WO2009107732A1 - チャネル情報予測システム及びチャネル情報予測方法 - Google Patents
チャネル情報予測システム及びチャネル情報予測方法 Download PDFInfo
- Publication number
- WO2009107732A1 WO2009107732A1 PCT/JP2009/053577 JP2009053577W WO2009107732A1 WO 2009107732 A1 WO2009107732 A1 WO 2009107732A1 JP 2009053577 W JP2009053577 W JP 2009053577W WO 2009107732 A1 WO2009107732 A1 WO 2009107732A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- prediction
- channel information
- filter coefficient
- component
- antenna
- Prior art date
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L25/00—Baseband systems
- H04L25/02—Details ; arrangements for supplying electrical power along data transmission lines
- H04L25/0202—Channel estimation
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B17/00—Monitoring; Testing
- H04B17/30—Monitoring; Testing of propagation channels
- H04B17/373—Predicting channel quality or other radio frequency [RF] parameters
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B7/00—Radio transmission systems, i.e. using radiation field
- H04B7/02—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas
- H04B7/04—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas
- H04B7/0413—MIMO systems
- H04B7/0426—Power distribution
- H04B7/0434—Power distribution using multiple eigenmodes
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L25/00—Baseband systems
- H04L25/02—Details ; arrangements for supplying electrical power along data transmission lines
- H04L25/03—Shaping networks in transmitter or receiver, e.g. adaptive shaping networks
- H04L25/03006—Arrangements for removing intersymbol interference
- H04L25/03178—Arrangements involving sequence estimation techniques
- H04L25/03248—Arrangements for operating in conjunction with other apparatus
- H04L25/03254—Operation with other circuitry for removing intersymbol interference
Definitions
- the present invention relates to a channel information prediction system and a channel information prediction method for calculating a prediction filter coefficient used for suppression of interference components included in a received radio signal in a multi-antenna transmission system.
- MIMO Muli-Input Multi-Output
- the channel reversibility in time division multiplexing (TDD) or the channel characteristics on the receiver side in a MIMO environment is demonstrated if a feedback channel is used in frequency division multiplexing (FDD).
- Channel information (hereinafter abbreviated as MIMO CSI as appropriate) can be estimated on the transmitter side. Therefore, there is known a method of predicting the future MIMO, that is, the MIMO CSI at the time of transmission of the next radio signal, using the past and current MIMO CSI (see Non-Patent Document 1 and Non-Patent Document 2).
- MMSE minimum mean square error
- the conventional MIMO-CSI prediction method described above has the following problems. That is, if the received radio signal includes interference components, specifically, many elementary waves (reflected waves existing on the propagation path), there is a problem that the prediction performance of MIMO-CSI is not sufficiently improved.
- the present invention has been made in view of such a situation, and in a multi-antenna transmission system that predicts channel information by calculating a prediction filter coefficient used for suppressing an interference component included in a received radio signal. It is an object of the present invention to provide a channel information prediction system and a channel information prediction method that are used and in which the prediction performance of the channel information is further improved.
- the present invention has the following features.
- a plurality of antennas (transmission antennas # 1 to #T or reception antennas #) are provided on at least one of a radio signal transmission side (transmitter 100) and reception side (receiver 200).
- a channel information prediction system for calculating a prediction filter coefficient used for suppressing interference components included in a received radio signal 10)
- a channel information prediction system channel information prediction system for calculating a prediction filter coefficient used for suppressing interference components included in a received radio signal 10.
- the filter coefficient calculation unit calculates the second prediction filter coefficient using a low elementary wave component in which the component of the elementary wave is suppressed by the channel information prediction unit, and the channel information prediction unit In summary, the second prediction filter coefficient is used to perform second prediction for predicting the channel information at the time of future transmission of the radio signal.
- Such a channel information prediction system performs prediction using the first prediction filter coefficient for the low elementary wave element, and suppresses the component of the elementary wave contained in the low elementary wave element. Then, after suppressing the component of the element wave included in the low element element, the second prediction filter coefficient is calculated, and prediction using the second prediction filter coefficient is performed. For this reason, it is possible to suppress the influence of the wave and improve the prediction performance of the channel information.
- a second feature of the present invention relates to the first feature of the present invention, wherein the channel information prediction unit uses the second prediction filter coefficient as the first prediction filter coefficient, and The gist is to execute prediction 1 and suppress the component of the wave.
- a third feature of the present invention relates to the first feature or the second feature of the present invention, wherein the filter coefficient calculation unit uses the inter-antenna element having the smallest number of element waves as the low element element. Is the gist.
- a fourth feature of the present invention relates to the first feature or the second feature of the present invention, wherein the channel information prediction unit is different from the low element element based on the prediction filter coefficient. And the channel information at the time of future transmission of the radio signal is predicted using the inter-antenna element in which the component of the elementary wave is suppressed. To do.
- a fifth aspect of the present invention in a multi-antenna transmission system in which a plurality of antennas are used on at least one of a radio signal transmitting side and a receiving side, prediction used for suppressing interference components included in the received radio signal.
