JP5263958B2 - Signal processing device - Google Patents

Description

  The present invention relates to a signal processing device, a signal processing method, and a program.

  With the widespread use of mobile terminals such as mobile phones, research and development of next-generation communication methods that enable high-speed communication of large volumes of data even in wireless communication and the use of multimedia data such as videos and voices on mobile terminals are thriving. Has been done.

  In 3GPP (3rd Generation Partnership Project), LTE (Long Term Evolution) based on the premise of a transmission rate of a maximum of 100 Mbps in the downlink is studied as a next generation communication method. A communication system using OFDM (Orthogonal Frequency Division Multiplexing) has attracted attention as a downlink radio transmission system in LTE.

  In the OFDM scheme, a band to be used is divided into a plurality of subcarriers, and each data symbol is assigned to each subcarrier for transmission. Since the subcarriers are arranged so as to be orthogonal to each other on the frequency axis, the frequency utilization efficiency is excellent. In addition, since each subcarrier has a narrow band, the influence of multipath interference can be suppressed, and high-speed and large-capacity communication is possible.

  In wireless communication, signal distortion due to multipath fading or the like occurs in a wireless communication path (channel). Therefore, when receiving an OFDM signal, an estimated value of channel characteristics (hereinafter referred to as a channel estimated value) is estimated, and distortion of the signal received on the channel is compensated based on the channel estimated value. If the accuracy of the channel estimation value is low, the signal distortion is not properly corrected, so that the demodulation accuracy of the received signal is lowered.

  FIG. 7 is a block diagram illustrating a configuration example of the channel estimation unit 101 that performs channel estimation in an LTE receiver (not illustrated) (see Non-Patent Document 1 and Patent Document 1). FIG. 8 is a flowchart showing the flow of channel estimation processing in the channel estimation unit 101.

  With reference to FIG.7 and FIG.8, the operation | movement of each part of the channel estimation part 101 and the channel estimation process in the channel estimation part 101 are demonstrated.

  The channel estimation unit 111 cancels the reference signal pattern multiplexed with the data symbol from the received signal of each subcarrier supplied to the channel estimation unit 101, and the subcarrier to which the reference signal is mapped (hereinafter referred to as the first carrier). A temporary channel estimation value (hereinafter referred to as a first subcarrier channel estimation value) is obtained (referred to as a first subcarrier channel estimation value hereinafter) (step S101).

  9 and 10 are references in MIMO (Multi-Input Multi-Output) transmission in which different signals are transmitted in parallel on a transmission path formed by a plurality of inputs and a plurality of outputs (that is, a plurality of transmission antennas and reception antennas). It is a figure which shows the example of the mapping of a signal. FIG. 9 shows an example of mapping of reference signals of the transmission antenna T0 and the transmission antenna T1, and FIG. 10 shows an example of mapping of reference signals of the transmission antenna T2 and the transmission antenna T3. In the figure, a reference signal is mapped to a black subcarrier.

  Thus, the reference signal is regularly distributed on the time axis and the frequency axis of the OFDM system. That is, the reference signal is transmitted at a predetermined time interval by using a subcarrier having a predetermined frequency.

  In the transmission antenna T0 and the transmission antenna T1, as shown in FIG. 9, for example, every two slots, two reference signals are mapped to subcarriers of a predetermined frequency, and subcarriers are assigned between the time directions. 6 data symbols are inserted. For example, in FIG. 9, the reference signals S1 and S2, the reference signals S3 and S4, and the reference signals S5 and S6 are mapped to subcarriers of a predetermined frequency, respectively. Is inserted.

  In the transmission antenna T2 and the transmission antenna T3, as shown in FIG. 10, one reference signal is mapped to a subcarrier of a predetermined frequency for every two slots, and between subcarriers arranged in the time direction at the same frequency. Are inserted 13 data symbols.

  The channel estimation unit 111 obtains a channel estimation value of the first subcarrier to which such a reference signal is mapped.

