multiple antenna techniques that can be used in an uplink base station, which is an embodiment of the present invention. In this case, the virtual SDMA scheme may use all resource allocation methods as shown in FIGS. 4A to 4C. When using a resource allocation method of a comb type and a random type, diversity gain may be obtained. In addition, even if the symbol synchronization is not performed in the cell, the phase rotation is estimated by the symbol timing offset in the channel estimation process, and thus performance degradation does not occur. In addition, the virtual smart antenna technique is a technique for estimating symbol timing offsets between multiple users in a cell or cell boundary region and then virtually forming a beam to remove interference of multiple users. It is preferable to use a cluster type resource allocation method. Do. Both the virtual SDMA technique and the virtual smart antenna technique can remove interference of multiple users and perform simultaneous demodulation of multiple users.

FIG. 11A is a flowchart illustrating a virtual multi-antenna method in downlink of an OFDM-based cellular system according to an embodiment of the present invention, and illustrates an embodiment of a virtual SDMA technique applied to a downlink base station. FIG. 11B illustrates a frequency domain resource allocation structure for explaining the embodiment of FIG. 11A, and FIG. 11C is a diagram for explaining a precoding process performed by a base station.

In the case of the downlink, a virtual SDMA technique may be applied similarly to the uplink, and the order of progress thereof will be described with reference to FIG. 11A.

In step S1100, the base station groups the subchannels of the OFDM symbol to generate at least one group consisting of G subchannels.

을 산출한다.In step S1102, the base station is a channel response matrix for the G subchannels between the N _c terminal and the base station

To calculate.

을 기초로, 수학식 11과 같이 N_c개의 심볼을 프리 코딩하여 하향링크 채널에 대한 영향을 선보상한다.In step S1104, the base station transmits the channel response matrix.

Based on Equation 11, N _c symbols are precoded to show the influence on the downlink channel.

In step S1106, the base station maps the G precoded symbols to each subchannel and transmits the OFDM symbols. After passing through the downlink, a vector U (m) consisting of multi-user received signals is given by Equation (24).

Here, the transmission signal vector Z (m) of the m-th group and the multi-user reception signal vector U (m) of the m-th group are given by Equation (25).

를 단위행렬로 만드는 F(m) 행렬을 선택함으로써 하향 링크에서 가상 SDMA 기법을 적용할 수 있다. 이러한 F(m) 행렬은 ZF(Zero Forcing) 기법, 직교화(Orthogonalization) 기법, DPC(Dirty Paper Coding) 기법, THP(Tomlinson-Harashima Precoding) 기법 등을 포함하는 선형/비선형 기법을 통하여 획득될 수 있다. 일예인 ZF 기법을 이용하는 경우 F(m) 행렬은 수학식 26으로 주어진다.The g-th transmission signal Z (J _mg ) of the m-th group represents a signal to which N _c signals presented are added, and U _i (m) represents the received signal of the I-th user of the m-th group. In Equation 25

The virtual SDMA technique can be applied in the downlink by selecting an F (m) matrix that makes a unit matrix. The F (m) matrix can be obtained through linear / nonlinear techniques including Zero Forcing (ZF), Orthogonalization, DPC (Dirty Paper Coding), and Tomlinson-Harashima Precoding (THP). have. For example, when using the ZF technique, the matrix F (m) is given by Equation 26.

의 의사역행렬 또는 역행렬로 프리코딩함으로써 하향 링크의 기지국에서 가상 SDMA 기법을 수행할 수 있다.That is, the channel coefficient matrix corresponding to the m th group

The virtual SDMA scheme can be performed at the base station in the downlink by precoding a pseudo inverse or an inverse of.

In step S1108, each terminal detects the signal by summing the received signals of the G subchannels.

In the present embodiment, when a comb type or random type resource allocation method is used, diversity gain can be obtained.

