JP2009171564A - Radio receiving apparatus and radio receiving method - Google Patents

Abstract

A radio receiving apparatus and radio receiving method capable of realizing an excellent bit error rate characteristic with a simple configuration in a MIMO system that needs to perform single carrier transmission.
The receiver 20 side performs SC-FDE by the FDE unit 22 after performing GI removal by the GI units 21-1 to 21-n on the signals received by the antennas Rx1 to Rxn, and performs a primary estimation result thereof. To the ISI canceller 23, the signal components at the time k dispersed in the time direction are aggregated, and the secondary estimation result is obtained by the MLD unit 24. The two-time estimation result is sequentially fed back to the ISI canceller 23 for each time and updated. After this process is repeated I times for each block of the FDE unit 22, the estimated signal by the MLD 24 is demodulated to obtain data.
[Selection] Figure 1

Description

  The present invention relates to a data transmission system in a digital wireless communication system. In particular, a radio receiving apparatus and a radio receiving method for realizing excellent bit error rate characteristics in a multiple-input multi-output (hereinafter referred to as MIMO) system in a frequency selective channel are provided. Is.

  As a first prior art, a maximum likelihood sequence estimation method using the Viterbi algorithm is raised. This scheme compares the received sequence with all sequences that may have been transmitted. As an example, to decode a binary sequence of length L bits, the likelihood of 2L different code sequences that may have been transmitted (the conditional probability that X was transmitted in the received sequence Y) p (Y / X)) are compared, and the most probable code sequence X that maximizes this likelihood is selected.

As a second conventional technique, there is an SC-FDE (Single-Carrier Frequency Domain Equalization) system. In this method, received data is converted into a frequency domain by Fast Fourier Transform (hereinafter referred to as FFT) on the receiver side, and frequency equalization and signal using a method called Nulling that multiplies a MMSE standard weight matrix. Separate at the same time. After this process, the data is returned to the time domain by Inverse Fast Fourier Transform (hereinafter referred to as IFFT), and the data is demodulated as it is (see Non-Patent Document 1).
Utsunomiya, Takahashi, Iwanami, Okamoto, “A Comparative Study on LDPC Coded MMO SC FDE System”, IEICE Society Conference, September 2007.

  The Viterbi algorithm increases the number of states required for equalization as the number of multipath delayed wave components to be considered increases exponentially. Therefore, the complexity of the system becomes a problem from the viewpoint of increasing the amount of computation and memory. , Difficult to use.

  The conventional SC-FDE method cannot completely eliminate the influence of time dispersion ISI, and equalization and signal separation are insufficient. However, in the SC-FDE transmission system, a method for performing maximum likelihood determination (hereinafter referred to as MLD) after frequency domain equalization has not been studied.

  In view of such circumstances, the present invention provides a radio reception apparatus and radio reception method capable of realizing excellent bit error rate characteristics with a simple configuration in a MIMO system that needs to perform single carrier transmission. Is.

The present invention is a receiving apparatus for receiving a wireless communication signal by MIMO single carrier transmission,
A frequency domain equalization unit that obtains a provisional decision value by performing frequency domain equalization of the received signal; an intersymbol interference removal unit that removes intersymbol interference from the reception signal using the provisional decision value; and the intersymbol interference. And a maximum likelihood determination unit that performs MIMO signal separation by performing maximum likelihood determination on the received signal from which the signal is removed.

  Here, the intersymbol interference removing unit removes the intersymbol interference using a temporary judgment value output from the maximum likelihood judging unit.

The present invention is also a wireless reception method for receiving a signal of wireless communication by MIMO single carrier transmission,
A frequency domain equalization procedure for performing a frequency domain equalization of a received signal to obtain a provisional decision value; an intersymbol interference removal procedure for removing intersymbol interference from the reception signal using the provisional decision value; and the intersymbol interference. And a maximum likelihood determination procedure for performing MIMO signal separation by performing maximum likelihood determination on the received signal from which the signal is removed.

  Here, the intersymbol interference cancellation procedure and the maximum likelihood determination procedure are repeatedly performed.

  As described above, in the MIMO single carrier transmission system, when the channel channel matrix is known by the receiver, the present invention performs the primary determination of the received signal by single carrier frequency equalization, and further performs intersymbol interference removal. By performing secondary determination by maximum likelihood determination, it is possible to realize spatial demultiplexing of signals in a MIMO communication channel and further equalization of intersymbol interference more accurately and with a simple system configuration than the prior art. .

  That is, in the present invention, the determination result by the single carrier frequency equalization is output to the intersymbol interference removing unit, the time-dispersive intersymbol interference is removed from the received signal of each antenna, and the processed signal is the maximum likelihood determining unit. Thus, interference between channels of each transmission antenna can be removed and spatial demultiplexing can be realized.

  Also, in the present invention, in order to obtain a gain by repeated equalization, a large iterative demodulation gain can be realized by repeatedly feeding back the determination result by the maximum likelihood determining unit to the intersymbol interference removing unit.