  Based on the channel estimation value of the first subcarrier to which the reference signal obtained by the channel estimation unit 111 is mapped, the channel estimation unit 112 is a subcarrier (hereinafter referred to as a subcarrier at a predetermined position between the first subcarriers). , A temporary channel estimation value (hereinafter referred to as a second subcarrier channel estimation value). In this example, a temporary channel estimation value of a subcarrier located at a hatched position in FIGS. 9 and 10 is obtained.

  Specifically, the average value of the difference between the channel estimation values of the first subcarriers adjacent in the time direction is compared with the average value of the difference between the channel estimation values of the first subcarriers adjacent in the frequency direction. (Step S102).

  If it is determined that the difference in channel estimation value of the first subcarrier in the time direction is smaller, the channel estimation value of the first subcarrier located next to the second subcarrier in the time direction is For example, the channel estimation value of the second subcarrier is obtained by linear interpolation (step S103). On the other hand, if it is determined that the difference in channel estimation value of the first subcarrier in the frequency direction is smaller, the channel estimation value of the first subcarrier located adjacent to the second subcarrier in the frequency direction is For example, the channel estimation value of the second subcarrier is obtained by linear interpolation (step S104).

  The channel estimation value of the first subcarrier (step S101) and the channel estimation value of the second subcarrier (step S103 or step S104) are input to the virtual waveform adding unit 113. The virtual waveform adding unit 113 adds a waveform such that the number of samples is a power of two so that a fast Fourier transform (FFT) process can be used for channel estimation value calculation (step S105).

  Specifically, the virtual waveform adding unit 113 adds a 0-value data string after the channel estimation value sample of the first subcarrier and the channel estimation value sample of the second subcarrier, and the total data length is 2. Try to be a power.

  Thereafter, the IFFT (Inverse Fast Fourier Transform) processing unit 114 converts (generates) the frequency component into a complex delay profile in the time domain (step S106).

  Next, the noise suppression unit 115 obtains a power delay profile from the complex delay profile, regards a sample whose power delay profile is equal to or less than a specified threshold value as noise, and replaces the sample of the complex delay profile with a zero value (step S107).

  The complex delay profile after the noise suppression process is converted back to a frequency component by the FFT processing unit 116 (step S108). As a result, a channel estimation value of each subcarrier in which noise is suppressed is obtained.

  As described above, since the channel estimation value of the first subcarrier is interpolated in the time direction or the frequency direction according to the channel condition, the channel estimation value of the second subcarrier is obtained. The value can be estimated with high accuracy.

  In addition to the channel estimation value of the first subcarrier, the number of IFFT processing points is increased by using the channel estimation value of the second subcarrier, so that the complex delay profile is expressed in detail. In a channel estimation method that suppresses the influence of noise by generating a complex delay profile by performing IFFT processing on the channel estimation value of a subcarrier, and replacing a component equal to or less than a predetermined threshold value as noise with a zero value (patent) The accuracy of noise suppression is improved and the channel estimation value can be estimated with high accuracy.

JP 2008-167088 A

Takashi Dateki, Daisuke Ogawa, Hideto Furukawa, "OFDM Channel Estimation by Adding a Virtual Channel Frequency Response", IEICE General Conference, B-5-94 , 2006

  However, when the insertion interval of the reference signal in the time direction is large, for example, as in the case of the transmission antenna T2 and the transmission antenna T3 in LTE (FIG. 10), when channel fluctuation occurs between two reference signals, When the channel estimation value of the second subcarrier is generated by extrapolating in the direction, the channel characteristics may not be improved even if the channel estimation value of the second subcarrier is used.

  The present invention has been made in view of such a situation, and makes it possible to accurately perform channel estimation even when the insertion interval of the reference signal in the time direction is large.