12A is a flowchart illustrating a virtual multiple antenna method in a downlink of an OFDM based cellular system according to an embodiment of the present invention, and shows an embodiment of a virtual MIMO detection technique applicable to a downlink terminal. FIG. 12B is a diagram for describing the embodiment of FIG. 12A. Referring to FIG. 12B, in a cellular system having a frequency reuse factor of 1, when a terminal is located in a cell boundary region, interference between a signal transmitted from an adjacent cell and a signal transmitted from a desired cell occurs. In this case, similar to the virtual SDMA technique applied to the uplink base station described in the embodiment of FIG. 7A, inter-cell interference may be eliminated by a virtual MIMO detection technique using spatial characteristics of channel frequency responses constituting resources.

This embodiment will be described with reference to FIG. 12A.

In step S1200, the Nc base station groups the subchannels of the OFDM symbol by the same grouping method to generate at least one group consisting of G subchannels.

In step S1202, each base station generates an OFDM symbol by mapping a symbol to be transmitted to a terminal belonging to its cell to the G subchannels, and transmits the generated OFDM symbol.

In step S1204, the terminal detects a signal transmitted from a cell to which the terminal belongs by applying a virtual SDMA technique that considers the received signals of the G subchannels as signals received from the G virtual antennas.

는 i 번째 기지국과 단말 사이의 채널 응답을 나타낸다. 이하 수학적 부분은 도 7a의 실시예에 기술된 내용과 동일하므로 여기서는 설명을 생략한다.This embodiment is similar to Figure 7a in mathematical terms, the difference is that in this embodiment N _c represents the number of adjacent multiple cells,

Denotes a channel response between the i-th base station and the terminal. Since the mathematical part is the same as that described in the embodiment of FIG. 7A, the description is omitted here.

13A is a flowchart illustrating a virtual multi-antenna method in downlink of an OFDM-based cellular system according to an embodiment of the present invention, and a virtual smart antenna technique capable of removing inter-cell interference applicable to a downlink terminal An embodiment of the is shown. FIG. 13B is a diagram for describing the embodiment of FIG. 13A. Referring to FIG. 13B, in a cellular system having a frequency reuse factor of 1, a symbol timing offset due to a transmission delay of each base station signal is generated in a signal received by a terminal located at a cell boundary. Using this characteristic, a virtual smart antenna technique may be applied to each terminal similarly to the embodiment of FIG. 9A.

The present embodiment will be described with reference to FIG. 13A.

In step S1300, the N _c base stations group at least one subchannel of the OFDM symbol in the same grouping manner to generate at least one group consisting of G subchannels.

In step S1302, each base station generates an OFDM symbol by mapping a symbol to be transmitted to a terminal belonging to its cell to the G subchannels, and transmits the generated OFDM symbol.

In step S1304, each terminal detects a signal transmitted from a cell to which the terminal belongs by applying a virtual smart antenna technique that regards the received signals of the G sub-channels as a signal received from the G virtual antenna.

는 i 번째 기지국과 단말 사이의 채널 응답을 나타낸다. 이하 수학적 부분은 도 9a의 실시예에 기술된 내용과 동일하므로 여기서는 설명을 생략한다.This embodiment is similar to Figure 9a in mathematical terms, the difference is that in this embodiment N _c represents the number of adjacent multiple cells,

Denotes a channel response between the i-th base station and the terminal. Since the mathematical parts are the same as those described in the embodiment of FIG. 9A, the description is omitted here.

Hereinafter, seven embodiments of a virtual multiple antenna method of an OFDM-based cellular system will be described.

On the other hand, the smaller the correlation between the channels, the greater the effect of the present invention is generally. That is, when channel responses of subchannels within a group are correlated with each other, the virtual SDMA technique or the virtual MIMO detection technique according to the present invention may not obtain the maximum diversity gain or may not effectively remove the interference signal. . Therefore, another embodiment of the present invention may further include a channel randomization process for the above-described embodiments to maximize the effect of the present invention. That is, another embodiment of the present invention is an embodiment using a virtual signature randomizer (VSR) technique, and makes a channel response between devices independently by multiplying a transmission symbol or an estimated channel response value by a predetermined weight in a transceiver. . This method is a method of applying a virtual multi-antenna technique based on the weighted channel response rather than applying the virtual multi-antenna technique based on the physical channel response itself.

