KR20020041217A - Method and apparatus for semi-blind transmit antenna array employing chip level forward link channel estimation - Google Patents

Abstract

PURPOSE: A semi-blind transmit antenna apparatus adopting chip level forward channel estimation in a mobile system and a method thereby are provided to effectively carry out channel estimation and data detection by directly executing chip level signal processing for forward signals without going through a despreading step. CONSTITUTION: A terminal forward processor(387) is composed of a receiving RF part(341), a forward fading estimator(343), a data determiner(345), and a decoder(347). The receiving RF part(341) converts a forward multipath fading channel signal received through a duplexer(337) into a baseband signal. The forward fading estimator(343) estimates and outputs forward fading for each path using the output of the receiving RF part(341). The data determiner(345) determines the data of the received signal using the output of the forward fading estimator(343). The decoder(347) decodes the determined data and restores a received message(348). A terminal signal processor(389) is composed of the forward fading estimator(343), the data determiner(345), and a forward fading encoder(349). The forward fading encoder(349) collects, encodes and outputs the forward fading coefficients estimated from the forward fading estimator(343). A terminal reverse processor(391) is comprised of an encoder(329), a multiplexer(331), a spreader(333), and a transmitting RF part(335). The encoder(329) encodes and outputs a transmitting message(328). The multiplexer(331) multiplexes and outputs the output of the forward fading encoder(349) and the output of the encoder(329). The spreader(333) spreads and outputs the output of the multiplexer(331). The transmitting RF part(335) converts the output of the spreader(333) into an RF signal.

Description

Semi-blind transmission antenna array device and method employing chip-level forward channel estimation in mobile communication system

The present invention relates to a transmission antenna array apparatus and method for a code division multiple access communication system, and more particularly, to a semi-blind transmission antenna array apparatus and method employing chip level forward channel estimation.

In order to meet the demand of Code Division Multiple Access (CDMA) mobile communication, which has shown a sharp increase in recent years, smart antennas that use beamforming using a plurality of base station antennas have been researched and developed and are ready for commercialization. In particular, since next-generation mobile communication has to provide various high-speed multimedia services, attention is focused on a forward beamforming technique that enables high-speed data transmission required for this. A method for forward beamforming using a smart antenna has already been proposed in the following literature.

(Reference 1: J. S. Thompson, J. E. Hudson, P. M. Grant, and B. Mulgrew, "CDMA Downlink Beamforming for Frequency Selective Channels," PIMRC'99, B2-3, Osaka, Japan, September 1999)

Figure 1 shows the structure of the Thompson's transmit array antenna. Looking at the operation of the transmission antenna array structure of the Thompson, the transmission message is formed in the transmission beam former 105 of the forward processor 131 in a specific transmission direction, and is propagated to the air through the antenna array 111. The reverse processor 133 receives and processes a signal of a reverse channel received through the antenna array 111. The forward fading power calculator 119 estimates a path fading coefficient of the signal received from the intermediate output of the reverse processor 133 and obtains an average power of the estimated fading coefficient. The average fading power is calculated using the assumption that the backward average fading power is equal to the forward average fading power. The array vector calculator 123 calculates an input direction (array vector) of a signal received from the intermediate output of the reverse processor 133. The transmission correlation matrix calculator 125 calculates a transmission correlation matrix using the calculated average fading power information and the array vector, and the weight vector calculator 127 calculates a weight vector using the calculated transmission correlation matrix. This is output to the transmission beam former 105. Accordingly, the transmission beam former 105 controls the formation of the transmission beam by adding a weight of the transmission signal to be output to the corresponding antenna element according to the weight vector.

Thompson's transmission antenna array system having a structure as shown in FIG. 1 is summarized as follows. First, the input direction (or array vector) of the signal received by the receiving antenna array is estimated. Second, the fading coefficient for each path of the received signal received by the receiving antenna array is estimated, and the average power thereof is obtained. Third, the weight vector for the transmission antenna array is calculated using the direction information and the average fading power information. Fourth, a method of transmitting a signal by controlling transmission beam formation using the calculated weight vector.

