JP2010057071A - Demodulator and demodulation method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a demodulator supporting, in a single architecture, reception and demodulation in all possible transmission-reception modes. <P>SOLUTION: On the basis of a channel estimation and reception signals from reception antennas, a channel matrix corresponding to a communication mode is calculated and on the basis of the reception signals, a reception vector corresponding to the communication mode is calculated. Furthermore, a unitary matrix Q and an upper triangular matrix R corresponding to the communication mode are derived by QR decomposition of the channel matrix, the calculated reception vector is multiplied by the Hermitian matrix of the unitary matrix, and a reception symbol vector is calculated. Moreover, on the basis of the received symbol vector and the upper triangular matrix R, soft bits are calculated by computing the Euclidean distance. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a demodulator and a demodulation method for demodulating data transmitted by a plurality of communication methods.

Receiver devices in the latest state-of-the-art systems for mobile communications, such as LTE (Long Term Evolution) and advanced LTE, which are high-speed data communications specifications for mobile communications, were transmitted using various different communication methods. It is desirable to require the ability to demodulate data. Therefore, it is highly desirable to have a demodulation method that is substantially similar across all transmission methods. In addition, the technique relevant to this invention is disclosed by patent document 1 and patent document 2. FIG.
US Patent Application Publication No. 2008/0069261 US Patent Application Publication No. 2008/0056396

Here, the current LTE system is:
1.2 Transmit-2 Receive Antenna (2x2) Multiple Input / Multiple Output (MIMO) System 2.1x1 Single Input / Single Output (SISO) System 3.2x1 Receive Diversity (Rx Diversity)
4.1 × 2 or 1 × 4 spatial frequency block code (SFBC) system 5.2 × 2 or 2 × 4 spatial frequency block code (SFBC) system 6.2 × 2 or 2 × 4 spatial multiplexing ( There is a need for a receiver that can demodulate data transmitted in one to six communication modes of a (MIMO) system with a single receiver device.

Among the communication modes 1 to 6 listed above, the system that requires the most complex demodulator is a MIMO system that requires a demodulator that simultaneously detects two streams of transmission data. One method for demodulating data transmitted in this communication mode is to use QR decomposition and maximum likelihood detector (MLD). A block diagram of the current state of the art QR-MLD demodulator is shown in FIG. The configuration of the demodulator shown in this figure can be summarized as several main functional configurations. In other words, the demodulator
(1) Calculation of channel matrix and received vector A process of taking channel estimation and received symbols from each receiving antenna and reconfiguring them so that appropriate transmitted symbols can be demodulated.
(2) QR decomposition Processing for performing QR decomposition on the channel matrix provided by the output of (1).
(3) Received symbol processing Processing to multiply the calculated received symbol vector by the Hermitian matrix Q obtained by QR decomposition processing.
(4) Calculation of soft bits A process of calculating soft bits for each transmission symbol using the Euclidean distance metric.
(5) Controller A process that controls how the functional block should operate, including determining how the channel matrix and received vector should be calculated.
It has the main functional configuration.

  The state-of-the-art QR-MLD in a MIMO system is well known, but its use of functional architecture blocks is limited to 2 × 2 MIMO (spatial multiplexing), and SIDO, SFBC (2Tx-1Rx, 4Tx-1Rx, 2Tx -4Rx, and 4Tx-2Rx), and other cases such as Rx diversity.

  Therefore, the present invention aims to provide a demodulator and demodulation method that can support reception and demodulation of all possible transmission-reception modes by a single architecture.

  To achieve the above object, the present invention provides channel matrix calculation means for calculating a channel matrix corresponding to a communication mode based on channel estimation and a received signal from each receiving antenna, and based on the received signal, A reception vector calculation means for calculating a reception vector according to the communication mode; a QR decomposition means for deriving a unitary matrix Q and an upper triangular matrix R according to the communication mode by QR decomposition of the channel matrix; and the calculated reception The received symbol processing means for calculating the received symbol vector by multiplying the vector by the Hermitian matrix of the unitary matrix Q, and the Euclidean distance calculation based on the received symbol vector and the upper triangular matrix R And a soft bit calculating means for calculating a bit.

  According to the present invention, in the demodulator described above, the soft bit calculation means calculates the first minimum distance from a constellation symbol having 1 for the i-th bit position of the soft bit. Second minimum distance calculating means for calculating a second minimum distance from a constellation symbol having 0 for the i-th bit position of the soft bit, and the first minimum distance and the second minimum distance And a minimum distance difference calculating means for calculating the difference.