  The signal processing apparatus according to the present invention is a signal processing apparatus that performs channel estimation in wireless communication using a plurality of subcarriers, and performs inverse Fourier transform processing that performs inverse Fourier transform processing on the channel estimation value of the subcarrier, and inverse Fourier transform processing. A removing means for removing a sample of a predetermined range including the center of the obtained complex delay profile, and a Fourier transform means for performing a Fourier transform process on the complex delay profile from which the sample of the predetermined range has been removed. .

  The channel estimation value of the first subcarrier to which the reference signal is mapped is calculated from the first estimation means for obtaining the channel estimation value from the reference signal and the second subcarrier at a predetermined position between the first subcarriers. Is further calculated by interpolating the channel estimation value of the first subcarrier, and the inverse Fourier transform unit reverses the channel estimation value of the first subcarrier and the second subcarrier. Fourier transform processing can be performed.

  The second estimation means compares the difference between the channel estimation values of the first subcarrier located in the time direction with respect to the second subcarrier and the difference between the channel estimation values of the first subcarrier located in the frequency direction. Then, the channel estimation value with the smaller difference is interpolated to estimate the channel estimation value of the second subcarrier, and the removal means uses the second estimation means to determine the channel of the first subcarrier located in the time direction. When the estimated value is interpolated to obtain the channel estimated value of the second subcarrier, samples in a predetermined range including the center of the complex delay profile obtained by the inverse Fourier transform process can be removed.

  The removing means obtains a sample of a predetermined range including the center of the complex delay profile obtained by the inverse Fourier transform process when the channel estimation value of the second subcarrier is obtained by extrapolation by the second estimating means. Can be removed.

  The removing means can replace the value of the sample in the predetermined range with a value smaller than the zero value or the predetermined value.

  The peak value after the effective delay path length of the complex delay profile can be replaced with a zero value or a value smaller than a predetermined value.

  According to the present invention, channel estimation can be performed with high accuracy even when the reference signal insertion interval in the time direction is large.

It is a block diagram which shows the structural example of a transmitter. It is a block diagram which shows the structural example of the receiver as embodiment of this invention. It is a block diagram which shows the structural example of the channel estimation part of FIG. It is a flowchart which shows the flow of the channel estimation process in the channel estimation part of FIG. It is a figure which shows the error of the channel estimated value of a 2nd subcarrier, and actual channel specification. It is a figure which shows that the influence of the error of the channel estimation value of a 2nd subcarrier is removed. It is a block diagram which shows the structural example of a general channel estimation part. It is a flowchart which shows the flow of the channel estimation process in the channel estimation part of FIG. It is a figure which shows the example of mapping of a reference signal. It is a figure which shows the other example of mapping of a reference signal.

  As an embodiment of the present invention, channel estimation in 3GPP LTE will be described below.

  FIG. 1 is a block diagram illustrating a configuration example of a general LTE transmitter 1. The transmitter 1 includes a channel encoding unit 11, a channel modulation unit 12, an IFFT (Inverse Fast Fourier Transform) processing unit 13, a CP (Cyclic Prefix) addition unit 14, a D / A (Digital / Analog) conversion unit 15, and a transmission. An antenna 16 is provided.

  Transmission data addressed to each user supplied to the transmitter 1 is input to the channel encoder 11. Transmission data input to the channel coding unit 11 is subjected to error detection coding and error correction coding by the channel coding unit 11, and is mapped to the I component or Q component together with the reference signal by the channel modulation unit 12. (FIG. 9 or FIG. 10).

  Next, after being converted into a signal wave in the time domain by the IFFT processing unit 13, a CP is added to the head of the OFDM symbol by the CP adding unit 14 in order to prevent the influence of intersymbol interference due to multipath. The OFDM symbol to which the CP is added is converted from a digital signal to an analog signal by the D / A converter 15 and then transmitted from the transmission antenna 16.