This method is called a blind transmission antenna array method as a forward beamforming method based on the reverse channel characteristics estimated from the reverse received signals without knowing the forward channel characteristics accurately. However, in the case of FDD (Frequency Division Duplex) communication systems having different transmission / reception frequency bands, transmission / reception channel characteristics are different so that the blind transmission method has a limitation in forming a transmission beam. That is, the base station needs to know the characteristics of the forward channel for faster forward data transmission, and is a semi-blind transmission antenna array method in which the base station forms a transmission beam using information fed back from the terminal station. The semi-blind transmission antenna array method has already been proposed in the following document.

(Reference Document 2): Choi Jin-ho, Chun Byung-jin, "Semi-intentional transmission antenna array apparatus and method using feedback information in a mobile communication system", Korean Patent Application No. P2000-11617.

FIG. 2 shows the structure of a transmit antenna array disclosed in the above-mentioned prior application patent P2000-1 1 617. FIG. The operation of the transmit antenna array structure shown in FIG. 2 is summarized as follows. The terminal station separates each multipath component from the forward signal received through the single antenna 239 using the rake receiver 243 for the forward channel estimation. The forward fading estimator 247 estimates the forward channel for each path, encodes it in the forward fading encoder 249, and feeds it back to the base station through the reverse channel. Then, the base station decodes the forward channel information for each path fed back from the terminal station in the forward channel decoder 219 and calculates the forward fading power for each path in the forward fading power calculator 221. Meanwhile, the array vector calculator 223 calculates an array vector for each path. Thereafter, the transmission correlation matrix calculator 225 calculates the transmission correlation matrix using the forward fading power and the array vector, and calculates the transmission weight vector in the weight vector calculator 227. In addition, the beamforming of the transmission signal is controlled by the transmission beam former 205 using the weight vector, and then transmitted to the desired terminal station through the antenna array 211.

However, in the case of code division multiple access (CDMA) communication, the spreading rate is lowered to a certain level (for example, 10) or less in order to increase the data rate while maintaining the chip rate constant. There is a number to fall. At this time, if the terminal station uses the rake receiver as it is, the interference signal components such as ISI (Inter Symbol Interference) or MAI (Mulitple Access Interference) are larger than the desired signal components of the despreader output, so that channel estimation is not properly performed. As a result, the anti-blind forward beamforming method is dependent on an incorrect channel estimate value, so that the effect of the rake receiver is greatly reduced. That is, in the fast forward data transmission environment, it can be seen that it is inappropriate for the terminal station to use an existing rake receiver for forward channel estimation and data detection.

Accordingly, an object of the present invention is to provide an apparatus and method for performing channel estimation and data detection for each path by performing chip level signal processing on a forward signal without despreading.

Another object of the present invention is to provide a semi-blind transmission antenna array apparatus and method employing chip level forward channel estimation.

According to a first aspect of the present invention for achieving the above objects, in a mobile communication system, a base station apparatus is fed back with forward fading information estimated at a chip level of a forward signal from a terminal station and received with the forward fading information. and a base station signal processor for generating a weight vector for the transmission beam forming with the estimation of the array vector from the reverse signal, and a transmit beamformer which controls the transmission beam forming of a transmission message to have the weight vector; A forward fading estimator for estimating forward fading information for each path at a chip level of the forward signal from the base station, a forward channel encoder for collecting and encoding the obtained forward fading information for each path, and the encoded And a reverse processor which multiplexes forward fading information with a transmission message and transmits the same to the base station.

According to a second aspect of the present invention, there is provided a communication method in a mobile communication system in which a base station transmits a signal having a directionality through an antenna array to a specific terminal station, wherein the terminal station is configured at a chip level of a forward signal from the base station. Estimating forward fading information for each path, combining and encoding the estimated forward fading information for each path, and transmitting the same to the base station, and the base station extracts the forward fading information from a reverse signal from the terminal station. Calculating a forward fading power using the extracted forward fading information, the base station estimating an array vector with a backward signal from the terminal station, and the base station calculates the forward fading power and the array vector. Generating a weight vector for forming a transmission beam with It characterized in that the base station comprises the step of transmitting on the antenna array to control the transmission beam of the transmission message to have the weight vector.

1 is a diagram illustrating a conventional transmission antenna array proposed by Thompson.

2 is a diagram showing a conventional transmission antenna array proposed by Choi.

3 is a diagram illustrating a configuration of a base station according to an embodiment of the present invention.