  According to the present invention, the demodulator includes a controller for setting the communication mode.

  Further, the present invention is a demodulation method in a demodulator, wherein the channel matrix calculation means of the demodulator calculates a channel matrix corresponding to a communication mode based on channel estimation and a received signal from each receiving antenna, The reception vector calculation means of the demodulator calculates a reception vector according to the communication mode based on the received signal, and the QR decomposition means of the demodulator includes a unitary matrix Q and an upper triangular matrix according to the communication mode. R is derived by QR decomposition of the channel matrix, and the received symbol processing means of the demodulator multiplies the calculated received vector by the Hermitian matrix of the unitary matrix Q to calculate a received symbol vector, Based on the received symbol vector and the upper triangular matrix R, the soft bit calculation means of the demodulator performs Euclidean distance The calculation is a demodulation method characterized by calculating the soft bits.

  According to the present invention, in the demodulation method described above, the soft bit calculation unit obtains a first minimum distance from a constellation symbol having 1 for the i-th bit position of the soft bit in the first minimum distance calculation unit. And a second minimum distance calculating unit calculates a second minimum distance from a constellation symbol having 0 for the i-th bit position of the soft bit, and a minimum distance difference calculating unit calculates the first minimum distance. The difference between the distance and the second minimum distance is calculated.

  According to the present invention, in the demodulation method described above, the controller of the demodulator sets the communication mode.

According to the present invention, according to the above-described demodulator processing, it is possible to receive and demodulate a physical channel transmitted in different communication modes, such as a SISO system, a SIMO system, an SFBC system, and a MIMO system. A consistent method using QR decomposition with detection can be provided. And the advantage of this consistent demodulation method is
(1) The chip size can be reduced by reusing functions and components.
(2) All communication modes can reduce overall processing of other functions such as CQI and threshold calculation by receiving the same processing chain.
(3) High-speed demodulation switching can be performed when the transmission mode is dynamically changed as frequently as for subframes.
Also, the above-described demodulator processing results in equal response (in frequency or time) of adjacent channels, thereby potentially reducing system error rate performance over conventional SFBC system decoding. It is possible to provide a method of demodulating a spatial frequency block code (SFBC) that is not supposed to be performed.
According to the processing of the above-described demodulator, the QR-MLD demodulator according to the communication mode of the existing 2 × 2 MIMO system is reused, and the SISO system, SIMO system, SFBC system, MIMO system (Tx-Rx diversity), etc. It is possible to construct a complex system from the viewpoint of the overall functional architecture.

Hereinafter, a demodulator according to an embodiment of the present invention will be described with reference to the drawings.
FIG. 1 is a block diagram showing a configuration of a demodulator according to the embodiment. In this figure, reference numeral 1 denotes a demodulator. The demodulator 1 according to the embodiment of the present invention includes a controller 11, a channel matrix / reception vector calculation unit 12, a QR decomposition processing unit 13, a received symbol processing unit 14, and a soft bit calculation unit 15. .

Next, details of the demodulator of this embodiment will be described.
The demodulator 1 according to the present embodiment improves the performance of function blocks used in 2 × 2 MIMO detection in QR-MLD demodulation, and selects the configuration of an appropriate communication mode for function processing. Functional configuration is introduced. Each processing unit of the demodulator 1 operates according to the following processing steps.

<Step S1> Calculation of Channel Matrix and Reception Vector First, the channel matrix / reception vector calculation unit 12 calculates a channel matrix and a reception vector suitable for calculation of QR decomposition processing for each communication mode. Here, the f-th frequency tone in the channel between the transmitting antenna n and the receiving antenna m is represented by hm _{, n} (f). The calculation of the channel matrix and the reception vector in each communication mode is performed by the following equations.

  (Communication mode A) In 2 × 2 MIMO for demodulating symbols transmitted from the first antenna, the channel matrix H is calculated by the following equation (1), and the reception vector is calculated by the following equation (2).

  (Communication mode B) In 2 × 2 MIMO for demodulating a symbol transmitted from the second antenna, the channel matrix H is calculated by the following equation (3), and the reception vector is calculated by the following equation (4).

  (Communication mode C) In the 1 × 1 SISO system, the channel matrix H is calculated by the following equation (5), and the reception vector is calculated by the following equation (6).

  (Communication mode D) In the 2 × 1 Rx diversity system, the channel matrix H is calculated by the following equation (7), and the reception vector is calculated by the following equation (8).

