JP6103843B2 - Noise removal circuit, receiver, and noise removal method - Google Patents

Description

  The present invention relates to a noise removal circuit, a receiver, and a noise removal method, and particularly suitable for efficiently removing spurious noise (hereinafter also referred to as spurious) contained in a reception signal in a reception circuit in wireless communication. The present invention relates to a noise removal circuit, a receiver, and a noise removal method.

  For example, in wireless communication such as terrestrial digital broadcasting, spurious with periodicity may be superimposed on received data. The period (or frequency) of the spurious is indefinite, and in order to remove this spurious, it is necessary to specify the frequency of the spurious.

Patent Document 1 discloses a technique for identifying the frequency of such a spurious and removing the spurious. In Patent Document 1, a signal α delayed by a time Δt by a cross product detector (CPD) with respect to a received signal α _m t (m is a spurious index) converted at a spurious frequency estimated with coarse accuracy. _m (t−Δt) and a cross product operation (“y = α _m t _I ^* α _m (t−Δt) _Q ^* α _m (t−Δt) _I ”), and the output is subjected to an IIR filter to obtain detailed information. The frequency is detected (“(y _IIR (p) = βx _IIR (p) + (1−β) y _IIR (p−1))”), and the spurious frequency is detected by repeatedly updating the coarse frequency. The spurious of the frequency is removed.

In the above equation, “t _I ” and “(t−Δt) _I ” represent the I component (I phase) of the complex signal, and “(t−Δt) _Q ” represents the Q component (Q phase) of the complex signal. Represents. Also, “p” is the discrete time index, “x _IIR ” is the IIR filter input, “y _IIR ” is the IIR filter output, and “β” is the tracking loop response time for tradeoff between accuracy and tracking speed. Is a loop gain constant (less than 1) that can be adjusted to change.

Special table 2011-520392

  For example, in a car navigation system or the like, in order to remove spurious signals while continuously receiving data during movement, it is necessary to specify a high-precision spurious frequency at high speed.

  However, in the technique described in Patent Document 1, the coarse frequency is updated by repeating the cross product operation with the delay signal delayed by the delay Δt by the cross product detector, and the detailed spurious frequency is detected. Therefore, if the delay Δt value is large, the frequency accuracy for detection becomes coarse instead of the detection convergence rate being high, and conversely, if the delay Δt value is small, the detection frequency accuracy is high but the detection convergence rate is slow. Become. As described above, when the frequency accuracy is rough, there is a problem that an error occurs in the removal of spurious components and the received signal is deteriorated. Further, when the convergence speed is low, there is a problem that it is impossible to follow the fluctuation of the spurious component.

  The present invention has been made to solve the above-described problems, and an object thereof is to enable detection and removal of unknown spurious frequencies at high speed and with high accuracy.

In order to achieve the above object, the noise removal circuit of the present invention comprises a plurality of sample signals stored by sampling a received signal received a plurality of times at a predetermined time interval, a received signal received, based on, includes a frequency detector for detecting the frequency of the spurious contained in the received signal, and a noise suppression unit for removing the spurs detected frequency by the frequency detecting unit from the received signal, said frequency The detection unit extracts the plurality of sample signals at a time interval obtained by multiplying the time interval by a power of a predetermined coefficient, and detects a spurious frequency included in the reception signal using the extracted plurality of sample signals. Do.
Further, the noise removal circuit of the present invention is based on each of a plurality of sample signals stored by sampling a received signal multiple times at a predetermined time interval, and the received signal received, A frequency detector that detects a frequency of spurious included in the received signal; and a noise suppressor that removes the spurious of the frequency detected by the frequency detector from the received signal, and the frequency detector A phase conversion unit for deriving phase information of the sample signal from the two-dimensional vector quantities of the I phase and Q phase of each of the plurality of sample signals; a storage unit for storing the phase information of each of the plurality of sample signals; The product of the complex conjugate of the unit vector having the phase information of each sample signal stored in the storage unit and the two-dimensional vector quantities of the I-phase and Q-phase of the received signal is individually provided. A plurality of complex conjugate sections to be calculated in a column, and a plurality of spurious frequencies in each sample signal that are individually determined in parallel based on the phase of each vector information individually calculated by the plurality of complex conjugate sections. A frequency determination unit, and detects a frequency obtained by adding the frequencies detected by the plurality of frequency determination units as the spurious frequency.

  On the other hand, in order to achieve the above object, the receiver of the present invention includes the noise removing circuit, a receiving unit that receives a signal transmitted by a carrier wave and generates the received signal to be input to the noise removing circuit, It has.