- the present invention includes a step of calculating a filter coefficient and a step of executing a second prediction for predicting the channel information at the time of future transmission of the radio signal by using the second prediction filter coefficient. .
- the present invention is used in a multi-antenna transmission system that predicts channel information by calculating a prediction filter coefficient used for suppressing an interference component included in a received radio signal, and the prediction performance of the channel information is further improved.
- An improved channel information prediction system and channel information prediction method can be provided.
- FIG. 1 is an overall schematic configuration diagram of a multi-antenna transmission system according to an embodiment of the present invention.
- FIG. 2 is a diagram illustrating the relationship between the estimated time of MIMO CSI and the transmission time according to the embodiment of the present invention.
- FIG. 3 is a functional block configuration diagram showing the configuration of the channel prediction unit according to the embodiment of the present invention.
- FIG. 4 is a conceptual diagram for explaining the function of the channel prediction unit according to the embodiment of the present invention.
- FIG. 5 is a flowchart showing the overall operation of the multi-antenna transmission system according to the embodiment of the present invention.
- FIG. 6 is a flowchart showing the operation of the channel prediction unit according to the embodiment of the present invention.
- FIG. 1 is an overall schematic configuration diagram of a multi-antenna transmission system according to an embodiment of the present invention.
- FIG. 2 is a diagram illustrating the relationship between the estimated time of MIMO CSI and the transmission time according to the embodiment of the present invention.
- FIG. 3
- FIG. 7 is a diagram illustrating prediction characteristics when the number of antennas is 4 ⁇ 4 and the order of the prediction filter is 20.
- FIG. 8 is a diagram illustrating the RMSE characteristics of each component of transformed MIMO CSI.
- FIG. 9 is a diagram showing characteristics when the number of antennas is 2 ⁇ 2 to 5 ⁇ 5.
- FIG. 10 is a diagram showing RMSE characteristics with respect to the order of the filter coefficient when the number of antennas is 4 ⁇ 4 and the delay time is 10 msec.
- FIG. 1 is an overall schematic configuration diagram of a multi-antenna transmission system 1 according to the present embodiment. As shown in FIG. 1, the multi-antenna transmission system 1 includes a transmitter 100, a receiver 200, and a channel information prediction system 10.
- the channel information prediction system 10 is provided separately from the transmitter 100 and the receiver 200, but either the transmitter 100 or the receiver 200, or the transmitter 100 and the receiver 200.
- the channel information prediction system 10 can be provided in a distributed manner.
- the channel information prediction system 10 uses the MIMO CSI indicating the transmission path characteristics between the transmitter 100 and the receiver 200 and the transformation matrix, and predicts the future MIMO CSI prediction value (hereinafter referred to as the MIMO CSI prediction value as appropriate). (Omitted) is calculated.
- the transmitter 100 is provided in one of the radio base station and the radio terminal, and the receiver 200 is provided in the remaining one.
- the receiver 200 is provided in a wireless terminal.
- the propagation path 4 fluctuates due to movement of the receiver 200 or movement of a scatterer existing between the transmitter 100 and the receiver 200.
- each of the transmitter 100 and the receiver 200 is provided with a plurality of antennas.
- the transmitter 100 includes a transmission processing unit 2 and a plurality of transmission antennas # 1 to #T (T ⁇ 2).
- An antenna array is configured by the transmission antennas # 1 to #T.
- the transmission processing unit 2 Based on the MIMO-CSI prediction value obtained from the channel information prediction system 10, the transmission processing unit 2 performs transmission data control, for example, adaptive modulation control and transmission beamforming by precoding.
- the receiver 200 receives a radio signal transmitted by the transmitter 100 via the propagation path 4.
- the receiver 200 includes a reception processing unit 6 and a plurality of reception antennas # 1 to #R (R ⁇ 2).
- the antenna array is constituted by the receiving antennas # 1 to #R.
- the reception processing unit 6 demodulates radio signals (reception signals) received by the reception antennas # 1 to #R and outputs reception data.
- the channel information prediction system 10 includes a channel estimation unit 8 and a channel prediction unit 9.
- the channel estimation unit 8 estimates the current MIMO ⁇ ⁇ CSI based on the received signal.
- MIMO CSI is expressed as a matrix composed of a plurality of inter-antenna elements that are distinguished by transmission path characteristics between any transmission antenna in the transmitter 100 and any reception antenna in the receiver 200.
- the channel prediction unit 9 calculates a MIMO CSI prediction value by a beam space linear prediction process based on the current MIMO CSI estimated by the channel estimation unit 8 and the past MIMO CSI. That is, MIMO CSI at the next transmission in transmitter 100 is predicted.
- the channel prediction unit 9 performs prediction on the transformed MIMO CSI obtained by transforming MIMO CSI using the transformation matrix, and inversely transforms the predicted value obtained for the transformed MIMO CSI, thereby obtaining the MIMO CSI. Get the predicted value (MIMO CSI predicted value).
- linear prediction (AR-LP) based on an autoregressive (AR) model is executed in each element of transformed MIMOMICSI obtained by multiplying MIMO CSI by a transformation matrix.
- radio waves (radio signals) radiated from the transmitter 100 are reflected by various scatterers and reach the receiver 200 through a plurality of propagation paths (multipaths).