  (Communication mode E) In a 1 × 2 SFBC system that demodulates symbols transmitted on the first frequency tone, the channel matrix H is calculated by the following equation (9), and the reception vector is calculated by the following equation (10).

  (Communication mode F) In a 1 × 2 SFBC system that demodulates symbols transmitted with the second frequency tone, the calculation of the channel matrix H is performed by the following equation (11), and the reception vector is calculated by the following equation (12).

  (Communication mode G) In a 2 × 2 SFBC system that demodulates symbols transmitted on the first frequency tone, calculation of the channel matrix H is performed by the following equation (13), and reception vector is calculated by the following equation (14).

  (Communication mode H) In a 2 × 2 SFBC system that demodulates symbols transmitted on the second frequency tone, the channel matrix H is calculated by the following equation (15), and the reception vector is calculated by the following equation (16).

<Step S2> QR Decomposition Next, in step S2, the QR decomposition processing unit converts the unitary matrix Q and the upper triangular matrix R corresponding to a specific communication mode into the QR of the channel matrix output from the channel matrix / reception vector calculation unit 12. Derived by performing decomposition. Mathematically, it is expressed by the following equation (17).

  Here, as in the case of Rx diversity, when the channel output by the channel matrix / reception vector calculation unit 12 is a vector, Q is a vector and R is a scalar value. Further, when the channel output by the channel matrix / reception vector calculation unit 12 is a scalar as in the case of the SISO system, Q is 1 and R is a channel coefficient.

<Step S3> Received Symbol Processing Next, in step S3, the received symbol processing unit 14 multiplies the calculated received vector by the Hermitian matrix of the unitary matrix Q to calculate a received symbol vector. This process can be expressed by equation (18).

  Here, when Q is a vector, as in the case of Rx diversity, z is a scalar value. Also, as in the SISO system, when Q is a scalar unit, z is a scalar value, but this step is self-evident.

<Step S4> Calculation of Soft Bit Next, in step S4, the soft bit calculation unit 15 uses the processed reception symbol processing result and the R matrix (vector or scalar) output by the QR decomposition processing unit 13. Calculate soft bits. That is, this is done by calculating the Euclidean distance as shown in equation (19).

Here, in Expression (19), c _j is a symbol derived from a set of constellations (for example, QPSK, 16QAM, and 64QAM). In the case of the SISO system (1Tx-1Rx) and the case of the SIMO system (1Tx-2Rx), r and z are scalar values, whereby the Euclidean distance is calculated as in Expression (20).

Further, in the SFBC system (2Tx-1Rx, 4Tx-1Rx, 2Tx-2Rx, or 4Tx-2Rx) and the MIMO system, R is a matrix, z is a vector, and the Euclidean distance is expressed by the equation (21). Is calculated as follows. Incidentally, the last row (r _{2, 2)} in the last column of the last row (z ₂₎ and the matrix R of the vector z is used to calculate the Euclidean distance.

  In addition, the soft bit calculation unit 15 performs the additional distance calculation represented by Expression (22) for each communication mode.

In Equation 22, x ^{^} _j is an estimate of a symbol related to another antenna (in the case of MIMO) or frequency (in the case of SFBC). This estimation can be calculated by the following equation (23).

The communication mode of the SFBC system is used, and if the soft bits of the "second" transmission symbol is calculated, the value of C _j, should be conjugated in the estimation of distance metrics and x ^{^} _j.

Also, if the communication mode of the SFBC system is used and the soft bits of the “first” transmission symbol are being calculated, the second distance metric should include the conjugate of x ^{^} _j .

Next, the soft bit calculation unit 15 derives the last distance.
Here, in the communication mode of the SISO system and the communication mode of Rx diversity, the last distance is equal to the calculated first distance metric.

  In the communication mode of the SFBC system and the communication mode of the MIMO system, the last distance metric is the sum of the two distance metrics calculated above.

And the soft bit calculation part 15 finally determines a soft bit by the following formula | equation. That is, first, the minimum distance is calculated from a constellation symbol having 1 at the i-th bit position (denoted as B ¹ _i ).

Then, the minimum distance is calculated from the constellation symbol having 0 at the i-th bit position (denoted as B ⁰ _i ).

  The soft bit calculation unit 15 calculates the difference between the two minimum distances.

In addition, the controller 11 implements the setting corresponding to the communication mode suitable for the input signal processing for each processing unit that performs Steps S1 to S4. Its main functions are as follows.
(1) A process for performing settings that constitute calculation of a channel matrix and a reception vector for an appropriate communication mode.
(2) A process for controlling which distance metric should be calculated according to the operating communication mode.
(3) A process that appropriately specifies whether an appropriate value of conjugate should be used in the SFBC mode.