14A is a block diagram illustrating an OFDM cellular system using a VSR technique and a virtual multiple antenna scheme according to an embodiment of the present invention. That is, FIG. 14A illustrates a process of randomizing a channel response by applying a VSR technique to an OFDM cellular system according to an embodiment of the present invention. Assuming that a signal transmitted from one target device is detected in a system that includes an interference signal of one interfering device, the present embodiment will be described with reference to FIG. 14A.

Referring to FIG. 14A, the OFDM cellular system according to the present embodiment, like FIG. 1, includes a transmitting end and an OFDM symbol including a transmission symbol generator 1400, an OFDM symbol generator 1410, and an OFDM symbol transmitter 1420. A receiver includes a receiver 1430, an OFDM symbol demodulator 1440, and a data detector 1450.

The transmission symbol generator 1400 generates transmission symbols to be carried on a channel (subchannel) according to each subcarrier similarly to the transmission symbol generator 100 of FIG. 1A. The grouping process according to the present invention will be described assuming that M groups having G subchannels are generated. In this case, the transmission symbol generator 1400 may transmit X ₀ , X ₁ ,... , X _{M -1}

The OFDM symbol generator 1410 performs X ₀ , X ₁ ,... An OFDM symbol including the information, X _{M -1} is generated. The difference between the OFDM symbol generator 1410 and the OFDM symbol generator 110 of FIG. 1 is that the OFDM symbol generator 1410 further includes a VSR 1412 and a scrambler 1413. Here, the scrambler 1413 is a block that multiplies a predetermined scrambling code for randomizing signals between cells in a cellular environment. It can be omitted. That is, according to the present embodiment, SDMA demodulation or virtual MIMO detection may be performed by reflecting the code of the scrambler block and the weight of the VSR block in the channel response.

The functions and functions of the remaining blocks S / P 1411, IFFT 1414, P / S 1415, and guard interval insertion unit 1416 are shown in FIG. 1, S / P 111, IFFT 112, and P. Since / S 113 and the protection section insertion unit 114 is the same, a description thereof will be omitted.

The VSR 1412 multiplies the allocated channel randomization code by a transmission symbol to be carried on each resource (subcarrier). In FIG. 14A, the VSR 1412 is located at the front end of the scrambler 1413, but may be located at the rear end of the scrambler 1413. Here, the assigned channel randomization code represents a column of weights for randomizing a channel as described above. In the embodiment of FIG. 14A, the target user may use c ₀ , c ₁ ,... , c _{G - 1} is assigned a channel randomization code, the interference user c ₁₀ , c ₁₁ ,. In this case, it is assumed that a channel randomization code of c _{1G - 1} is allocated. The VSR 1412 stores X ₀ , X ₁ ,... Symbols, which are transmission symbols to be carried on each resource (subcarrier). , X _{M -1,} X ₀ , X ₁ ,... , X _{M -1,} ... , X ₀ , X ₁ ,.. , X _{M - 1 for} each channel randomization code C ₀ ,... , C ₀ , C ₁ ,... , C ₁ ,… , C _G-1 ,... Multiply by C _G-1 . In FIG. 14A, it is assumed that a comb-type resource allocation method is used and all groups of each user use the same channel randomization code. However, the present invention may use a different resource allocation scheme or each of the M groups may use different channel randomization codes.

Equation 27 shows an example of weights used in the VSR 1412.

Here, c _i (n) represents a weight multiplied by the n th resource of the i th user, and rand (i, n) has an arbitrary value from 0 to 8. That is, C ₀ , C ₁ ,... , C _G-1 is given by the equation (27) c ₁ (0), c ₁ (1), ... , c ₁ (G-1). In Equation 27, an example of weights of 8-PSK (Phase Shift Keying) is given, but is not necessarily limited thereto.

The scrambler 1413 multiplies each input symbol with a scrambling code S _n . In FIG. 14A, it is assumed that an interfering user is assigned a scrambling code S _{1 , n} .