4 is a diagram illustrating a terminal station configuration according to an embodiment of the present invention.

5 is a diagram illustrating a base station procedure according to an embodiment of the present invention.

6 is a diagram illustrating a terminal station procedure according to an embodiment of the present invention.

7 illustrates an EM based channel estimation process.

8 is a diagram illustrating a process of calculating forward fading power for each path.

9 illustrates an HMM signal model for explaining an algorithm according to an embodiment of the present invention.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. First of all, in adding reference numerals to the components of each drawing, it should be noted that the same reference numerals have the same reference numerals as much as possible even if displayed on different drawings. Also, in the following description, many specific details such as components of specific circuits are shown, which are provided to help a more general understanding of the present invention, and the present invention may be practiced without these specific details. It is self-evident to those of ordinary knowledge in Esau. In the following description of the present invention, if it is determined that a detailed description of a related known function or configuration may unnecessarily obscure the subject matter of the present invention, the detailed description thereof will be omitted.

In general, in a base station employing a transmission antenna array system, it is known that the weight vector for optimal transmission beamforming corresponds to an eigenvector corresponding to the maximum eigenvalue of the next transmission correlation matrix G. (See references 1 and 2)

here, Is an array-by-path array vector, Is the Angle of Arrival, Is the forward fading power for each path, M is the number of paths, Represents the Hermitiam operator, respectively. Therefore, the base station needs to know the array vector for each path and the forward fading power for each path for optimal transmission beam formation. Here, we assume the following Reciprocity between the forward and reverse channels.

(Assumption 1) The number of multipaths between the forward and reverse fading channels is the same, and the transmission and reception directions to and from each path are also the same.

(Assumption 2) In the FDD system, when the transmission / reception frequency band difference is larger than the correlation band, the instantaneous fading coefficients for the same path in the forward and reverse directions are independent of each other.

According to the reversibility, an array vector for each path (equivalently, an angle of arrival for each path) can be estimated from an uplink signal received from a terminal station (provision 1). Since the estimation is difficult (assumption 2), it is necessary for the terminal station to receive the forward signal from the base station, estimate the forward channel characteristics for each path, and then feed it back to the base station.

However, in order for the terminal station to effectively perform forward channel estimation during high-speed forward data transmission, a signal processor that performs predetermined chip level signal processing without despreading is required instead of the conventional rake receiver that performs symbol level signal processing after despreading. Do. According to the necessity, the forward beamforming algorithm proposed in the present invention performs a forward channel estimation based on an EM (expectation maximization) method, and then feeds it back to the base station, and the base station forwards the received forward channel information and the feedback to the base station. It is a semi-blind forward beamforming algorithm that performs forward beamforming by obtaining the eigenvector of Equation 1 using the estimated path-by-path array vector information.

3 is a transmission array antenna system of a base station that receives a forward fading coefficient from a terminal station to obtain an optimal weight vector in a mobile communication system according to an embodiment of the present invention, and then efficiently forms a beam of a transmitted signal using the same. The structure of the is shown. On the other hand, Figure 4 is after receiving the forward transmission signal of the base station in the mobile communication system according to an embodiment of the present invention, after estimating the forward fading coefficient from the received signal, the estimated forward fading coefficient value to a specific reverse channel The structure of a terminal station transmitting to a base station through the network is shown. Here, the forward fading coefficient estimated by the terminal station may be a forward fading magnitude.

Referring to FIG. 3, the base station includes a base station forward processor 381 consisting of an encoder 301, a spreader 303, a transmit beamformer 305, a transmit RF unit 307, a forward channel decoder 319, A base station signal processor 383 consisting of a forward fading power calculator 321, an array vector calculator 323, a transmission correlation matrix calculator 325, a weight vector calculator 327, a receiving RF unit 313, and a lake receiver ( 315), base station reverse processor 385 composed of demultiplexer 316, decoder 317, etc., other duplexer 309, antenna array 311 composed of N identical antenna elements, and the like. In the drawing, the thick lines represent N signal lines required by using the antenna array.

An operation based on the configuration of the base station is as follows.