- Each reflected wave is called an elementary wave, and the multipath propagation path 4 is expressed by superposition of the elementary waves.
- each subcarrier has a sufficiently narrow band. Therefore, in this embodiment, a flat fading environment is assumed in which a delay time difference between paths is negligible.
- each elementary wave undergoes a Doppler shift.
- the Doppler frequency of the elementary wave l is given by
- the multipath propagation path 4 at time t can be expressed by the following equation using L elementary waves.
- ⁇ l is the complex scattering coefficient of the elementary wave l.
- ⁇ r, l and ⁇ t, l are reception / transmission array response vectors for the elementary wave l, and the sizes thereof are the number of reception antennas R and the number of transmission antennas T, respectively.
- the transmission array response vector for the elementary wave l is expressed by the following equation.
- Expression (3) can be expressed as the following expression by matrix expression.
- a r and A t are each R ⁇ L, and T ⁇ L, the reception for the L rays, a transmit array response set. Since the estimated MIMO CSI includes an error due to the influence of noise, it is expressed by the following equation.
- AR-LP Linear prediction based on AR model
- d (j) is a linear prediction filter coefficient.
- the optimum filter coefficient d [d (1)... D (P)] minimizes the mean square error of the predicted value and is given by the following equation.
- an autocorrelation matrix and an autocorrelation vector are calculated by sample averaging using past MIMO CSI.
- the prediction is repeated by using the prediction value as a filter input.
- the side lobe of the autocorrelation characteristic has a much larger value. This is more conspicuous as the dominant wave number is smaller.
- the dominant wave number is L
- each component of the MIMO channel is affected by L wave waves.
- the prediction performance should be improved if the number of elementary waves existing in each component is reduced.
- W r and W t are R ⁇ R and T ⁇ T reception side and transmission side conversion matrices, respectively.
- a predicted value of MIMO CSI (MIMO CSI predicted value) is obtained by performing AR-LP on each component of transformed MIMO CSI and performing inverse transformation as in the following equation.
- Y ⁇ (t + ⁇ ) is the predicted value in the conversion MIMO CSI.
- FIG. 3 is a functional block configuration diagram showing the configuration of the channel prediction unit 9.
- the channel prediction unit 9 includes a direction estimation unit 91, an array response matrix calculation unit 92, a transformation matrix calculation unit 93, a CSI conversion unit 94, a prediction filter coefficient calculation unit 95, a linear prediction unit 96, and a prediction CSI.
- An inverse conversion unit 97 is included.
- the direction estimation unit 91 estimates an arrival direction (DoA) and a radiation direction (DoD) of a radio signal (elementary wave) using a direction estimation technique.
- the array response matrix calculation unit 92 calculates an array response matrix from the arrival direction (DoA) and the radiation direction (DoD) estimated by the direction estimation unit 91. For example, when the ESPRIT algorithm is used as the direction estimation technique, the arrival direction of the elementary wave is obtained, and the array response of the antenna array is obtained from the arrival direction. Such an array response is associated with the direction of arrival of the elementary waves. Alternatively, when the minimum norm method is used as the direction estimation technique, the array weight associated with the arrival direction is obtained instead of the array response.
- the conversion matrix calculation unit 93 calculates a conversion matrix based on the array response or array weight of the antenna array.
- the transformation matrix preferably has a good condition, that is, the product of the norm of the matrix and the norm of the inverse matrix is small.
- a unitary matrix is used as such a transformation matrix.
- the transformation matrix calculation unit 93 performs QR decomposition of the calculated array response matrix and converts the unitary matrix obtained by the QR decomposition. Used as a matrix.
- the CSI conversion unit 94 multiplies the MIMO CSI estimated by the channel estimation unit 8 and the conversion matrix calculated by the conversion matrix calculation unit 93 to obtain a conversion MIMO CSI.
- the number of elementary waves in the component is reduced compared to MIMO CSI.
- each component (h 11 , h 12 , h 21 , h 22 ) of the MIMO CSI includes the influence of the elementary waves # 1 to # 3.
- a transformation MIMO CSI as shown in FIG. 4C is obtained.
- the component y 22 is affected by the wave # 1
- the component y 21 is affected by the waves # 1 and # 2
- the component y 12 is the wave # 1 and # influenced by 3
- component y 11 is affected by the rays # 1 to # 3.
- the prediction filter coefficient calculation unit 95 first calculates a prediction filter coefficient using a low elementary wave element having the smallest number of elementary waves among the components of the transformed MIMO CSI. With respect to low-element elements, since the number of element waves is small, high-precision prediction is possible. In the example of FIG. 4C, the component y 22 is a low element element having the smallest number of element waves.
- the linear prediction unit 96 performs channel prediction using the prediction filter coefficient for the low wave element.
- the prediction filter coefficient calculation unit 95 and the linear prediction unit 96 use the prediction filter coefficient for the component y 22 for the component y 12 with the next smallest number of elementary waves, and perform channel processing. Run predictions. Channel prediction is performed not only for the prediction target time but also for the estimated time. Then, the component y 12, by subtracting the prediction value for the estimated time obtained by channel estimation, suppresses the component of rays # 1 in component y 12.
