JP4836251B2 - Signal equalization apparatus, reception apparatus, signal equalization method, and signal equalization program - Google Patents

Description

  The present invention relates to a signal equalization apparatus, a signal equalization method, and a signal equalization program that receive a signal including a pilot signal and perform equalization processing.

In the transmission method in which the pilot signal is included in the transmission signal, the received signal is corrected (equalized) based on the pilot signal with the configuration as shown in FIG. That is, in the receiving apparatus 100, the analog signal processing circuit 2 receives a signal including a pilot signal via the antenna 1. The equalization processing unit 3 performs Fourier transform on the output signal of the analog signal processing circuit 2 by the FFT processing unit 3a, and outputs the result to the equalization unit 3c and the pilot signal acquisition unit 3b. The pilot signal acquisition unit 3b acquires a pilot signal from the signal after the FFT processing and outputs the pilot signal to the equalization unit 3c.
The equalization unit 3c determines a correction amount so that the signal level of the actually received pilot signal matches the known signal level, and corrects the output signal of the FFT processing unit 3a based on this correction amount. In addition, a technique for controlling the directivity of an antenna is known in order to suppress the influence of signals transmitted through a plurality of transmission paths (see, for example, Patent Document 1).
JP 2005-252767 A

However, the conventional technique cannot suppress the influence of noise.
That is, although the above equalization processing is performed based on the pilot signal, the pilot signal is also affected by noise applied in the transmission path. Therefore, when these effects are large, the correction amount is estimated based on the pilot signal. An error may occur and the reception accuracy may be reduced.

In addition, even if the influence of multiple transmission paths is suppressed by controlling the antenna directivity so that the received power is maximized as in the technique disclosed in the above publication, noise mixed in a specific transmission path The influence of can not be suppressed. Furthermore, the configuration for controlling the directivity of the antenna is complicated and expensive, and it cannot be applied to a device that does not include an antenna, such as a video deck.
The present invention has been made in view of the above points, and an object thereof is to provide a general-purpose technique for improving the effectiveness of equalization processing with a simple configuration.

  In order to achieve the above object, a signal equalizer according to the present invention acquires a delay profile from a pilot signal, and removes noise by paying attention to a profile for each delay time in the delay profile. At this time, since the time change characteristic is different between the amplitude component and the phase component in the profile for each delay time, noise removal processing is performed in accordance with each characteristic. As a result, noise can be reliably removed for each of the plurality of transmission lines. If the pilot signal is estimated based on the signal after removing the noise for each transmission path in this way, and equalization processing is performed, the influence of noise can be suppressed and demodulated. Therefore, the demodulation process can be performed accurately.

  In other words, the major degradation factors of radio signals include signal interference caused by transmission through multiple transmission paths and noise mixed in each transmission path during transmission. Usually, reception is affected by these degradation factors. The received power varies greatly depending on the position and frequency of the machine. In the equalization process, in order to suppress the influence of this fluctuation, each signal is corrected so that the signal level of the pilot signal is equal to a known level. However, as described above, the signal deterioration factor is complicated, and the noise level However, when the characteristics of the transmission line are drastically changed, the conventional technique cannot perform appropriate equalization processing.

  However, in the present invention, since the delay profile is acquired based on the pilot signal, the pilot signal can be analyzed as a profile for each delay time (amplitude component and phase component of the signal propagated through each transmission path). When the pilot signal is decomposed into profiles for each delay time, the profile at each delay time is also affected by noise, but this noise has the same tendency for the same delay time (ie, for each same transmission line). There are many. However, the tendency is different between the amplitude component and the phase component.

  Therefore, by referring to the time change of the profile for each delay time and suppressing the time change for each of the amplitude component and the phase component, noise can be individually removed for each transmission line. If noise can be removed for each transmission path with a profile for each delay time, the degradation factor of the signal remaining in the delay profile is only the influence of transmission through a plurality of transmission paths, and the degradation factor is extremely simplified. Therefore, it is possible to accurately perform equalization processing by estimating the pilot signal based on the delay profile after removing the noise.