  FIG. 2 is a block diagram showing a configuration example of an LTE receiver 2 to which the present invention is applied. The receiver 2 includes a receiving antenna 21, an A / D (Analog / Digital) converter 22, an FFT timing detector 23, a CP remover 24, an FFT (Fast Fourier Transform) processor 25, a channel estimator 26, and a demodulator 27. , And a decoding unit 28.

  A received signal received from the receiving antenna 21 is converted from an analog signal to a digital signal by the A / D converter 22 and then input to the FFT timing detector 23. The FFT timing detection unit 23 detects the timing for performing the fast Fourier transform (FFT) by means for detecting the autocorrelation peak of the received signal.

  Using this Fast Fourier Transform (FFT) timing information, the CP removal unit 24 cuts out a signal sequence in the effective symbol section from the received signal and outputs it to the FFT processing unit 25. The FFT processing unit 25 performs fast Fourier transform (FFT) processing on the input signal sequence in the effective symbol period, and as a result, a reception signal of each subcarrier is obtained.

  The reception signal of each subcarrier obtained by the FFT processing unit 25 is input to the channel estimation unit 26, and the channel estimation unit 26 uses the reference signal multiplexed and transmitted with the data symbol to transmit each subcarrier. A channel estimate is estimated.

  The reception signal of each subcarrier is multiplied by the complex conjugate number of each channel estimation value. As a result, signal distortion in the communication path due to multipath fading or the like is compensated. Thereafter, the demodulation unit 27 performs demodulation processing of the data signal based on the received signal of each subcarrier, and the demodulated data signal is error-corrected and decoded by the decoding unit 28 to restore the transmission data signal.

  FIG. 3 is a block diagram illustrating a configuration example of the channel estimation unit 26. FIG. 4 is a flowchart showing a flow of channel estimation processing in the channel estimation unit 26. In this channel estimation unit 26, a channel estimation unit 51 and a noise suppression unit 52 are provided instead of the channel estimation unit 112 and the noise suppression unit 115 of the channel estimation unit 101 of FIG. 7, and a control unit 53 is further provided. .

  With reference to FIG.3 and FIG.4, the operation | movement of each part of the channel estimation part 26 and the channel estimation process in the channel estimation part 26 are demonstrated.

  The channel estimation unit 111 cancels the reference signal pattern multiplexed together with the data symbols from the received signal of each subcarrier supplied to the channel estimation unit 26, and the channel of the first subcarrier to which the reference signal is mapped An estimated value is obtained (step S11).

  The channel estimation unit 51 obtains the channel estimation value of the second subcarrier located at a predetermined position between the first subcarriers based on the channel estimation value of the first subcarrier.

  Specifically, similar to the channel estimation unit 112 in FIG. 7, the average value of the difference between the channel estimation values of the first subcarriers adjacent in the time direction and the first subcarriers adjacent in the frequency direction The average values of the channel estimation value differences are compared (step S12).

  If it is determined that the difference in channel estimation value of the first subcarrier in the time direction is smaller, the channel estimation value of the first subcarrier located next to the second subcarrier in the time direction is For example, the channel estimation value of the second subcarrier is obtained (generated) by linear interpolation (step S13). On the other hand, if it is determined that the difference in channel estimation value of the first subcarrier in the frequency direction is smaller, the channel estimation value of the first subcarrier located adjacent to the second subcarrier in the frequency direction is For example, the channel estimation value of the second subcarrier is obtained (generated) by linear interpolation (step S14).

  When the channel estimation unit 51 obtains the channel estimation value of the second subcarrier in this way, an interpolation method for obtaining the channel estimation value, that is, the channel estimation of the first subcarrier located next in the time direction is performed. Interpolated values (i.e., interpolated in the time direction) to obtain channel estimates for the second subcarrier, or interpolated channel estimates for the first subcarrier located next in the frequency direction (i.e., Information indicating whether the channel estimation value of the second subcarrier has been obtained by interpolating in the frequency direction is notified to the control unit 53 (step S15).