Since the OFDM symbol transmitter 1420 and the OFDM symbol receiver 1430 are the same in function and operation as the OFDM symbol transmitter 1420 and the OFDM symbol receiver 1430 of FIG. 1, description thereof will be omitted.

In FIG. 14A, the channel response value on the frequency between the receiver and the target device is denoted by H _n , and the channel response value on the frequency between the receiver and the interfering device is denoted by H _{1, n} .

를 검출하여 데이터 검출부(1450)에 제공한다. OFDM 심볼 복조 부(1440)가 도 1의 OFDM 심볼 복조부(140)와 다른 점은 디스크램블러(1444) 및 VSR 복호화부(1447)을 더 포함하고, 설명을 간단히 하기 위하여 동기&채널추정부(146) 대신 채널추정부(1445)가 존재하는 것이다. 또한, 본 발명의 가상 다중 안테나 기법에 따른 신호 검출이 수행되는 SDMA 블록(1446)이 등화기(144) 대신 존재하는 것도 다르다. OFDM symbol demodulator 1410 includes X ₀ , X ₁ ,... , The signal corresponding to X _{M -1}

Is detected and provided to the data detector 1450. The difference between the OFDM symbol demodulator 1440 and the OFDM symbol demodulator 140 of FIG. 1 further includes a descrambler 1444 and a VSR decoder 1447. 146) Instead, the channel estimation unit 1445 exists. In addition, the SDMA block 1446 in which signal detection is performed according to the virtual multiple antenna scheme of the present invention is also present instead of the equalizer 144.

Here, the descrambler 1413 performs an inverse process of the scrambler 1413, and the VSR decoder 1447 multiplies the channel value multiplied by the transmitting end with the result of the channel response estimated by the channel estimator 1445. The multiplication code is multiplied and provided to the SDMA block 1446.

은 수학식 28과 같이 주어진다.As a output of the VSR decoder 1447, a channel matrix used for virtual SDMA demodulation or virtual MIMO detection.

Is given by Equation 28.

는 i번째 사용자의 채널 응답을 나타내며 도 14a를 참조하면, [H_mC₀, H_{M + m}C₁,…,H_{M (G-1)+ m}C_{G -1}]^T 이에 해당된다. 물리적 채널이 변하지는 않았지만, 상기 곱하여진 가중치를 포함한 채널 응답을 이용하여 SDMA 복조 또는 가상 MIMO 검출을 수행하므로 가중치까지 곱해진 성분을 채널로 간주할 수 있게 된다. 그러므로, 한 그룹내에 속하는 부반송파 채널 응답, H_i(J_mg)이 일반적으로 서로 유사하게 되는 클러스터 타입의 자원 할당 방식을 사용하는 경우에는 다이버시티 이득을 얻을 가능성이 적지만, 이러한 클러스터 타입의 자원 할당 방식을 사용한 가상 다중 안테나 기법에 VSR 과정을 추가한다면우 한 그룹의 채널 응답 c_i(J_mg)H_i(J_mg)이 서로 독립적이게 되어 다이버시티 이득을 얻을 수 있다. here,

Denotes the channel response of the i-th user, and referring to FIG. 14A, [H _m C ₀ , H _{M + m} C ₁ ,... , H _{M (G-1) + m} C _{G -1} ] ^T Although the physical channel has not changed, since the SDMA demodulation or the virtual MIMO detection is performed using the channel response including the multiplied weight, the component multiplied by the weight can be regarded as the channel. Therefore, in the case of using a cluster type resource allocation scheme in which subcarrier channel responses, H _i (J _mg ), which belong to a group are generally similar to each other, diversity gains are less likely to be obtained. If the VSR process is added to the virtual multi-antenna scheme, a group gain of channel response c _i (J _mg ) H _i (J _mg ) becomes independent of each other, thereby achieving diversity gain.

를 출력한다. As a result, the SDMA block 1446 detects a desired signal from a signal composed of the sum of the respective user signals passing through the randomized channel.

Outputs

The remaining blocks, the protection section removal unit 1441, the S / P 1442, the FFT 1443, and the P / S 1484, are the protection section removal unit 141, S / P 142, and FFT (FIG. 1). 143) and the P / S 145 are the same in terms of function and operation, and thus description thereof will be omitted.