  Here, the signal acquisition means only needs to be able to acquire a signal including a pilot signal, and a signal transmitted by a transmission method for transmitting a pilot signal at a specific rule (for example, a specific time and frequency interval) at the time of communication. Can be adopted. In addition, in the pilot signal, the reference value may be known so that equalization processing can be performed. For example, a signal with known transmission timing, frequency, amplitude, phase, and the like can be employed.

  The pilot signal acquisition means only needs to be able to acquire a plurality of pilot signals transmitted at different times, which are pilot signals included in the received signal. For example, a configuration in which a signal at a specific time is Fourier transformed to extract a spectrum at a known frequency can be employed.

The delay profile acquisition unit only needs to be able to acquire a delay profile indicating the intensity of the signal for each delay time included in each of the plurality of pilot signals. For example, it can be obtained by performing an inverse Fourier transform on a pilot signal expressed by a spectrum with respect to frequency.

  In the noise removing means, a profile corresponding to a common delay time is extracted from delay profiles corresponding to a plurality of times, and noise is suppressed by suppressing a steep change in each of the amplitude component and the phase component based on the time change. As long as it can be removed. For example, a configuration for acquiring a profile for each delay time from a plurality of delay profiles, averaging the change of the amplitude component for each common delay time, and removing the noise by approximating the change of the phase component to a linear change. It can be adopted.

  That is, the applicant converts a plurality of pilot signals into a delay profile and analyzes the profile for each delay time in each delay profile, and if the profile is for each delay time, the amplitude component does not vary greatly in time, It was found that the phase component fluctuates with a certain tendency in time. More specifically, since the amplitude component has a small element that changes with time in a specific tendency and means that the time change is mainly caused by noise, the noise component is calculated by calculating the time average for the amplitude component. Can be removed.

  Further, since the phase component changes with time in a specific tendency, noise is superimposed on this change. Accordingly, noise can be removed by approximating the phase component so as to change linearly with respect to time. As a result, noise can be accurately removed for each of the amplitude component and the phase component. Therefore, by referring to the delay profile after removing the noise, the pilot signal after removing the noise can be estimated. it can.

For example, it is possible to obtain a frequency spectrum of the pilot signal after the noise removal if Fourier transform the delay profile after removing the noise further. Therefore, using the frequency spectrum as a reference, the signal level after conversion and the level of the known pilot signal are compared to calculate a correction amount, and the signal level of other timing and frequency may be corrected with this correction amount.

  Of course, the above apparatus can be applied to various apparatuses such as a receiving apparatus, and the signal equalization described above proceeds with processing according to a procedure peculiar to the present application. It can also be realized as an invention of the method described above. Moreover, it is realizable also as invention of the program for making a computer implement | achieve the procedure. Further, the signal equalization apparatus, method, and program may be realized as a part of another apparatus, method, and program, or may be realized by combining a part of a plurality of apparatuses, methods, and programs. Various aspects can be adopted. Of course, the present invention can be realized as a recording medium on which the program is recorded.

Hereinafter, embodiments of the present invention will be described in the following order.
(1) Configuration of receiving apparatus:
(2) Configuration of equalization processing unit:
(3) Operation of equalization processing unit:
(4) Other embodiments:

(1) Configuration of receiving apparatus:
FIG. 1 is a block diagram showing a receiving apparatus 10 including a signal equalizing apparatus according to an embodiment of the present invention. The receiving apparatus 10 includes an antenna 11, an analog signal processing circuit 12, an equalization processing unit 13, a demodulation unit 14, and a data processing unit 15, and receives radio waves via the antenna 11. In the embodiment described below, the reception device 10 is a digital broadcast reception device, and receives broadcast radio waves obtained by multiplexing encoded data by an OFDM (Orthogonal Frequency Division Multiplexing) transmission method.