- the component y 12 which components of rays # 1 is suppressed, the prediction filter coefficients are calculated, predicted values are calculated.
- the predicted value is added to the prediction value obtained by the channel prediction using prediction filter coefficients for the component y 22.
- SC serial canceller
- the prediction CSI inverse conversion unit 97 calculates the MIMO CSI prediction value by inversely converting the prediction value of the conversion MIMO CSI obtained in this way.
- this inverse matrix When this inverse matrix is used as a transformation matrix, the received array response submatrix is diagonalized, and the number of rays existing in each component of the transformation MIMO CSI is reduced.
- this inverse matrix generally has a poor condition number. Therefore, by performing QR decomposition on the equation (16), it is decomposed into a product of an R ⁇ R unitary matrix expressed by the following equation and an upper triangular matrix of R ⁇ (R ⁇ 1).
- the recipient array response sub-matrix can be converted into upper triangular matrix R r.
- a MIMO CSI conversion is performed using the Rth column of Q r , a conversion MISO CSI is obtained. Since the elementary waves 1 to R-1 are suppressed by nulls, it is assumed that T-1 DoDs can be newly estimated from other elementary waves. In other words, the converted wave amplitude
- MIMO CSI is converted as follows.
- the prediction value of MIMO ⁇ CSI (MIMO CSI prediction value) is obtained by inversely transforming the prediction value of transformation MIMO ⁇ ⁇ CSI.
- [ ⁇ ] X, y means the (x, y) component of the matrix.
- the second term on the right side is a replica of the elementary wave R + T-1 to L contained in the (R-1, T) component. is there. Therefore, linear prediction corresponding to the CSI estimated time is also performed, and when the predicted value is subtracted from the estimated value,
- AR-LP is applied to the CSI in (25) and the next predicted value
- the reason for () is to guarantee the operation in the case where the null by the transmission side conversion matrix is turned to the elementary wave whose DoA is estimated on the reception side. Ideally, AR-LP is only performed on LR ⁇ T + 2 elementary waves at the maximum, and an improvement in prediction performance can be expected.
- FIG. 5 is a flowchart showing the overall operation of the multi-antenna transmission system 1.
- step S100 the reception processing unit 6 of the receiver 200 demodulates the radio signals (reception signals) received by the reception antennas # 1 to #R and outputs reception data.
- step S200 the channel estimation unit 8 estimates MIMO ⁇ CSI.
- step S300 the channel prediction unit 9 calculates a MIMO CSI prediction value by a beam space linear prediction process.
- step S400 the transmitter 100 performs transmission data control, for example, adaptive modulation control or transmission beamforming by precoding, based on the MIMO-CSI prediction value calculated by the channel prediction unit 9.
- transmission data control for example, adaptive modulation control or transmission beamforming by precoding
- FIG. 6 is a flowchart showing the operation of the channel prediction unit 9.
- step S301 the direction estimation unit 91 estimates the arrival direction (DoA) and the radiation direction (DoD) of the elementary wave using the direction estimation technique.
- step S302 the array response matrix calculation unit 92 calculates an array response matrix from the DoA / DoD estimated by the direction estimation unit 91.
- step S303 the transformation matrix calculation unit 93 calculates a transformation matrix from the array response matrix calculated by the array response matrix calculation unit 92.
- step S304 the CSI conversion unit 94 multiplies the MIMO CSI estimated by the channel estimation unit 8 and the conversion matrix calculated by the conversion matrix calculation unit 93 to obtain a conversion MIMO CSI.
- Loop B is a loop that uses the calculated prediction filter coefficient. For example, in FIG. 4D, the loop is omitted for the component y 22 , the loop is executed once for the components y 12 and y 21 , and the loop is executed twice for the component y 11. become. Since the prediction process starts from the low elementary wave element (component) of the transformed MIMO CSI, steps S307 to S309 will be described first.
- step S307 the prediction filter coefficient calculation unit 95 calculates a prediction filter coefficient for a prediction target component in the transformed MIMO CSI.
- step S308 the linear prediction unit 96 performs linear prediction on the component to be predicted in the transformed MIMO-CSI using the prediction filter coefficient calculated by the prediction filter coefficient calculation unit 95.
- step S309 the linear prediction unit 96 calculates a prediction value for a prediction target component in the transformed MIMO-CSI by linear prediction.
- step S305 the linear prediction unit 96 performs linear prediction on the next prediction target component of the transformed MIMO CSI using the prediction filter coefficient calculated in step S307. Note that this linear prediction is performed not only for the predicted time (future time to be predicted) but also for the estimated time (time when MIMO CSI is estimated).
- step S306 the linear prediction unit 96 subtracts the predicted value (predicted value with respect to the estimated time of MIMOSICSI) obtained using the calculated prediction filter coefficient from the pair as the next prediction target component.
- the predicted value predicted value with respect to the estimated time of MIMOSICSI
- step S308 the linear prediction unit 96 performs linear prediction using the prediction filter coefficient for the next component to be predicted whose number of rays is suppressed, and calculates a prediction value.