First, referring to an operation of the base station forward processor 381, the encoder 301 encodes the transmission message 300, and the spreader 303 spreads and outputs the output of the encoder 301. The transmission beam former 305 copies the output from the spreader 303 into a plurality of antenna elements so as to correspond to the plurality of antenna elements in order to efficiently communicate with the terminal station receiving the transmission message, and transmits the plurality of signals. And the weight vector to be applied to the forward multipath from the weight vector calculator 327 is output. The transmission RF unit 307 converts the output of the transmission beam former 305 into a radio frequency (RF) signal and outputs the radio frequency (RF) signal. The output of the transmitting RF unit 307 is separated into a transmission signal through the duplexer 309, forms a beam through the antenna array 311, and is transmitted to the corresponding terminal station.

Next, referring to the base station reverse direction processor 385, the reception RF unit 313 converts the received signal received through the duplexer 309 into a baseband signal and outputs the frequency. The rake receiver 315 despreads and outputs the multipath signal output from the reception RF unit 313. The demultiplexer 316 separates the channel signal including the forward fading information from the output from the rake receiver 315 and provides it to the forward channel decoder 319. The decoder 317 decodes and outputs the output of the demultiplexer 316.

Next, referring to the base station signal processor 383, the forward channel decoder 319 decodes the output of the rake receiver 315 and extracts the forward fading coefficient transmitted from the terminal station. Here, the fading coefficient may be a forward fading size estimated by the terminal station. The forward fading power calculator 321 calculates and outputs a forward fading power with the fading coefficient from the forward channel decoder 319. Here, the forward fading power calculator 321 is a forward fading coefficient extractor (not shown) for extracting a forward fading coefficient from the information from the forward channel decoder 319 and the forward fading power from the extracted forward fading coefficient value. It is composed of a power calculator (not shown) for calculating the. On the other hand, the forward fading power converter 321 has only the forward channel information is extracted on the forward channel repeat call group 319 as described above may compute a forward fading power, this as well as the predetermined signal process from the uplink signal received Additional fading power can be used to calculate forward fading power.

The forward channel information fed back from the terminal is complex information In this case, the forward fading coefficient extractor is a weight vector used by the base station in forming a corresponding transmission beam. Array vector estimates from the array vector calculator 323 Complex forward fading coefficient using Extract On the other hand, the size of the forward fading information fed back from the terminal information In this case, the forward fading coefficient extractor is a weight vector used by the base station in forming a corresponding transmission beam. Array vector estimates from the array vector calculator Forward fading size using Extract

The array vector calculator 323 inputs an intermediate output of the rake receiver 315 to derive an input direction (or derived from) of a received signal. Calculate and output the array vector). Here, the array vector calculator is an array vector directly from a reverse received signal.Can be generated after predetermined signal processing,Is generated after predetermined signal processing, and then array vectorYou can also calculate

The transmission correlation matrix calculator 325 obtains a transmission correlation matrix based on the fading power from the forward fading power calculator 321 and the array vector from the array vector calculator 323, and the weight vector calculator 327 calculates the transmission correlation matrix. A weight vector to be applied to the forward multipath is calculated and output using the transmission correlation matrix from the transmission correlation matrix calculator 325.

Referring to FIG. 4, the terminal station includes a receiving station RF processor 341, a forward fading estimator 343, a data determiner 345, a decoder 347, and a forward fading estimator. 343, a terminal station signal processor 389 consisting of a data determiner 345, a forward fading encoder 349, an encoder 329, a multiplexer 331, a spreader 333, and a transmitting RF unit 335 And a terminal station direction processor 391 and the like, a double unit 337, a single antenna 339, and the like.

First, referring to the terminal station forward processor 387, the reception RF unit 341 converts the downlink multipath fading channel signal received through the duplexer 337 into a baseband signal by outputting the frequency down-regulated signal. The forward fading estimator 343 has an output of the reception RF unit 341 and estimates and outputs forward fading for each path. The data determiner 345 determines the data of the received signal with the forward fading from the forward fading estimator 345, and decodes the determined data by the decoder 347 to recover the received message 348.

Next, referring to the terminal station signal processor 389, as described above, the forward fading estimator 343 estimates and outputs forward fading for each path. Here, the forward channel estimator uses complex forward fading information when the transmit beamformer of the base station forms the non-directional transmit beam. Or size forward fading information Estimate On the other hand, when the transmission beamformer of the base station forms the directional transmission beam, the forward fading coefficient , Angle of arrival , Array vector , Weight vector Complex forward channel information Or size forward channel information Estimate The data determiner 345 detects the forward transmission data. The forward fading encoder 349 collects, encodes, and outputs a forward fading coefficient for each path estimated from the forward fading estimator 343.