First, the outline of the present invention will be described with reference to FIG.
FIG. 1 is a block diagram showing a transmitter and a receiver of a MIMO single carrier transmission scheme. The transmitter 10 includes a PSK or QAM modulation unit 11 that modulates input data into PSK or QAM, an S / P (serial / parallel) conversion unit 12, a GI (Guard Interval) addition unit 13-1 to 13-n, and an antenna Tx1. ~ Txn. The receiver 20 includes antennas Rx1 to Rxn, GI (Guard Interval) removing units 21-1 to 21-n, an FDE (Frequency Domain Equalization) unit 22, an ISI (Inter-Symbol Interference). A canceller (also referred to as an intersymbol interference removing unit) 23, an MLD (maximum likelihood determination) unit 24, and a PSK or QAM demodulating unit 25 are provided. Further, the FDE unit 22 includes S / P conversion units 31-1 to 31-n, FFT (fast Fourier transform) units 32-1 to 32-n, Nulling (signal separation) unit 33, and IFFT (inverse fast Fourier transform). Sections 34-1 to 34-n and P / S (parallel / serial) conversion sections 35-1 to 35-n.

  On the transmitter 10 side, the S / P unit 12 divides the data modulated by the PSK or QAM modulation unit 11 into blocks composed of M symbols, and the GI addition units 13-1 to 13-n perform GI ( Guard Interval) is added and transmitted from each antenna Tx1 to Txn.

  On the receiver 20 side, after GI removal by the GI removal units 21-1 to 21-n is performed on the signals received by the antennas Rx1 to Rxn, the FDE unit 22 performs SC-FDE, and the primary estimation result (formula 1 ) To the ISI canceller 23, the signal components at time k dispersed in the time direction are aggregated, and the secondary estimation result (Formula 2) is obtained by the MLD unit 24. Formula 2 is fed back to the ISI canceller 23 sequentially for each time, and Formula 3 is updated. When this process is repeated I times for each block of the FDE unit 22 and the estimated signal by the MLD 24 is expressed by Equation 4, this signal is finally demodulated to obtain data.

The signal y (k) obtained after GI removal on the receiver 20 side is expressed by the following equation (Equation 5).

  In Equation 5, y (k) is the received signal vector of the k-th symbol, x (k) is the transmitted signal vector, w (k) is the noise vector, and the multipath delay wave number is L. When the number of transmission antennas nT and the number of reception antennas nR, the channel matrix Hi is represented by a matrix of nT × nR as in the following formula (Formula 6), and is a block unit of FFT processing by the FFT units 32-1 to 32-n. Takes the elements of a time-invariant complex Gaussian variable.

  Next, after converting the received signal y (k) independent for each multipath (l = 0, 1,..., L) into the frequency domain by p-point FFT, the nulling unit 33 obtains the following formula (formula 7 Frequency domain equalization and signal separation (Nulling) are simultaneously performed using an MMSE weight matrix w (k) such as

  In Equation 7, G (j) is a MIMO channel matrix at the j-th point on the frequency axis, σ2 is noise variance, and In is an nT-th unit matrix. Thereafter, the IFFT units 34-1 to 34-n return to the time domain to obtain a primary estimation result (Formula 1).

  Using the primary estimation result (Formula 1), the ISI canceller 23 extracts signal components dispersed in the time direction due to the influence of the multipath delay wave. The output of the ISI canceller 23 is expressed by the following formula (Formula 8).

  In Expression 8, the output result (Expression 9) is an nR (L + 1) × 1 column vector.

  Thus, by subtracting the signal represented by Equation 10 estimated by SC-FDE from the received signal represented by Equation 9 and the replica made up of the communication channel matrix Hi, the time dispersive ISI is removed, Only the signal component at the desired time k can be extracted from each delayed wave component.

  The ISI canceller 23 operates in block units M of the FDE unit 22. Therefore, in Equation 8, values can be obtained for Equation 11 regarding the processing of both ends (k = ML + 1,..., M−1, M) and (k = 1,..., L−1) of the block. There is a part that cannot be done.

  Therefore, dummy data is generated by copying the symbols at both ends of the block by the number L of delay waves as shown in FIG.

  The MLD unit 24 performs determination by MLD using the output result (represented by Expression 12) from the ISI canceller 23 in the previous stage.

When an MLD signal point replica is expressed by Equation 13,

  When the number of antennas is nT × nT and the multi-level modulation number is m, there are nTm Equations (13). In MLD, the following formula (Formula 14) is used.

  Formula 13 that minimizes Formula 14 is output as the discrimination result of the MLD unit 24. At this time, Formula 13 is fed back sequentially to the ISI canceller 23 at the time of determination of the symbol at the next time.