- step S309 the linear prediction unit 96 uses the predicted value obtained by using the calculated prediction filter coefficient (predicted value by the first prediction) for the component to be predicted next, and after suppressing the wave.
- the predicted value (predicted value by the second prediction) is added to obtain the final predicted value for the component that is the next prediction target.
- step S310 the prediction CSI inverse conversion unit 97 calculates the MIMO CSI prediction value by performing inverse conversion on the prediction value for each component of the conversion MIMO CSI.
- AR-LP Linear prediction based on an AR model, which does not use a beam space conversion matrix (prior art 1)
- B BS-AR: Linear prediction based on an AR model, using a simple beam space transformation matrix (prior art 2)
- C DBS-AR: Linear prediction based on an AR model, using a beam space (DBS) transformation matrix based on DoA / DoD (this embodiment)
- D DBS-AR w / SC: Linear prediction based on the AR model, using a beam space (DBS) transformation matrix based on DoA / DoD, and introducing a serial canceller (SC) (this implementation Form)
- the above method is evaluated by computer simulation.
- the channel prediction error with respect to the delay time ⁇ is evaluated by the root mean square error (RMSE) defined by the following equation.
- RMSE root mean square error
- the signal-to-noise power ratio (SNR) of the estimated CSI is defined as follows:
- Evaluation is performed with 1000 snaps.
- the DoA / DoD of each elementary wave and the moving direction of the receiver 200 are randomly set according to a uniform distribution of 360 degrees, and the complex scattering coefficient is randomly set according to a complex Gaussian distribution with an average of 0 and a variance of 1.
- the set parameters are constant in the snap.
- the number of rays is 10, the maximum Doppler frequency is 100Hz, and the estimated CSI rate is 500Hz.
- Prediction filter coefficients are calculated from the estimated CSI of 1000 samples, and the prediction error is evaluated in the subsequent 100 samples of CSI.
- the estimated CSI SNR is 30 dB.
- Fig. 7 shows the prediction characteristics when the number of antennas is 4x4 and the order of the prediction filter is 20.
- the simple BS-AR improves the prediction performance by reducing the number of rays.
- the prediction performance is improved by actively reducing the number of rays.
- the number of rays does not decrease sufficiently, so the degree of deterioration with respect to the delay time has not improved much.
- the prediction performance characteristics with respect to an increase in delay time are greatly improved.
- Fig. 8 shows the RMSE characteristics of each component of transformation MIMOMICSI.
- the prediction performance is not improved in the upper left component, but by introducing the serial canceller, high prediction performance improvement is obtained in any component.
- Figure 9 shows the characteristics when the number of antennas is 2 ⁇ 2 to 5 ⁇ 5.
- Figure 10 shows the RMSE characteristics with respect to the order of the filter coefficient when the number of antennas is 4x4 and the delay time is 10 msec.
- the system configuration in which the transmitter 100 and the receiver 200 each have a plurality of antennas has been described.
- the transmitter 100 has one antenna and the receiver 200 has a plurality of antennas.
- the present invention can be applied to a SIMO system or a MISO system in which the transmitter 100 has a plurality of antennas and the receiver 200 has one antenna.
- the unitary matrix is used as the transformation matrix.
- other matrices may be used as long as the matrix is well-conditioned.
- the “good condition” refers to the following case, for example. That is, for matrix A, the condition number is
- indicates the norm of the matrix.
- a “good condition” can be considered if the condition number is 5 or less, and a “bad condition” if it is larger than that.
- the condition number is also 1, which is a “good condition”.
- the channel information prediction system and the channel information prediction method according to the present invention can further improve the prediction performance of channel information, and thus are useful in wireless communication such as mobile communication.