14B is a block diagram illustrating an embodiment in which a VSR process is added to the embodiment of FIG. 11A. When the base station transmits an OFDM symbol generated by demonstrating the influence of the channel response to the Nc user terminals, each user terminal detects a desired signal.

Referring to FIG. 14B, similarly to FIG. 1, the OFDM cellular system according to the present embodiment includes a transmission symbol generator 1500_1, 1500_2,..., 1500_N _c , an OFDM symbol generator 1510, and an OFDM symbol transmitter 1520. The transmitter includes a transmitter including an OFDM symbol receiver 1530, an OFDM symbol demodulator 1540, and a data detector 1550. Here, the transmitting end means a transmitting end of the base station, the receiving end means a target user terminal and the description will be described on the assumption that the target user terminal has a value of the user index I = 1.

Each of the transmission symbol generators 1500_1, 1500_2,..., 1500_N _c generates a transmission symbol to be transmitted to each user terminal. For example, the transmission symbol generator 1500_i may include X _i0 , X _i1 ,... , X _{1M -1} . The VSRs 1512_1, 1512_2,..., 1512_N _c multiply the channel randomization code assigned to each user by the input transmission symbol. When the comb type resource allocation method is used, examples of the output of the VSR 1512_i include c _i (0) X _i0 , c _i (0) X _i1 ,... , c _i (0) X _1M-1, c _i (1) X _i0 , c _i (1) X _i1 ,... , c _i (1) X _1M-1, ... c _i (G-1) X _i1 ,... A vector comprising a _{c i (G-1) X} 1M- 1.

The scrambler 1513_i multiplies the output of the VSR 1512_i by the scrambling code S _n . Since the technique is performed at a base station in a cell, the scrambling code is the same for each user.

The presentation unit 1514 estimates H _i (k), which is a channel response between each user terminal and the base station, and then precodes the input signal based on the estimated result. Since the embodiment of FIG. 14B includes VSRs 1512_1, 1512_2,..., 1512_N _c , the precoding matrix is calculated based on the result of multiplying the channel response H _i (k) by the channel randomization code. Precode the input signal with the calculated precoding matrix.

Thanks to the precoding of the transmitting end, the receiving end can detect the transmitted symbol transmitted from the transmitting end to itself without any special signal processing. If the receiving end of FIG. 14B is the receiving end of the user terminal having user index i = 1, N _c = G, and Equation 26 is used as the precoding matrix, the output of the FFT 1543 is Y (m) = S _m X _1m. + N _m , Y (M + m) = S _{M + m} X _1m + N _{M + m} ,... , Y (M (G-1) + m) = S _{M (G-1) + m} × _1m + _{NM (G-1) + m} . Accordingly, the data detector 1550 may detect the data by summing the outputs of the descrambler 1544 for each group.

Meanwhile, the same demodulation process may be performed at the receiving end regardless of whether the VSR is performed at the transmitting end. That is, this embodiment does not need the VSR decoder 1447 shown in Fig. 14A.

The invention can also be embodied as computer readable code on a computer readable recording medium. Computer-readable recording media include all kinds of recording devices that store data that can be read by a computer system. Examples of computer-readable recording media include ROM, RAM, CD-ROM, magnetic tape, floppy disks, optical data storage devices, and the like, which are also implemented in the form of carrier waves (for example, transmission over the Internet). Include. The computer readable recording medium can also be distributed over network coupled computer systems so that the computer readable code is stored and executed in a distributed fashion. In addition, functional programs, codes, and code segments for implementing the present invention can be easily inferred by programmers in the art to which the present invention belongs.

Such a method and apparatus of the present invention have been described with reference to the embodiments shown in the drawings for clarity, but these are merely exemplary, and various modifications and equivalent other embodiments are possible to those skilled in the art. Will understand. Therefore, the true technical protection scope of the present invention will be defined by the appended claims.