  Specifically, when k = k + 1 in Equation 8, the portion of Equation 3 is replaced with Equation 2. In this way, the FDE unit 22 sequentially corrects the SC-FDE estimation results to improve the operation of the ISI canceller 23. After the calculation of the ISI canceller 23 and the detection by the MLD unit 24 are repeated I times for each block of the FDE unit 22, the PSK or QAM demodulation unit 25 outputs the output result (the output result represented by the equation 13) by the MLD unit 24. Demodulate and get data.

  Hereinafter, although the effect of the present invention is explained concretely based on an example, the present invention is not limited to these examples from the first.

  The communication channel is a MIMO system, and each transmission / reception antenna is a 1 dB exponentially damped quasi-static Rayleigh fading channel with 16 multipaths. A model is assumed. The delay profile of this communication path is shown in FIG.

  One block of SC-FDE is M = 64 symbols, Guard Interval length is 16 symbols, and the number of FFT points is p = 64. It is also assumed that channel estimation of the communication channel is complete on the receiving side.

  FIG. 4 shows an error rate (Bit Error Rate, BER) characteristic by computer simulation when BPSK is used for modulation and FIG. 5 is used when QPSK is used for modulation. In both cases, the number of antennas is a 4 × 4 MIMO system, and the characteristics in the case of the AWGN communication channel of each system and the characteristics in the case of demodulating the output of the SC-FDE as they are for comparison are shown. When the MLD feedback is repeated five times, the gain of about 9 dB in FIG. 4 and about 9 dB in FIG. 5 is obtained at BER = 10−4 when the SC-FDE output is demodulated as it is. It has been.

  FIG. 6 shows BER characteristics by computer simulation in the case of a MIMO system with 16QAM modulation and 2 × 2 antennas. As comparison targets, the characteristics in the case of the AWGN communication path and the characteristics in the case of demodulating the output of the SC-FDE as it is are used. When MLD feedback is repeated five times, a gain of about 5 dB is obtained at BER = 10-4, compared to the case where the SC-FDE output is demodulated as it is.

  The present invention relates to a data transmission system in a digital wireless communication system. In particular, high-quality frequency-selective channel compensation and spatial multiplexing signal separation can be considered, and it is used as a method to achieve excellent bit error rate characteristics with a simple configuration in MIMO systems that require single-carrier transmission. there is a possibility.

It is a block diagram which shows the system model of a transmitter / receiver. It is a figure which shows the processing method of the both ends of the block of SC-FDE in an ISI canceller. It is a model figure which shows the outline of a multipath channel model. It is the figure which showed the comparison result of the BER characteristic by computer simulation at the time of using BPSK for the modulation | alteration in a 4x4 MIMO system. It is the figure which showed the comparison result of the BER characteristic by computer simulation at the time of using QPSK for the modulation | alteration in a 4x4 MIMO system. It is the figure which showed the comparison result of the BER characteristic by computer simulation at the time of using 16QAM for the modulation | alteration in a 2 * 2 MIMO system.

Explanation of symbols

10 Transmitter 11 PSK or QAM Modulation Unit 12 S / P (Serial / Parallel) Conversion Units 13-1 to 13-n GI (Guard Interval) Addition Units Tx1 to Txn Antenna 20 Receivers 21-1 to 21-n GI ( Guard Interval) removal unit 22 FDE unit 23 ISI canceller 24 MLD (maximum likelihood determination) unit 25 PSK or QAM demodulation unit 31-1 to 31-n S / P (serial / parallel) conversion unit 32-1 to 32-n FFT (Fast Fourier Transform) Unit 33 Nulling (Signal Separation) Unit 34-1 to 34-n IFFT (Inverse Fast Fourier Transform) Unit 35-1 to 35-n P / S (Parallel / Serial) Conversion Unit

Claims (4)

A receiving device for receiving a wireless communication signal by MIMO single carrier transmission,
A frequency domain equalization unit that obtains a provisional determination value by performing frequency domain equalization of the received signal;
An intersymbol interference removing unit that removes intersymbol interference from the received signal using the temporary determination value;
A maximum likelihood determination unit that performs MIMO signal separation by performing maximum likelihood determination on the received signal from which the intersymbol interference is removed;
A radio receiving apparatus comprising: The intersymbol interference removing unit
The radio reception apparatus according to claim 3, wherein intersymbol interference is removed using a temporary determination value output by the maximum likelihood determination unit. A radio reception method for receiving a signal of radio communication by MIMO single carrier transmission,
A frequency domain equalization procedure for obtaining a provisional determination value by performing frequency domain equalization of a received signal;
An intersymbol interference removal procedure for removing intersymbol interference from the received signal using the temporary determination value;
Maximum likelihood determination procedure for performing MIMO signal separation by performing maximum likelihood determination on the received signal from which the intersymbol interference is removed,
A wireless reception method characterized by the above. The radio reception method according to claim 1, wherein the intersymbol interference cancellation procedure and the maximum likelihood determination procedure are repeated.

Images