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Power Engineering (AREA)
- Quality & Reliability (AREA)
- Physics & Mathematics (AREA)
- Electromagnetism (AREA)
- Radio Transmission System (AREA)
Abstract
Description
T. Eyceoz, S. Hu, and A. Duel-Hallen, "Performance Analysis of Long Range Prediction for Fast Fading Channels," Proc. of 33rd Annual Conf. on Inform. Sciences and Systems CISS'99, Vol. II, pp. 656 - 661、1999年3月 T. Svantesson, A. L. Swindlehurst,"A Performance Bound for Prediction of MIMO Channels", IEEE Trans. Signal Process., vol.54, no.2, pp.520-529、2006年2月
図1は、本実施形態に係るマルチアンテナ伝送システム1の全体概略構成図である。図1に示すように、マルチアンテナ伝送システム1は、送信機100、受信機200及びチャネル情報予測システム10を含む。
引き続き図1を参照して、本実施形態に適用されるチャネルモデルについて説明する。
ここでは、本実施形態で共通して使用するAR-LPと、ビームスペース変換について説明する。
チャネル係数がARモデルに従うと想定することは、多くのチャネル予測にける共通のアプローチである。AR-LPは、チャネルの自己相関特性におけるサイドローブを利用することで実現されている。MIMO CSIの推定時刻と送信時刻の関係を図2に示す。
無限数の素波がレイリーフェージングを形成するチャネルの場合、その自己相関特性は第1種0次ベッセル関数で表される。一方で、実環境、特に端末の移動が予想される屋外環境のチャネル変動は、電力の大きい有限数の素波による影響が支配的となる。
図3は、チャネル予測部9の構成を示す機能ブロック構成図である。
ここでは、本実施形態に係る変換行列、および、線形予測手法について説明する。本実施形態では、推定されたMIMO CSIに対して方向推定技術を用いることにより、それぞれR-1個、T-1個の素波のDoA/DoD推定が可能であるものとする。
ここでは、推定された素波のDoA/DoDに基づき、変換MIMO CSIの各成分に存在する素波数を減少させることができ、かつ、条件数の良い変換行列を提案する。
一方で、アレイ応答小行列は対角化されず、三角化されるのみであるため、変換MIMO CSIの左上成分では、存在する素波数があまり減少しない。このため、予測性能が十分には改善しない。
次に、マルチアンテナ伝送システム1の動作について説明する。
図5は、マルチアンテナ伝送システム1の全体動作を示すフローチャートである。
図6は、チャネル予測部9の動作を示すフローチャートである。
ステップS305において、線形予測部96は、変換MIMO CSIのうち次の予測対象となる成分に対し、ステップS307で計算された予測フィルタ係数を用いて線形予測を行う。なお、この線形予測は、予測時刻(予測の対象となる未来の時刻)に対してだけでなく、推定時刻(MIMO CSIを推定した時刻)に対しても行う。
再度ステップS307において、予測フィルタ係数計算部95は、素波の数が抑圧された次の予測対象となる成分について、予測フィルタ係数を計算する。
ステップS309において、線形予測部96は、次の予測対象となる成分について、計算済みの予測フィルタ係数を用いて得られた予測値(第1の予測による予測値)と、素波の抑圧後の予測値(第2の予測による予測値)とを加算し、当該次の予測対象となる成分についての最終的な予測値を得る。
次に、本実施形態によって得られる効果について、比較例を挙げて説明する。具体的には、次の(a)~(d)の各チャネル予測方式を比較して説明する。
(b)BS-AR:ARモデルに基づく線形予測であって、シンプルなビームスペース変換行列を使用する方式(従来技術2)
(c)DBS-AR:ARモデルに基づく線形予測であって、DoA/DoDに基づくビームスペース(DBS)変換行列を使用する方式(本実施形態)
(d)DBS-AR w/SC:ARモデルに基づく線形予測であって、DoA/DoDに基づくビームスペース(DBS)変換行列を使用し、かつ、シリアルキャンセラ(SC)を導入した方式(本実施形態)
上記の方式を、計算機シミュレーションによって評価する。遅延時間τに対するチャネル予測誤差を、次式で定義するルート平均二乗誤差(RMSE)によって評価する。
上述したように、本発明の一実施形態を通じて本発明の内容を開示したが、この開示の一部をなす論述及び図面は、本発明を限定するものであると理解すべきではない。この開示から当業者には様々な代替実施の形態が明らかとなろう。
る。そして、ユニタリ行列の場合、ユークリッドノルムが共に1であるため、条件数も1であり、「良条件」である。
Claims (5)
- 無線信号の送信側または受信側の少なくとも何れか一方において複数のアンテナが用いられるマルチアンテナ伝送システムにおいて、受信した無線信号に含まれる干渉成分の抑圧に用いられる予測フィルタ係数を計算するチャネル情報予測システムであって、
前記送信側における何れかの送信アンテナと、前記受信側における何れかの受信アンテナとの間における伝搬路特性を示すチャネル情報によって区別される複数のアンテナ間要素のうち、前記無線信号に含まれる素波の数が所定数よりも少ない前記アンテナ間要素である低素波要素を用いて前記予測フィルタ係数を計算するフィルタ係数計算部と、
前記フィルタ係数計算部によって計算された第1の前記予測フィルタ係数を用いて、前記無線信号の未来の送信時における前記チャネル情報を予測する第1の予測を実行し、かつ、前記低素波要素に含まれている前記素波の成分を抑圧するチャネル情報予測部と
を備え、
前記フィルタ係数計算部は、前記チャネル情報予測部で前記素波の成分が抑圧された低素波成分を用いて第2の前記予測フィルタ係数を計算し、
前記チャネル情報予測部は、前記第2の予測フィルタ係数を用いて、前記無線信号の未来の送信時における前記チャネル情報を予測する第2の予測を実行するチャネル情報予測システム。 - 前記チャネル情報予測部は、前記第2の予測フィルタ係数を、前記第1の予測フィルタ係数として用いて、次の前記第1の予測を実行し、かつ前記素波の成分を抑圧する請求項1に記載のチャネル情報予測システム。
- 前記フィルタ係数計算部は、前記素波の数が最も少ない前記アンテナ間要素を前記低素波要素として用いる請求項1または2に記載のチャネル情報予測システム。
- 前記チャネル情報予測部は、前記予測フィルタ係数に基づいて、前記低素波要素と異なる他の前記アンテナ間要素に含まれている前記素波の成分を抑圧し、前記素波の成分が抑圧された前記アンテナ間要素を用いて、前記無線信号の未来の送信時における前記チャネル情報を予測する請求項1または2に記載のチャネル情報予測システム。
- 無線信号の送信側または受信側の少なくとも何れか一方において複数のアンテナが用いられるマルチアンテナ伝送システムにおいて、受信した無線信号に含まれる干渉成分の抑圧に用いられる予測フィルタ係数を計算するチャネル情報予測方法であって、
前記送信側における何れかの送信アンテナと、前記受信側における何れかの受信アンテナとの間における伝搬路特性を示すチャネル情報によって区別される複数のアンテナ間要素のうち、前記無線信号に含まれる素波の数が所定数よりも少ない前記アンテナ間要素である低素波要素を用いて前記予測フィルタ係数を計算するステップと、
前記計算された第1の前記予測フィルタ係数を用いて、前記無線信号の未来の送信時における前記チャネル情報を予測する第1の予測を実行し、かつ、前記低素波要素に含まれている前記素波の成分を抑圧するステップと、
前記素波の成分が抑圧された低素波成分を用いて第2の前記予測フィルタ係数を計算するステップと、
前記第2の予測フィルタ係数を用いて、前記無線信号の未来の送信時における前記チャネル情報を予測する第2の予測を実行するステップと
を備えるチャネル情報予測方法。
Priority Applications (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2009801067877A CN101960760A (zh) | 2008-02-27 | 2009-02-26 | 信道信息预测***以及信道信息预测方法 |