  That is, the analog signal processing circuit 12 receives a broadcast wave modulated by OFDM via the antenna 11, performs analog-digital conversion, removal of guard intervals, etc., and performs a plurality of subcarrier signals modulated by a predetermined modulation method. Is generated. The equalization processing unit 13 extracts a pilot signal from this signal, performs correction to remove noise, and performs equalization processing.

  In this embodiment, at the time of this noise removal, by acquiring a delay profile from a plurality of pilot signals, the transmission characteristic for each transmission path in each pilot signal is acquired, and noise removal processing is performed for each transmission path. At this time, separate noise removal processing is performed for each of the amplitude component and the phase component based on the fact that the time change characteristics are different from each other.

  That is, since noise is mixed independently for each transmission path, it is possible to reliably remove noise by performing noise removal separately for each transmission path. Further, since appropriate noise removal is performed according to the time change characteristics of the amplitude component and the phase component, noise can be reliably removed. Although the delay profile after removing noise in this way does not suppress the influence of multiple transmission paths, the main cause of signal degradation is only the influence of multiple transmission paths because the influence of noise is excluded. . If the pilot signal is estimated from the profile after removing the noise and the equalization process is performed, the equalization process can be performed accurately.

  The demodulator 14 demodulates each subcarrier based on the equalized signal, and the data processor 15 performs data decoding processing and data processing based on the decoded data based on the demodulated signal. In the present embodiment, since the equalization processing is performed based on the signal after removing the noise as described above, the demodulation in the demodulation unit 14 and the decoding in the data processing unit 15 are performed very accurately. .

  As the data processing in the data processing unit 15, any data processing on the decoded data can be assumed. For example, if the receiving device 10 is a digital broadcast video viewing device, the decoded data Based on the above, data processing for displaying an image on the screen is performed, and the image is displayed on a display screen (not shown). Of course, the receiving device 10 is not limited to a digital broadcast video viewing device, and various devices such as a digital broadcast video recording device and a portable terminal can be the receiving device 10 of the present invention.

(2) Configuration of equalization processing unit:
Next, the configuration of the equalization processing unit 13 will be described. The equalization processing unit 13 includes an FFT unit 13a, a pilot signal acquisition unit 13b, an inverse FFT unit 13c, a noise removal unit 13d , an FFT unit 13e, an equalization unit 13f, and a buffer memory 13g, and outputs an output signal of each unit. Equalization processing is performed while accumulating in the buffer memory as necessary. The FFT processing unit 13a acquires the signal output from the analog signal processing circuit 12, and performs FFT (Fast Fourier Transform) processing.

  The pilot signal acquisition unit 13b acquires an output signal of the FFT processing unit 13a, acquires a pilot signal from the acquired signal, and outputs the pilot signal. That is, since the transmission timing and frequency of the pilot signal are known, the pilot signal acquisition unit 13b acquires a signal having a predetermined frequency at a predetermined time from the output signal of the FFT processing unit 13a. Of course, the frequency range of the signal to be extracted as the pilot signal can be adjusted as appropriate.

The inverse FFT processing unit 13c acquires the output signal of the pilot signal acquisition unit 13b, performs an inverse FFT process, and outputs it. As a result, a profile (delay profile) in which the amplitude component and the phase component of each pilot signal are expressed with respect to time is obtained. The peaks in this profile are discretely arranged with respect to time, and the time of each peak corresponds to the delay time of the signal. Therefore, the profile for each delay time corresponds to the profile for each transmission path.

The noise removing unit 13d acquires the output signal of the inverse FFT processing unit 13c, and performs noise removal based on the profile for each delay time. That is, a profile corresponding to a common delay time is extracted from each of the delay profiles generated from a plurality of pilot signals, and noise is removed by averaging the amplitude component and the phase component in each delay time individually. Here, it is assumed that the amplitude component does not vary greatly with respect to time, and the time average is calculated. On the other hand, the phase component is assumed to fluctuate with a constant tendency with respect to time, and the change with respect to time is linearly approximated.