  The virtual waveform adding unit 113 adds the number of samples to the channel estimation value of the first subcarrier (step S11) and the channel estimation value of the second subcarrier (step S13 or step S14) input from the channel estimation unit 51. A waveform is added so that becomes a power of 2 (step S16).

  Thereafter, the IFFT processing unit 114 converts (generates) the frequency component into a complex delay profile in the time domain (step S17).

  Next, similarly to the noise suppression unit 115 in FIG. 7, the noise suppression unit 52 obtains a power delay profile from the complex delay profile, regards samples whose power delay profile is equal to or less than a specified threshold as noise, The sample is replaced with 0 value (step S18).

  The noise suppression unit 52 further determines an interpolation method when generating the channel estimation value of the second subcarrier (step S19). When it is obtained by interpolation in the time direction (that is, when interpolation is performed in the direction in which the reference signal insertion interval is large), the noise suppression unit 52 does not depend on the size of the power delay profile, but the complex delay profile. A sample in a predetermined range including the center of is replaced with 0 value (step S20). The interpolation method is appropriately notified by the control unit 53.

  When the insertion interval of the reference signal in the time direction is large, such as the transmission antenna T2 and the transmission antenna T3 in the LTE system, channel fluctuation may occur between the two reference signals. In such a case, if the channel estimation value of the second subcarrier is obtained by interpolation in the time direction, an error between the actual channel characteristics and the channel estimation value may increase. An example is shown in FIG.

  The solid line in FIG. 5 shows the actual channel characteristics. That is, in the example of FIG. 5, since channel variation occurs between the reference signals, an error between the channel estimation value of the second subcarrier obtained by interpolation in the time direction and the actual channel characteristic is large.

  The interpolation method is determined based on the difference between the channel estimation values of the first subcarrier in the time direction and the channel estimation value of the first subcarrier in the frequency direction. Even when the insertion interval is large, interpolation may be performed in the time direction.

  As described above, when the error between the channel estimation value of the second subcarrier and the actual channel characteristic is large, when the channel estimation value is plotted on the IQ plane, the accurate first subcarrier obtained from the reference signal is obtained. 6 and the channel estimation value of the second subcarrier having a large error are alternately arranged, so that a sawtooth waveform is obtained as shown in the uppermost part of FIG.

  In the power delay profile of such a waveform, as shown in the middle part of FIG. Since this peak is larger than the noise suppression threshold, it cannot be removed by the noise suppression process (step S18 in FIG. 4), which causes reception characteristic deterioration.

  Therefore, the accuracy of channel estimation can be improved by forcibly replacing this central portion with a zero value and removing the sample at that portion. The lowermost part of FIG. 6 is obtained by replacing the entire 1/3 size range including the center of the power delay profile in the middle part of FIG. 6 with a zero value, performing FFT processing, and converting the frequency domain channel estimation value again. It is. Compared with the channel estimation value shown at the top of FIG. 6, it can be seen that the waveform is smooth and the error of the channel estimation value of the second subcarrier is removed.

  Returning to FIG. 4, when the channel estimation value of the second subcarrier is obtained by interpolation in the frequency direction, the process of replacing the central portion of the complex delay profile with the 0 value is not performed (that is, the process of step S20 is skipped). ).

  The complex delay profile after the noise suppression processing (the complex delay profile in which the central portion is replaced with 0 value in step S20 or the complex delay profile in which the processing is skipped) is converted into frequency components again by the FFT processing unit 116. (Step S21). As a result, noise is suppressed and a channel estimation value for each subcarrier is obtained.

  As described above, in the wireless communication system using the OFDM communication scheme, the sample at the center of the complex delay profile of the channel estimation value is removed, so that channel estimation can be performed with high accuracy.

  In the above description, a channel estimation method for obtaining a complex delay profile and performing noise suppression as in the channel estimation methods described in Non-Patent Document 1 and Patent Document 1 has been described as an example. However, noise suppression processing is not necessarily performed. There is no need to do it. However, if noise suppression processing is used, the accuracy of channel estimation can be improved and the reception characteristics can be improved with a slight change.