Next, referring to the terminal station reverse direction processor 391, the encoder 329 encodes and outputs the transmission message 328. The multiplexer 331 multiplexes the output of the forward fading encoder 349 and the output of the encoder 329. The diffuser 333 diffuses and outputs the output of the multiplexer 331. The transmitter RF unit 335 adjusts the frequency of the output of the spreader 333 to convert the radio frequency (RF) signal and outputs the radio frequency (RF) signal. The radio frequency signal is separated from the duplexer 337 into a transmission signal and transmitted to the base station through the single antenna 339.

Although not shown in the configuration of FIGS. 3 and 4, the encoders 301 and 329 include a convolutional encoder, a repeater, an interleaver, etc., and the transmitting RF units 307 and 335 include a D / A converter and an up frequency. The receiver RF unit 313, 341 includes a low noise amplifier, a downlink frequency converter, an A / D converter, and the like, and the rake receiver 315 includes a path finder, a despreader, a channel estimator, and the like. And a combiner, and the decoders 317 and 347 include a deinterleaver, a convolution decoder, and the like.

In the above configurations, the core configuration according to the present invention comprises a base station signal processor 383, a terminal station signal processor 389, a transmission beam former 305, and the like. Signal processor. " The rest are the components typically used in code division multiple access mobile communication systems. In addition, the multiplexer 331 and the demultiplexing 316 are usually formed using an orthogonal Walsh code in the case of code division multiple access communication.

5 and 6 illustrate a procedure performed in a base station and a terminal station having the structures of FIGS. 3 and 4. In order to clarify the characteristic elements of the present invention, only the generation of the encoded transmission data of the base station to the encoded data detection and forward channel estimation of the terminal station, and only the calculation of the weight vector of the base station from the generation of the forward channel estimation information of the terminal station do.

Referring to FIG. 5, the base station initializes the weight vector in step 403. )do. Then, at every i th transmission time, the base station transmits new transmission data in step 405. In step 407, the generated data is spread with a spreading rate L. Diffused signal Generates. In operation 409, the weight vector previously calculated on the spread signal. Forward beamforming signal to a specific terminal station by multiplying Is then transmitted to the space through the antenna array.

In step 411, the base station receives a reverse signal including forward channel vector information from the terminal station after a predetermined time (D unit time) delay after the forward signal transmission. The base station determines the array vector for each path from the reverse signal in step 417. And estimate the forward channel estimation information for each path from the reverse signal in step 419. Forward fading power by path by extracting Calculate

In step 421, the base station calculates a transmission correlation matrix according to Equation 1 using the array vector for each path and the forward fading power for each path, and uses the transmission correlation matrix for the next forward beamforming in step 423. Weight vector Calculate The weight vector mathematically corresponds to an eigenvector corresponding to the maximum eigenvalue of the transmission correlation matrix, and the method of calculating the weight vector includes a power calculation method, a Lagrangian calculation method, a Conjugate Gradient Method (CGM) calculation method, and a Tangential GradientMethod (TGM) calculation method. There are many things, so choose the right one.

Referring to FIG. 6, the terminal station initializes a channel vector in step 433. And the forward beamformed signal vector from the base station in step 435 at every i < th &gt; reception time. Receive The terminal station uses the received signal vector in step 437 based on an EM algorithm. In step 439, the transmission data are estimated using the estimated channel vector. Detect. In step 441, the terminal station encodes the channel vector estimated in step 437, and in step 443, the encoded forward channel estimation vector. Is fed back to the base station through the reverse channel. The channel estimation process will be described later in detail along with the signal model. However, it should be noted that the forward channel estimation method of the terminal station according to the present invention does not use a conventional rake receiver combining channel estimation and MRC (Maximum Ratio Combining) by symbol level signal processing after despreading. Undespread chip level forward signal Is to try forward channel estimation for. This is different from the forward channel estimation method of the existing terminal station. If the channel estimation is performed after despreading in the small state (L < 10) according to the conventional method, the interference signal composed of ISI or MAI becomes too large compared with the desired signal, and thus, the reliable channel estimation cannot be performed.