F FT processor 13e obtains the output signal of the noise removal unit 13d, and outputs performs F FT process. That is, the pilot signal is estimated from the delay profile from which noise is removed. In the delay profile after the noise is removed, the influence of the signal transmitted through the plurality of transmission paths is not excluded, but the influence of the plurality of transmission paths is corrected by the processing of the equalization unit. .

Equalization unit 13f acquires an output signal of the F FT processing unit 13e, calculates a correction amount for correcting the output signal to match the predetermined amplitude and phase for the pilot signal, the FFT processing unit The output signal 13a is corrected. That is, based on correction amounts calculated based on a plurality of pilot signals, correction amounts for other signals are calculated by interpolation or the like, and each signal is corrected.

In this embodiment, the FFT processing unit 13a and the pilot signal acquisition unit 13b correspond to a pilot signal acquisition unit, the inverse FFT processing unit 13c corresponds to a delay profile acquisition unit, and the noise removal unit 13d serves as a noise removal unit. corresponds to, F FT processor 13e corresponds to the pilot signal estimation unit, an equalization unit 13f corresponds to the equalizing means.

(3) Operation of equalization processing unit:
Next, the operation in the above equalization processing unit 13 will be described. FIG. 2 is a flowchart showing processing in the equalization processing unit 13. When the signal is output by the analog signal processing circuit 12, the FFT processing unit 13a acquires the signal and performs a Fourier transform (step S100). FIG. 3 is a schematic diagram showing the signal after the Fourier transform. In each graph, the horizontal axis represents frequency and the vertical axis represents carrier power.

  Note that the FFT processing unit 13a sequentially performs FFT processing in units of a certain time interval, and obtains amplitude components and phase components corresponding to a plurality of carriers by performing Fourier transform for each time interval. Can do. In FIG. 3, the power component (the square of the amplitude component) is calculated from the FFT processing results of the signals at different times, and arranged in a separate graph so that time elapses from top to bottom. Further, the transmission timing and frequency of the pilot signal are known, and indicated as P in FIG.

  The amplitude component and phase component of each carrier fluctuate depending on the frequency and time, and the power component fluctuates as shown in FIG. 3 to reflect this fluctuation. There are two factors, that is, transmission and that different noise is mixed for each transmission path, and the variation appearing in the frequency spectrum shown in FIG. 3 is extremely complicated.

  Therefore, in the present embodiment, in order to simplify and correct this complicated factor, first, the pilot signal acquisition unit 13b specifies the pilot signal (P in FIG. 3) (step S105), and as an individual signal spectrum A pilot signal is acquired (step S110). Since the timing and frequency of pilot signal P are known, pilot signal acquisition unit 13b can specify a plurality of pilot signals by specifying a predetermined frequency at a predetermined timing.

The inverse FFT processing unit 13c performs inverse Fourier transform on the plurality of pilot signals acquired in this way (step S115). The result obtained is the delay profile described above. Here, the coefficient of exp when the signal after inverse FFT processing is displayed as a pole is the amplitude component, and the exponent of exp is the phase component. FIG. 4 schematically shows the result of calculating the power component from the result of performing the inverse Fourier transform on a pilot signal of a certain frequency on the graph, and shows the calculation results for the pilot signal at different times side by side. Yes. In these graphs, the horizontal axis indicates the delay time, and the vertical axis indicates the power of the signal that has passed through the transmission line of the delay time. Here, the time component within the same delay profile is set as the delay time, and is distinguished from the time indicating the transmission timing of the pilot signal.

  As shown in FIG. 4, in the delay profile, a plurality of signal levels are present in a delay time different from a delay time at which a signal having the highest signal level (a bold arrow shown in FIG. 4 is normally treated as a desired wave) appears. The peak shift of each signal appears corresponding to the transmission time when transmitting through a plurality of transmission paths. Therefore, if a profile at each peak (or a profile within a predetermined range around it) is analyzed, analysis for each transmission path having different transmission characteristics is performed.