  Further, in the above, only when the channel estimation value of the second subcarrier is obtained by interpolation in the time direction, the process of replacing the central portion of the complex delay profile with the zero value is performed. It is also possible to always perform processing for replacing the central portion with a zero value. Furthermore, in a system in which the reference signal insertion interval differs depending on the transmission antenna, such as LTE, as described above, the process of replacing the central portion of the delay profile with a zero value may be performed only when the insertion interval is large. It is also possible to perform the processing for all transmission antennas.

  In the above description, when the channel estimation value of the second subcarrier is obtained by interpolation in the time direction, the center portion of the complex delay profile is replaced with 0 value. Can be used to replace the central portion of the complex delay profile with a zero value when the moving speed is high and the channel time fluctuation is severe.

  In the above description, the present invention has been described as a countermeasure when the insertion interval of the reference signal is large. However, the second subcarrier channel estimation value is obtained by extrapolation regardless of the insertion interval. The present invention can also be used as an improvement method when the accuracy of the channel estimation value of the subcarrier is poor.

  In the above, the range of 1/3 of the whole including the center of the complex delay profile is replaced with 0 value, but it is not necessarily limited to this range. Depending on the channel conditions, the range to be replaced with the zero value may be changed or moved. Alternatively, a peak after an effective delay path length (CP length) may be detected, and the peak position and its surroundings may be replaced with zero values. Further, the replacement value is not limited to the 0 value, and may be another value as long as the value is sufficiently small to be regarded as being removed.

  In the above description, the LTE discussed in 3GPP has been described as an example, but the present invention is not necessarily limited thereto. The present invention can be similarly applied to systems using other OFDM communication systems and other wireless communication systems.

  The above processing functions can be realized by a computer. In this case, a program describing the processing contents of the functions that the signal processing apparatus should have is provided. By executing the program on a computer, the above processing functions are realized on the computer.

  26 channel estimation unit, 51 provisional estimation value generation unit, 52 noise suppression unit, 53 control unit

Claims (3)

In a signal processing apparatus that performs channel estimation in wireless communication using a plurality of subcarriers,
An inverse Fourier transform means for performing an inverse Fourier transform process on the channel estimation value of the subcarrier;
And removing means for removing a sample of the predetermined range including the center of the complex delay profile obtained by the prior Kigyaku Fourier transform,
Before SL predetermined range of the previous SL complex delay profile samples have been removed have a Fourier transform means for Fourier transform processing,
First estimation means for obtaining a channel estimation value of a first subcarrier to which a reference signal is mapped from the reference signal;
Second estimation means for obtaining a channel estimation value of a second subcarrier at a predetermined position between the first subcarriers by interpolating the channel estimation value of the first subcarrier;
Further comprising
The inverse Fourier transform means performs an inverse Fourier transform process on the channel estimation values of the first subcarrier and the second subcarrier,
The second estimation means includes a difference between channel estimation values of the first subcarrier located in the time direction with respect to the second subcarrier and a channel estimation value of the first subcarrier located in the frequency direction. And estimating the channel estimation value of the second subcarrier by interpolating the smaller channel estimation value,
When the second estimation means interpolates the channel estimation value of the first subcarrier located in the time direction and obtains the channel estimation value of the second subcarrier, the removing means A signal processing apparatus that removes samples in a predetermined range including the center of a complex delay profile obtained by Fourier transform processing . The signal processing device according to claim 1,
The signal processing apparatus according to claim 1, wherein the removing unit replaces the value of the sample in the predetermined range with a value of 0 or smaller than a predetermined value. The signal processing device according to claim 1,
A signal processing apparatus, wherein a peak value after an effective delay path length of the complex delay profile is replaced with a zero value or a value smaller than a predetermined value.

Images