Hereinafter, the EM-based channel estimation algorithm proposed in the present invention (step 437 of FIG. 6) will be described in detail. On the other hand, the signal model considered for channel estimation in the present invention is as shown in FIG.

first, To be sent in the first transmission session Data sequence consisting of two pieces of data Data cycle Assume that it occurs while having (step 701). Unless otherwise confused, Superscript (i) indicating the i-th transmission / reception time will be omitted one or more times.

Thereafter, the transmission data chip period With, thus the diffusion rate Diffusion code Spread Signal Sequence (Step 702) Here, the case where L = 4 for convenience.

The weight vector of the spread signal is prepared at a previous time. Signal sequence multiplied by Is generated and supplied to the antenna array (step 703). The beamformed signal is transmitted to the space through the antenna array to pass through a wireless multiple fading channel. Here, the channel memory length is M chip section and each fading channel independent of each other Assuming that the signal is received by the terminal station Is the same as Equation 2 (step 704).

here,toIs the forward fading channel of the mth pathAnd array vector Is multiplied byIs 0 and the variance isAWGN (Additive White Gaussian Noise). And more than oneFor example,

On the other hand, the terminal station generates the k-th received signal vector as shown in Equation 3 below by combining the received signals with L spread values received during the k-th data period (step 705).

Here, Equations 2 and 3 can be summarized as in Equation 4 below.

Where the transmission signal matrix , Channel matrix , Noise vector T is defined as a Transpose operation.

As shown in FIG. 9 above. Data section with one channel memory If abnormal, a plurality of transmission data is received signal vector Is reflected in the number of data reflected (here, Denotes the integer part of a + 1). Where L = 4 and M = 7, Is calculated.

Then, one received signal vector Reflected inSignal vector ToI can think of it (see a).May be represented by a Markov chain as shown in Equation 5 below.

Then, the above Is the transmission data vector And spreading code As a function of It can be briefly expressed as follows. Further, the channel matrix H and the weight vector Multiply the channel vector by Called It is defined as The reason for this definition is that the terminal station can estimate the channel matrix H or the weight vector. Channel vector multiplied by these two values Because it is. According to the two definitions, Equation 4 may be summarized as Equation 6 below.

In general, the <Equation 5> and <Equation 6> Source EquationandIt is called Observation Equation, and these equations are collectively called Hidden Markov Model (HMM).

Using the HMM as a signal model of a high speed forward data transmission environment of interest, the following problems are set. That is, a received signal vector group consisting of K received signal vectorsAfter receiving (see 707), knowledge from the terminal station is known, i.e., the transmission data vector.Markov Properties of Liver (Equation 5) and Diffusion CodeUsing the desired Channel VectorIs estimated by the EM algorithm.

A channel estimation algorithm based on the HMM signal model and the EM algorithm will be described with reference to FIGS. 6 and 7 as follows.

First, when the terminal station starts operation, the channel vector is initialized in step 433, and in step 435, the K forward signal vector group from the base station is transmitted at every transmission / reception time (i-th). Receive In step 437, the EM-based channel vector is estimated using the forward signal vector group.

The EM based channel estimation process is illustrated in detail in FIG. 7. First, the terminal station determines whether the data sequence transmitted in step 601 is known in advance. If the transmitted data sequence is known in advance, step 603 is performed. Otherwise, step 605 is performed. First, the transmitted data sequenceConsider a case where is unknown to the terminal station. In this case, the received signal vector group in step 605.Estimated Channel Values for TransceiversProvided thatIs a specific valueA posteriori probabilityBy conditional probability as shown in Equation 7 Obtain

And in step 607Is the specific valueCan be pre-calculated when Transmit Signal MatrixAnd received signal vectorCross Correlation Vector with Auto Correlation Matrix R usingIs obtained through the following Equations 8 and 9

In step 607, the terminal station obtains the updated channel estimation value using Equation 8 and Equation 9 by Equation 10 below.

Where J is If is a binary data sequence Is given by Post-mortem probability The procedure for obtaining is detailed in (Ref. 3), so the explanation is omitted here.