Therefore, the noise removing unit 13d specifies a delay time in which peaks appear in a plurality of delay profiles (step S120). For example, as in the example shown in FIG. 4, numbers t _n (n is an integer) are defined in order from the peak corresponding to a small delay time. The delay profile is acquired for a plurality of pilot signals having different transmission times. If a profile corresponding to a common number is acquired from the delay profile corresponding to each pilot signal, the time change of the profile for each transmission path is analyzed. be able to. Therefore, it is possible to remove noise by suppressing the temporal variation.

  FIG. 5 is a diagram illustrating temporal changes in the amplitude component and the phase component of the profile at a specific delay time. In FIG. 5, the amplitude component is shown in the upper part and the phase component is shown in the lower part, and the horizontal axis is time. As shown in the upper part of FIG. 5, the amplitude component has a specific tendency and does not fluctuate greatly in the profile at a specific delay time, but the value fluctuates due to noise mixing. On the other hand, in the profile at a specific delay time, the value of the phase component fluctuates with a certain tendency as shown in the lower part of FIG. 5 and further fluctuates due to noise.

In order to remove noise according to such characteristics, the noise removing unit 13d acquires a profile corresponding to a common number from the delay profile corresponding to each pilot signal, performs time averaging on the amplitude component, and obtains the profile. Is corrected (step S125). Further, the phase component is approximated so as to change linearly (step S130). For example, when a certain pilot signal is to be corrected, the profile corresponding to the delay time t _n is extracted with reference to the delay profile of the pilot signal to be corrected and the delay profile of the pilot signal at the previous and subsequent times.

Then, an average is calculated for the amplitude component, and correction is performed so that the amplitude component of the profile of the delay time t _n becomes the average value in the delay profile corresponding to the pilot signal to be corrected. As a result, the amplitude component of the profile of the delay time t _n is corrected as indicated by the broken line shown in the upper part of FIG. 5, and noise is removed. In addition, regarding the phase component, it is considered that the temporal change of the phase can be approximated by a linear function, and a linear function indicating the temporal change of the phase is calculated from a plurality of profiles corresponding to the delay time t _n, and the pilot signal to be corrected In the corresponding delay profile, correction is performed so that the phase component of the profile of the delay time t _n becomes a value calculated from a linear function. As a result, the phase component of the profile of the delay time t _n is corrected as indicated by the broken line shown in the lower part of FIG. 5, and noise is removed.

Noise removal unit 13d, the above noise removal process performed on the plurality of the delay time t _n. In other words, since the noise characteristics tend to be different for each transmission path specified by the delay time, accurate noise removal corresponding to the characteristics of each transmission path is performed by removing noise by averaging and linear approximation for each delay time. It can be performed. In the above-described averaging of amplitude components and linear approximation of phase components, the number of delay profiles to be considered can be appropriately changed within a range in which noise can be effectively removed.

When noise is removed for each delay time, the noise is removed from the delay profile. In this embodiment, it is estimated that this state is a correct delay profile. Then, F FT processor 13e is a delay profile corresponding estimated conversion Fourier (step S135), estimates a pilot signal. When the pilot signal is estimated, the equalization unit 13f compares the predetermined pilot signal with the estimated pilot signal to calculate a correction amount (step S140), and uses the correction amounts of a plurality of pilot signals, and the surrounding data. The signal is corrected (step S145). It should be noted that a known method can be employed for the equalization processing in the equalization unit 13f.

  In the above processing, the noise mixed in the pilot signal is effectively removed before the equalization processing, and the influence of the noise can be reduced to about half. That is, when such noise removal is not performed, noise is mixed in the calculated correction amount because noise is mixed in the pilot signal compared with the reference value in calculating the correction amount. For this reason, at the time of correction, correction based on information in which noise is further mixed is performed on a signal in which noise is mixed. However, in this embodiment, since noise is removed before calculating the correction amount, the calculated correction amount is not affected by noise. Therefore, the main cause of noise can be halved, and extremely effective correction can be performed.