According to the present invention, it is possible to derive the effect of canceling an interference signal similar to using a multi-antenna physically even when using a single antenna for the transceiver. Therefore, hardware complexity does not increase according to the use of physical multi-antennas, and there is an advantage that the cancellation performance of interference signals is not sensitive to the synchronization performance.

According to the virtual SDMA scheme of the base station in the uplink of the present invention, the overall cell spread spectrum efficiency is increased due to effective multi-user interference cancellation. In addition, when the number of multi-users is smaller than the number of subcarriers forming a group, diversity gain can be obtained, thereby increasing the performance of the system.

According to the virtual MIMO technique of the base station in the uplink of the present invention, it is possible to effectively remove the signal of the interference user in the cell boundary region using the virtual MIMO technique. In addition, if the number of multi-users is smaller than the number of subcarriers in the group, the diversity gain can be obtained, and the performance of the system is also increased.

According to the virtual smart antenna technique of the base station in the uplink of the present invention, an array gain can be obtained and performance deterioration due to symbol synchronization error of multiple users does not occur.

According to the virtual smart antenna technique of the base station in the uplink of the present invention, it is possible to remove the signal of the interference user of the cell boundary area, and obtain an array gain. In addition, there is an advantage that the performance degradation due to the symbol synchronization error of the multi-user does not occur.

According to the virtual SDMA technique of the base station in the downlink of the present invention, performance deterioration due to a channel can be reduced by performing a virtual SDMA technique by performing a pilot. In addition, diversity gain may be obtained when the number of multi-users is smaller than the number of subcarriers forming a group.

According to the virtual MIMO detection technique and the virtual smart antenna technique of the UE in the downlink according to the present invention, when the frequency reuse coefficient is 1, interference between neighbor cells is effectively eliminated so that interference between cells does not occur. In addition, since signals of adjacent cells can be demodulated at the same time, it is effective for handoff and the like. In addition, diversity gain and array gain can be obtained according to channel conditions, and performance deterioration due to symbol synchronization error does not occur.

According to the virtual multi-antenna technique using the VSR scheme of the present invention, performance degradation due to inter-channel correlation may be reduced by virtually randomly changing a channel response corresponding to a group.

Claims (44)