(Ref. 3): G. Kaleh and R. Vallet, "Joint Parameter Estimation and Symbol Detection for Linear or Non-linear Unknown Dispersive Channels", IEEE Trans. on Communications, January 1994

Meanwhile, the transmission data sequence Is known to the terminal station in advance (i.e., when the transmission data sequence is a pilot sequence), Since it is obvious that the probability of having a known value is 1 and the probability of having a remaining value is 0, the following Equation 11 and Equation 12 are calculated using Equation 8 and Equation 9 in step 603, and the channel estimation value is used. Is updated in the same manner as in Equation (10).

After performing channel estimation as described above, the channel estimation value in step 439 of FIG. Detect the transmission data by using the data value that has the greatest post probability. Decide on In general, such a detection method is called Maximum A Posteriori (MAP) detection, and can be obtained as shown in Equation 13.

If the transmission data vector is determined, transmission data Above From the definition of It's easy to see that this is the first element of.

On the other hand, the channel estimation value in step 441 Is a coder with a predetermined encoding logic because it is difficult to transmit the data backward as it is. Encoding is performed as in Equation 14 below.

In operation 443, the encoded channel estimate value. Transmits backwards. After transmitting the channel estimation value backward, in step 445, the index i is increased to prepare to receive the next received signal vector, and the above processes are repeated.

In addition, the channel estimation value is fed back to the base station and used to form the transmission beam, and the coded channel estimation information fed back to the base station in FIG. The process of calculating the forward fading power from is shown. First, in step 501, the base station estimates forward channel estimation information. Is the encoder Decoder, counteracting Decoded by Equation 15 below.

And in step 503Weight vector And array vectorsForward channel estimation vector without componentsCalculate As pointed out above, the channel estimation vector is a fading channel matrix H and a weight vector.It should be noted that since the value is multiplied, the operation of step 503 must be performed. In operation 505, forward fading power is calculated using the calculated forward estimation vector. The forward fading power can be easily calculated from the fading coefficient. The forward fading power is used to obtain the transmission correlation matrix of Equation 1 as described above.

As described above, in a frequency division duplex-code division multiple access (FDD-CDMA) mobile communication system employing a transmission antenna array, a terminal station uses chip level signal processing instead of a channel estimation method using a conventional rake receiver method. In order to achieve high performance of fast forward data transmission, the base station estimates the forward channel estimation value estimated by the terminal station through the reverse channel and reflects it in the weight vector calculation for forward beamforming. The effects of increasing capacity of communication system, improving call quality, and saving transmission power of terminal station are produced.

Claims (20)