(4) Other embodiments:
The above-described embodiment is an embodiment of the present invention, and the embodiment of the present invention is not limited to the above-described embodiment. That is, various configurations can be adopted as long as the equalization processing can be performed by removing the influence of noise for each transmission path based on the pilot signal. For example, the present invention can be applied to devices other than digital broadcast receivers as in the above-described embodiments, and receive wireless communication devices other than digital broadcasts and transmission signals other than wireless broadcasts such as wired signals. The present invention may be applied to an apparatus.

  Furthermore, various methods other than the above-described embodiment can be adopted as a method for removing noise based on the above-described amplitude component and phase component. That is, it is only necessary that the time variation characteristic of the amplitude component and the time variation characteristic of the phase component are different and noise removal corresponding to each characteristic can be performed. For example, noise may be removed by interpolating the amplitude component and the phase component with different interpolation functions. In other words, if an interpolation function corresponding to each time variation characteristic of the amplitude component and the phase component is specified in advance and the interpolation function is calculated based on a specific delay time profile, noise corresponding to each characteristic is calculated. Removal can be performed.

It is a block diagram which shows the signal equalization apparatus concerning one Embodiment of this invention. It is a flowchart of an equalization process. It is a figure which shows the signal after a Fourier-transform. It is a figure which shows the example of calculation of a delay profile. It is a figure which shows the time change of the profile in specific delay time. It is a block diagram of the conventional signal equalization apparatus.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 ... Reception apparatus 11 ... Antenna 12 ... Analog signal processing circuit 13 ... Equalization processing part 13a ... FFT processing part 13b ... Pilot signal acquisition part 13c ... Inverse FFT processing part 13d ... Noise removal part 13e ... FFT processing part 13f ... etc. 13 g ... Buffer memory 14 ... Demodulator 15 ... Data processor

Claims (5)

Signal acquisition means for acquiring a signal including a pilot signal;
Pilot signal acquisition means for acquiring a plurality of pilot signals transmitted at different times from the signal;
Delay profile acquisition means for acquiring delay profiles in the plurality of pilot signals;
A noise removing unit that obtains a profile for each delay time in the delay profile, averages the amplitude component thereof, and approximates a time change of the phase component to a linear change to remove noise;
Pilot signal estimating means for estimating a pilot signal based on a delay profile after removing the noise;
Equalizing means for equalizing the signal based on the estimated pilot signal. The pilot signal acquisition means performs a Fourier transform process on the signal, extracts a signal having a predetermined time and frequency as a pilot signal,
The signal equalizer according to claim 1, wherein the delay profile acquisition unit acquires a delay profile by performing inverse Fourier transform on the extracted pilot signal.   A radio signal receiving apparatus comprising the signal equalization apparatus according to claim 1. A signal equalization method for equalizing a signal,
A signal acquisition step of acquiring a signal including a pilot signal;
A pilot signal acquisition step of acquiring a plurality of pilot signals transmitted at different times from the signal;
A delay profile acquisition step of acquiring a delay profile in the plurality of pilot signals;
A noise removal step of obtaining a profile for each delay time in the delay profile and averaging the amplitude component thereof, removing the noise by approximating the time change of the phase component to a linear change,
A pilot signal estimating step of estimating a pilot signal based on the delay profile after removing the noise;
An equalization step of equalizing the signal based on the estimated pilot signal. A signal equalization program for equalizing signals,
A signal acquisition function for acquiring a signal including a pilot signal;
A pilot signal acquisition function for acquiring a plurality of pilot signals transmitted at different times from the signal;
A delay profile acquisition function for acquiring a delay profile in the plurality of pilot signals;
A noise removal function that obtains a profile for each delay time in the delay profile, averages the amplitude component thereof, and approximates the time change of the phase component to a linear change to remove noise;
A pilot signal estimation function for estimating a pilot signal based on a delay profile after removing the noise;
A signal equalization program for causing a computer to realize an equalization function for equalizing the signal based on the estimated pilot signal.

Images