Note that the above algorithm for the SFBC system does not assume that the response (for frequency) of adjacent channels is equal as generally assumed. This in some cases improves the performance of highly frequency selective channels and can thus outperform conventional SFBC demodulation methods.
The above algorithm can also be extended to space time block codes (STBC).
Further, the configuration of the demodulator shown in FIG. 3 is not limited to the maximum number of two transmission antennas and two reception antennas, and can be extended to a larger number of transmission antennas and reception antennas.

Although the embodiments of the present invention have been described above, according to the above-described demodulator processing, physical channels transmitted in different communication modes such as a SISO system, a SIMO system, an SFBC system, and a MIMO system are received and demodulated. In order to do so, a consistent method of using QR decomposition with maximum likelihood detection can be provided. And the advantage of this consistent demodulation method is
(1) The chip size can be reduced by reusing functions and components.
(2) All communication modes can reduce overall processing of other functions such as CQI and threshold calculation by receiving the same processing chain.
(3) High-speed demodulation switching can be performed when the transmission mode is dynamically changed as frequently as for subframes.
Also, the above-described demodulator processing results in equal response (in frequency or time) of adjacent channels, thereby potentially reducing system error rate performance over conventional SFBC system decoding. It is possible to provide a method of demodulating a spatial frequency block code (SFBC) that is not supposed to be performed.
According to the processing of the above-described demodulator, the QR-MLD demodulator according to the communication mode of the existing 2 × 2 MIMO system is reused, and the SISO system, SIMO system, SFBC system, MIMO system (Tx-Rx diversity), etc. It is possible to construct a complex system from the viewpoint of the overall functional architecture.

It is a block diagram which shows the structure of a demodulator. It is a block diagram of a QR-MLD demodulator.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 ... Demodulator 11 ... Controller 12 ... Channel matrix and received vector calculation part 13 ... QR decomposition processing part 14 ... Received symbol processing part 15 ... Soft bit calculation part

Claims (6)

Channel matrix calculating means for calculating a channel matrix according to the communication mode based on channel estimation and a received signal from each receiving antenna;
Received vector calculation means for calculating a received vector according to the communication mode based on the received signal;
QR decomposition means for deriving a unitary matrix Q and an upper triangular matrix R according to the communication mode by QR decomposition of the channel matrix;
Received symbol processing means for calculating the received symbol vector by multiplying the calculated received vector by the Hermitian matrix of the unitary matrix Q;
Soft bit calculating means for calculating a soft bit by calculation of Euclidean distance based on the received symbol vector and the upper triangular matrix R;
A demodulator. The soft bit calculation means includes
First minimum distance calculating means for calculating a first minimum distance from a constellation symbol having 1 for the i-th bit position of the soft bit;
Second minimum distance calculating means for calculating a second minimum distance from a constellation symbol having 0 for the i-th bit position of the soft bit;
The demodulator according to claim 1, further comprising minimum distance difference calculation means for calculating a difference between the first minimum distance and the second minimum distance.   The demodulator according to claim 1, further comprising a controller that sets the communication mode. A demodulation method in a demodulator,
The channel matrix calculation means of the demodulator calculates the channel matrix according to the communication mode based on the channel estimation and the received signal from each receiving antenna,
The reception vector calculation means of the demodulator calculates a reception vector corresponding to the communication mode based on the reception signal,
QR decomposition means of the demodulator derives a unitary matrix Q and an upper triangular matrix R according to the communication mode by QR decomposition of the channel matrix,
The received symbol processing means of the demodulator multiplies the calculated received vector by the Hermitian matrix of the unitary matrix Q to calculate a received symbol vector,
The demodulating method characterized in that the soft bit calculating means of the demodulator calculates a soft bit by calculating the Euclidean distance based on the received symbol vector and the upper triangular matrix R. The soft bit calculation means includes
In a first minimum distance calculating means, a first minimum distance is calculated from a constellation symbol having 1 for the i-th bit position of the soft bit,
A second minimum distance calculating means for calculating a second minimum distance from a constellation symbol having 0 for the i-th bit position of the soft bit;
The demodulation method according to claim 4, wherein the minimum distance difference calculation means calculates a difference between the first minimum distance and the second minimum distance. The demodulation method according to claim 4 or 5, wherein a controller of the demodulator sets the communication mode.

Images