In the base station transmitter having an antenna array, A base station signal processor receiving feedback fading information estimated at a chip level of a forward signal from a terminal station and generating a weight vector for forming the transmission beam with an array vector estimated from the forward fading information and a received reverse signal; And a transmission beam former for controlling transmission beam formation of a transmission message with the weight vector. The method of claim 1, And the transmission beamformer copies the transmission message by the number of antennas (N), multiplies each by each element of the weight vector, and outputs N signals to the antenna array. The method of claim 1, wherein the base station signal processor, A forward channel decoder for decoding forward fading information for each path fed back from the terminal station; A forward fading power calculator for calculating forward fading power from the decoded forward fading information; An array vector calculator for calculating an array vector for each path from the received reverse signal; A transmission correlation matrix calculator for calculating a transmission correlation matrix with the forward fading power and the array vector; And a weight vector calculator that calculates a maximum eigenvector corresponding to the maximum eigenvalue of the transmission correlation matrix from the transmission correlation matrix calculator and determines the weight eigenvector. The method of claim 3, The transmission correlation matrix calculator is an array vector obtained from the array vector calculator. And forward fading power obtained from the forward fading power calculator. Transmit correlation matrix using Device for calculating the. The method of claim 3, The forward fading power calculator includes a forward fading coefficient extractor for extracting a forward fading coefficient from information from the forward channel decoder, and a power calculator for calculating forward fading power from the extracted forward fading coefficients. Device. The method of claim 5, In the forward fading coefficient extractor, the forward channel information fed back from the terminal station is complex information. In this case, the weight vector used by the base station in forming the corresponding transmission beam. And array vector estimates from the array vector calculator. Complex forward fading coefficient using Device for extracting the. The method of claim 5, The forward fading coefficient extractor, the forward fading information fed back from the terminal station is the size information In this case, the weight vector used by the base station in forming the corresponding transmission beam. Array vector estimates from the array vector calculator Forward fading size using Device for extracting the. A terminal station apparatus for receiving a signal transmitted with directionality from a base station, A forward fading estimator for estimating forward fading information for each path at a chip level of the forward signal from the base station; A forward channel encoder for collecting and encoding the obtained forward fading information for each path; And a reverse processor which multiplexes the encoded forward fading information with a transmission message and transmits the same to the base station. The method of claim 8, The forward fading estimator uses complex forward fading information when the base station forms a non-directional transmission beam. Or size forward fading information Device for estimating the. The method of claim 8, The forward fading estimator, the forward fading coefficient , Angle of arrival , Array vector , Weight vector , When the number of paths is M, complex forward fading information Or size forward fading information Device for estimating the. The method of claim 8, The forward channel estimator is characterized in that for estimating the forward fading information by the EM (Expectation Maximization) method. The method of claim 11, The EM method estimates the transmission data sequence and the channel vector according to a given rule with an HMM signal model consisting of a generation expression for the transmission data vector sequence having Markov properties and an observation expression for the reception signal vector passing through the radio channel. Device characterized in that. The method of claim 11, In the EM scheme, when the terminal station does not know the transmission data sequence in advance, it calculates a posterior probability of values that the transmission data may have, and correlates with the autocorrelation matrix R. And calculate the channel vector The device characterized in that the calculation. The method of claim 13, And the transmission data vector sequence is estimated as a vector sequence having the largest posterior probability. The method of claim 11, In the EM scheme, when the terminal station knows the transmission data sequence in advance, the autocorrelation matrix R and the cross-correlation vector are not required without calculating the post probability. And calculate the channel vector The device characterized in that the calculation. In the mobile communication system, The base station apparatus, Being fed back for the forward fading information estimated at the chip level of the forward signal from the terminal station, and the base station the signal processor to have the estimation of the array vector from the reverse signal to be received and the forward fading information generates a weight vector for the transmission beam forming , A transmission beam former for controlling transmission beam formation of a transmission message with the weight vector; The terminal station apparatus, A forward fading estimator for estimating forward fading information for each path at a chip level of the forward signal from the base station; A forward channel encoder for collecting and encoding the obtained forward fading information for each path; And a reverse processor which multiplexes the encoded forward fading information with a transmission message and transmits the same to the base station. A method of transmitting a forward signal by a base station apparatus of a mobile communication system having an antenna array and receiving forward fading information from a terminal station, Receiving, from the terminal station, forward fading information estimated at a chip level of a forward signal; Estimating an array vector from a reverse signal received from the terminal station; Generating a weight vector for forming the transmission beam with the forward fading information and the array vector; And controlling the transmission beam of the transmission message with the weight vector to transmit to the terminal station through the antenna array. In the terminal station transmission method, Estimating forward fading information for each path at the chip level of the forward signal; Collecting and encoding forward fading information for each estimated path; And multiplexing the encoded forward fading information with a transmission message to transmit to the base station. A communication method in a mobile communication system in which a base station transmits a directional signal to a specific terminal station through an antenna array, Estimating, by the terminal station, forward fading information for each path at the chip level of the forward signal from the base station, combining and encoding the estimated forward fading information for each path, and transmitting the information to the base station; Extracting, by the base station, the forward fading information from the backward signal from the terminal station, and generating a weight vector for forming a transmission beam using the extracted forward fading information; And controlling, by the base station, the transmission beam of the transmission message with the weight vector to transmit through the antenna array. A communication method in a mobile communication system in which a base station transmits a directional signal to a specific terminal station through an antenna array, Estimating, by the terminal station, forward fading information for each path at the chip level of the forward signal from the base station, combining and encoding the estimated forward fading information for each path, and transmitting the information to the base station; Extracting, by the base station, the forward fading information from the reverse signal from the terminal station, and calculating forward fading power using the extracted forward fading information; Estimating an array vector by the base station using a reverse signal from the terminal station; Generating, by the base station, a weight vector for forming a transmission beam with the forward fading power and the array vector; And controlling, by the base station, the transmission beam of the transmission message with the weight vector to transmit through the antenna array.