US12/920,087 US8462869B2 (en) | 2008-02-27 | 2009-02-26 | Channel information prediction system and channel information prediction method |
KR1020107019061A KR101088602B1 (ko) | 2008-02-27 | 2009-02-26 | 채널 정보 예측 시스템 및 채널 정보 예측 방법 |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2008046928A JP4481336B2 (ja) | 2008-02-27 | 2008-02-27 | チャネル情報予測システム及びチャネル情報予測方法 |
JP2008-046928 | 2008-02-27 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2009107732A1 true WO2009107732A1 (ja) | 2009-09-03 |
Family
ID=41016119
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2009/053577 WO2009107732A1 (ja) | 2008-02-27 | 2009-02-26 | チャネル情報予測システム及びチャネル情報予測方法 |
Country Status (5)
Country | Link |
---|---|
US (1) | US8462869B2 (ja) |
JP (1) | JP4481336B2 (ja) |
KR (1) | KR101088602B1 (ja) |
CN (1) | CN101960760A (ja) |
WO (1) | WO2009107732A1 (ja) |
Families Citing this family (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
FR2936121B1 (fr) * | 2008-09-15 | 2010-11-05 | Commissariat Energie Atomique | Methode d'estimation aveugle de parametres de modulation ofdm selon un critere de maximun de vraisemblance. |
FR2936122B1 (fr) * | 2008-09-15 | 2010-10-15 | Commissariat Energie Atomique | Methode d'estimation aveugle de parametres de signal ofdm par filtrage adapte. |
US9219533B2 (en) * | 2011-10-25 | 2015-12-22 | Transpacific Ip Management Group Ltd. | Systems and methods for downlink scheduling in multiple input multiple output wireless communications systems |
US9456369B2 (en) | 2012-07-31 | 2016-09-27 | Hewlett Packard Enterprise Development Lp | Management of modulation and coding scheme implementation |
CN103760519B (zh) * | 2014-01-24 | 2016-02-03 | 深圳大学 | 高分辨率doa估计方法及*** |
WO2015169358A1 (en) * | 2014-05-08 | 2015-11-12 | Huawei Technologies Co., Ltd. | Channel estimation in wireless communication network node |
CN107182068A (zh) * | 2016-03-11 | 2017-09-19 | 索尼公司 | 用于无线通信的装置和方法、参数优化装置和方法 |
EP3605867A4 (en) | 2017-03-21 | 2021-03-10 | Mitsubishi Electric Corporation | COMMUNICATION SYSTEM |
WO2022162152A2 (en) * | 2021-01-29 | 2022-08-04 | Telefonaktiebolaget Lm Ericsson (Publ) | Method for transmitting link adaptation state information in telecommunication networks |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2006303625A (ja) * | 2005-04-15 | 2006-11-02 | Matsushita Electric Ind Co Ltd | 無線送信器、無線送信器のバースト送信方法、プログラムおよび記録媒体 |
WO2007007249A2 (en) * | 2005-07-08 | 2007-01-18 | Koninklijke Philips Electronics N.V. | Transmission over a multiple input multiple output broadcast channel (mimo-bc) |
JP2008193340A (ja) * | 2007-02-02 | 2008-08-21 | Matsushita Electric Ind Co Ltd | 無線基地局装置、無線端末装置、無線通信システム、及びチャネルクオリティインジケータ推定方法 |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2009505560A (ja) * | 2005-08-19 | 2009-02-05 | ミツビシ・エレクトリック・リサーチ・ラボラトリーズ・インコーポレイテッド | エラーを有するフィードバックによる送信アンテナ選択のための最適なシグナリング及び選択検証 |
EP2060019A1 (en) * | 2006-08-31 | 2009-05-20 | The Governors Of The University Of Alberta | Decision-feedback detection for block differential space-time modulation |
-
2008
- 2008-02-27 JP JP2008046928A patent/JP4481336B2/ja active Active
-
2009
- 2009-02-26 KR KR1020107019061A patent/KR101088602B1/ko not_active IP Right Cessation
- 2009-02-26 WO PCT/JP2009/053577 patent/WO2009107732A1/ja active Application Filing
- 2009-02-26 CN CN2009801067877A patent/CN101960760A/zh active Pending
- 2009-02-26 US US12/920,087 patent/US8462869B2/en not_active Expired - Fee Related
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2006303625A (ja) * | 2005-04-15 | 2006-11-02 | Matsushita Electric Ind Co Ltd | 無線送信器、無線送信器のバースト送信方法、プログラムおよび記録媒体 |
WO2007007249A2 (en) * | 2005-07-08 | 2007-01-18 | Koninklijke Philips Electronics N.V. | Transmission over a multiple input multiple output broadcast channel (mimo-bc) |
JP2008193340A (ja) * | 2007-02-02 | 2008-08-21 | Matsushita Electric Ind Co Ltd | 無線基地局装置、無線端末装置、無線通信システム、及びチャネルクオリティインジケータ推定方法 |