(a) grouping subchannels in the frequency domain of the OFDM symbol to generate at least one group consisting of G subchannels; And and (b) virtually applying the multiple antenna scheme to the transmission and reception of the OFDM symbol by considering the G subchannels belonging to the group as multiple channels of the multiple antenna scheme. Antenna method. The method of claim 1, wherein the multi-antenna technique applied in the step (b), A virtual multiple antenna method in an OFDM system comprising a spatial multiplexed division access (SDMA) technique, a MIMO detection technique, and a smart antenna technique. According to claim 1, wherein step (b), (b1) a transmitting device estimating a channel response for the G subchannels between Nc receiving devices and the transmitting device; (b2) the transmitting device presenting an influence on the channel by precoding Nc symbols to be transmitted to the G subchannels based on the channel matrix composed of the estimated channel response; And and (b3) the transmitting device transmitting an OFDM symbol including the precoded Nc symbols to the Nc receiving devices. According to claim 1, wherein step (b), (b1) multiplying each of the Nc transmitting devices by the G weights for randomizing the channel between the transmitting devices and the symbols carried on the G subchannels, and transmitting an OFDM symbol including the multiplied symbols; (b2) a receiving device estimating channel responses for the G subchannels between the Nc transmitting devices and the receiving devices, and multiplying the estimated channel response values by a weight used by a target transmitter; And and (b3) detecting, by the receiving device, a signal transmitted from the target transmitter by applying a virtual multiple antenna scheme based on the multiplied channel response value. Way. According to claim 1, wherein the step (a), And generating the group by using any one of a comb type, a cluster type, and a random type. (a) generating, by the Nc terminals, at least one group consisting of G subchannels by grouping subchannels in a frequency domain of an OFDM symbol by the same grouping method; (b) the Nc terminal generating an OFDM symbol by mapping a symbol to a subchannel belonging to the group, and transmitting the generated OFDM symbol to a base station; And and (c) detecting, by the base station, a signal by applying a multi-antenna technique that considers the received signals of the G subchannels as signals received from the G virtual antennas. The virtual multiple antenna method in the uplink of the. The method of claim 6, The step (b) includes the steps of mapping the multiplied symbols to the G subchannels by multiplying the symbols by the Nc terminals by G weights for randomizing an uplink channel. In the step (c), after the base station estimates the channel responses of the G subchannels, the estimated channel response values are multiplied by the G weights, respectively, and are based on the multiplied channel response values. A method of virtual multi-antenna in the uplink of an OFDM-based cellular system, comprising applying a technique. 8. The method of claim 7, wherein each of said weights is Virtual multi-antenna method in the uplink of the OFDM-based cellular system, characterized in that the same size and the value of the M-PSK series. According to claim 6, wherein step (a), And generating the group by using any one of a comb type, a cluster type, and a random type. The method of claim 6, wherein the virtual multiple antenna technique A virtual multi-antenna method in the uplink of an OFDM-based cellular system, characterized in that it is a virtual SDMA technique. The method of claim 10, wherein step (c) comprises: Virtual multiplexing in the uplink of an OFDM based cellular system, comprising removing interference signals and detecting signals of a desired user using signal detection techniques including ZF, MMSE, SIC, PIC, and ML Antenna method. The method of claim 10, wherein step (c) comprises: Detecting a signal of multiple users simultaneously using signal detection techniques including ZF, MMSE, SIC, PIC, and ML. The method according to any one of claims 6 to 12, The Nc terminal comprises a terminal located in a cell boundary region and a terminal located in an adjacent cell boundary region, the virtual multiple antenna method in the uplink of the OFDM-based cellular system. The method of claim 6, wherein the virtual multiple antenna technique Virtual multi-antenna method in the uplink of the OFDM-based cellular system, characterized in that the virtual smart antenna technique. The method of claim 14, wherein step (c) comprises: (c1) estimating an autocorrelation matrix of a vector including the received signals of the G subchannels; (c2) estimating a symbol timing offset between terminals; And (c2) detecting a signal using a weight of a virtual smart antenna calculated based on the estimated autocorrelation matrix and a symbol timing offset; and multi-antenna in an uplink of an OFDM based cellular system Way. The method of claim 15, wherein step (c1), Estimating the autocorrelation matrix using a characteristic in which a symbol timing offset affects a received signal as a phase rotation between adjacent subcarriers. . The method of claim 15, wherein step (c2), Estimating the symbol timing offset by virtually applying a smart antenna technique including MUSIC and ESPRIT to the estimated autocorrelation matrix. Multiple antenna method. The method of claim 14, wherein step (c) comprises: Computing the weight of the virtual smart antenna using a training signal-based method including the LMS, RLS and SMI techniques. The method of claim 14, wherein step (c) comprises: calculating a weight of a virtual smart antenna using a symbol timing offset based technique including null-steering and MVDR. The method of claim 14, wherein step (c) comprises: A method for virtual multi-antenna in uplink of an OFDM based cellular system, comprising the step of removing interference signals and detecting signals of a desired