Non-Patent Citations (4)
Title |
---|
"Acoustics, Speech, and Signal Processing, 2005. Proceedings. (ICASSP '05). IEEE International Conference on, 2005.03.18", article SVANTESSON T. ET AL.: "capacity ot spatio-temporally structured MIMO channels with estimation errors", pages: 401 - 404 * |
"Personal, Indoor and Mobile Radio Communications, 2008. PIMRC 2008. IEEE 19th International Symposium on, 2008.09.15", article KENTA OKINO ET AL.: "Direction based beamspace MIMO channel prediction with ray cancelling", pages: 1 - 5 * |
LARSEN M.D. ET AL.: "A performance bound for MIMO-OFDM channel estimation and prediction", SENSOR ARRAY AND MULTICHANNEL SIGNAL PROCESSING WORKSHOP, 2008. SAM 2008. 5TH IEEE, 21 July 2008 (2008-07-21), pages 141 - 145 * |
SAVAZZI S. ET AL.: "A Comparative Analysis of Spatial Multiplexing Techniques for Outdoor MIMO-OFDM Systems with a Limited Feedback Constraint", VEHICULAR TECHNOLOGY, IEEE TRANSACTIONS ON, January 2009 (2009-01-01), pages 218 - 230 * |
Also Published As
Publication number | Publication date |
---|---|
CN101960760A (zh) | 2011-01-26 |
US8462869B2 (en) | 2013-06-11 |
US20110090980A1 (en) | 2011-04-21 |
JP4481336B2 (ja) | 2010-06-16 |
JP2009206833A (ja) | 2009-09-10 |
KR20100116201A (ko) | 2010-10-29 |
KR101088602B1 (ko) | 2011-11-30 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP5073809B2 (ja) | チャネル情報予測システム及びチャネル情報予測方法 | |
JP4481336B2 (ja) | チャネル情報予測システム及びチャネル情報予測方法 | |
US8417191B2 (en) | Method and system for beamforming communication in high throughput wireless communication systems | |
US7822141B2 (en) | Multiple input, multiple output communications systems | |
EP2741429B1 (en) | Method for determining beamforming parameters in a wireless communication system and to a wireless communication system | |
US8040278B2 (en) | Adaptive antenna beamforming | |
US8638871B2 (en) | System and method for combination multiple input, multiple output (MIMO) and beamforming | |
US8175184B2 (en) | Method for transmitting beam forming information and a method for tracking position of a mobile station in multi input multi output system using codebook-based beam forming scheme | |
JP4763386B2 (ja) | Mimoフェージングシミュレータ | |
US8606175B2 (en) | RF relay of full-duplex and method for removing interference of EM level thereof | |
US8274431B1 (en) | Beamforming with partial channel knowledge | |
Withers et al. | Echo-MIMO: A two-way channel training method for matched cooperative beamforming | |
US20100323630A1 (en) | Reception apparatus, mobile equipment, communication system, and communication method | |
US7738595B2 (en) | Multiple input, multiple output communications systems | |
US8014981B2 (en) | Angular-domain channel model and channel estimation | |
JP2006025328A (ja) | 空間多重伝送用送信方法および装置 | |
Okino et al. | Direction based beamspace MIMO channel prediction with ray cancelling | |
Singh et al. | Beamforming Showing Effect on BER with Change in Antenna Configuration | |
Fedosov et al. | Investigation of the Influence of Spatial Correlation on the Performance of the MIMO System When Using the Adaptation Algorithm |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
WWE | Wipo information: entry into national phase |
Ref document number: 200980106787.7 Country of ref document: CN |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 09714798 Country of ref document: EP Kind code of ref document: A1 |
|
ENP | Entry into the national phase |
Ref document number: 20107019061 Country of ref document: KR Kind code of ref document: A |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
WWE | Wipo information: entry into national phase |
Ref document number: 12920087 Country of ref document: US |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 09714798 Country of ref document: EP Kind code of ref document: A1 |