user using a virtual smart antenna technique. The method of claim 14, wherein step (c) comprises: And simultaneously detecting a signal of multiple users using a virtual smart antenna technique. The method according to any one of claims 14 to 21, The Nc terminal comprises a terminal located in a cell boundary region and a terminal located in an adjacent cell boundary region, the virtual multiple antenna method in the uplink of the OFDM-based cellular system. (a) the base station grouping subchannels of the OFDM symbol to generate at least one group consisting of G subchannels; (b) the base station calculating a channel response matrix for the G subchannels between Nc terminals and the base station; (c) the base station presenting an effect on a downlink channel by precoding Nc symbols based on the channel response matrix; And and (d) mapping, by the base station, the precoded G symbols to each subchannel and transmitting the OFDM symbols as OFDM symbols. The method of claim 23, wherein and (e) summing, by each terminal, the received signals of the G subchannels to detect a signal. The method of claim 23, wherein step (a) comprises: And generating the group by using any one of a comb type, a cluster type, and a random type. The method of claim 23 wherein the precoding is Virtual multi-antenna method in downlink of an OFDM-based cellular system characterized by ZF, orthogonalization, DPC and THP. The method of claim 23, wherein The step (b) includes the base station calculating the channel response matrix by multiplying the estimated channel response values of the G subchannels of each terminal by the G weights for randomizing the downlink channel. A virtual multi-antenna method in downlink of an OFDM based cellular system. 28. The method of claim 27, wherein each of said weights is Virtual multi-antenna method in downlink of an OFDM-based cellular system, characterized in that the same size and the value of the M-PSK series. (a) Nc base stations grouping subchannels of an OFDM symbol in the same grouping manner to generate at least one group consisting of G subchannels; (b) generating, by each base station, an OFDM symbol by mapping a transmission symbol to the G subchannels, and transmitting the generated OFDM symbol; And (c) a terminal for detecting a desired signal by applying a virtual multi-antenna technique in which received signals of the G subchannels are regarded as signals received from the G virtual antennas; Virtual Multiple Antenna Method on Link. The method of claim 29, wherein the step (a), And generating the group by using any one of a comb type, a cluster type, and a random type. 30. The method of claim 29, wherein the virtual multiple antenna technique is A virtual multi-antenna method in downlink of an OFDM-based cellular system, characterized in that it is a virtual SDMA technique. The method of claim 29, wherein step (c) Canceling an interference signal using a signal detection technique including ZF, MMSE, SIC, PIC, and ML, and detecting a signal transmitted from a cell to which the UE belongs. Virtual Multiple Antenna Method on Link. The method of claim 29, wherein step (c) Simultaneously detecting a signal of a neighboring cell along with a signal of a target cell using a signal detection technique comprising ZF, MMSE, SIC, PIC, and ML, in a downlink of an OFDM-based cellular system. Multiple antenna method. The method of claim 29, The step (b) includes the step of mapping the multiplied symbols to the G subchannels by multiplying each symbol by G weights for randomizing a downlink channel. In the step (c), after each terminal estimates the channel responses of the G subchannels, the estimated channel response values are multiplied by the G weights, respectively, based on the multiplied channel response values. A method of virtual multi-antenna in downlink of an OFDM-based cellular system, comprising applying an antenna technique. 35. The method of claim 34, wherein each of said weights is Virtual multi-antenna method in downlink of an OFDM-based cellular system, characterized in that the same size and the value of the M-PSK series. 30. The method of claim 29, wherein the virtual multiple antenna technique is A virtual multi-antenna method in downlink of an OFDM-based cellular system characterized by a virtual smart antenna technique. The method of claim 36, wherein step (c) comprises: (c1) estimating an autocorrelation matrix of a vector including the received signals of the G subchannels; (c2) estimating a symbol offset between adjacent cells; And (c2) detecting a signal using a weight of a virtual smart antenna calculated based on the estimated autocorrelation matrix and symbol timing offset; Way. The method of claim 37, wherein the step (c1), Estimating the autocorrelation matrix using a characteristic in which a symbol timing offset affects a received signal as a phase rotation between adjacent subcarriers. . The method of claim 37, wherein step (c2), Estimating the symbol timing offset by virtually applying a smart antenna scheme including MUSIC and ESPRIT to the estimated autocorrelation matrix. . The method of claim 36, wherein step (c) comprises: Calculating a weight of a virtual smart antenna using a training signal-based technique including an LMS, an RLS, and an SMI technique. The method of claim 36, wherein step (c) comprises: calculating a weight of a virtual smart antenna using a symbol timing offset based technique including null-steering and MVDR. The method of claim 36, wherein step (c) comprises: A method of virtual multi-antenna in downlink of an OFDM based cellular system, comprising the step of removing interference signals of adjacent cells using a virtual smart antenna technique and detecting a desired signal. The method of claim 36, wherein step (c) comprises: And detecting a signal of a neighbor cell and a signal of a target cell simultaneously using a virtual smart antenna technique. A computer-readable recording medium containing a program for performing the method according to any one of claims 1 to 43.

Images