On the other hand, in order to achieve the above object, the noise removal method of the present invention includes a plurality of sample signals stored by sampling the received signal received a plurality of times at predetermined time intervals, and the received signal received. And a first step of detecting a spurious frequency included in the received signal, and a second step of removing the spurious of the frequency detected in the first step from the received signal. only contains the first step, the extracting the plurality of sample signals at a time interval multiplied by the power of predetermined coefficients to the time interval, the extraction plurality of received signal using a sample signal The spurious frequency included is detected .
Further, the noise removal method of the present invention is based on each of a plurality of sample signals stored by sampling a received reception signal a plurality of times at a predetermined time interval and the received reception signal. a first step of detecting the frequency of the spurious contained in the received signal, seen containing a second step, the removing the spurious frequency detected by the first step from the received signal, the first The step of deriving phase information of the sample signal from the I-phase and Q-phase two-dimensional vector quantities of each of the plurality of sample signals, and storing the phase information of each of the plurality of sample signals The product of the complex conjugate of the unit vector having the phase information of each sample signal stored in the unit and the two-dimensional vector quantities of the I-phase and Q-phase of the received signal are individually parallelized And a step of individually determining in parallel each spurious frequency in each sample signal based on the phase of each vector information calculated individually, and detected in the first step. A frequency obtained by adding the respective frequencies is detected as the spurious frequency.

  According to the present invention, the determination of the spurious frequency included in the received signal based on each of the plurality of sample signals and the received signal is performed in parallel, and a desired spurious frequency is detected based on the determined frequencies and detected. It is possible to remove spurious at a high speed and with high accuracy by detecting an unknown spurious frequency.

It is a block diagram which shows the structure of the OFDM demodulator of the receiver which concerns on embodiment. It is a block diagram which shows the structure of the noise removal circuit which concerns on embodiment. It is explanatory drawing which shows the example of a processing content of the phase conversion part in FIG. It is explanatory drawing which shows the 1st processing content of the frequency determination part in FIG. It is explanatory drawing which shows the 2nd processing content of the frequency determination part in FIG. It is explanatory drawing which shows the 3rd processing content of the frequency determination part in FIG. It is a block diagram which shows the structure of the noise suppression part in FIG. It is explanatory drawing which shows the 1st processing content of the noise removal circuit which concerns on embodiment. It is explanatory drawing which shows the 2nd processing content of the noise removal circuit which concerns on embodiment. It is explanatory drawing which shows the 3rd processing content of the noise removal circuit which concerns on embodiment. It is a flowchart which shows the noise removal process example by the noise removal circuit which concerns on embodiment.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. FIG. 1 shows a configuration of a receiver 10 according to the present embodiment. The receiver 10 includes an antenna 1, an RF 2, an ADC 3, a mixer / filter 4, and an FFT 5, and further includes a spurious detection unit 6 a and a noise suppression unit 6 b, and a spurious noise removal unit 6 as a noise removal circuit according to the present invention. It has.

  The receiver 10 performs, for example, OFDM (Orthogonal Frequency Division Multiplexing) demodulation, and constitutes an OFDM demodulator that performs OFDM demodulation on a received signal transmitted on a carrier wave. The mixer / filter 4, FFT 5, and A spurious noise removing unit 6 is provided. In addition, although the case where it applies to an OFDM system is described as an example here, it is not limited to OFDM format, It is possible to apply also to the receiver of a single carrier transmission system, etc.

  Note that OFDM modulation / demodulation is used in, for example, terrestrial digital broadcasting, and is a digital modulation / demodulation system that can provide video signals and audio signals efficiently by providing a large number of sub-carriers within one channel band. is there. The OFDM signal is a complex signal.

  The antenna 1 receives an RF (Radio Frequency) signal (radio wave), and the RF 2 converts the frequency of the RF signal received by the antenna 1. The ADC 3 is an analog-digital converter, and converts an analog signal obtained by frequency conversion with the RF 2 into a digital signal at a predetermined sampling frequency.

  The mixer / filter 4 performs a predetermined filter process on the digital signal generated by the ADC 3 and performs quadrature demodulation on the digital signal to form a base composed of a real axis component (I-phase signal) and an imaginary axis component (Q-phase signal). A band OFDM signal is generated.

  The FFT 5 normally receives the baseband OFDM signal generated by the mixer / filter 4, but when the baseband OFDM signal includes spurious power having a large power value, the baseband output from the spurious noise removing unit 6. An OFDM signal is input, fast Fourier transform (FFT operation) is performed, and the signal is converted into a frequency domain complex signal including a data signal and a pilot signal. Note that demodulated data is generated by a decoder (not shown) using the frequency domain signal output from the FFT 5.

  The spurious noise removal unit 6 receives the baseband OFDM signal generated by the mixer / filter 4, detects the spurious frequency from the baseband OFDM signal by the spurious detection unit 6a, and detects the spurious frequency by the noise suppression unit 6b. The noise of the baseband OFDM signal generated by the mixer / filter 4 is removed using the spurious frequency and input to the FFT 5.

  In the present embodiment, the demodulator including the FFT 5 is used. Although not shown in FIG. 1, the subsequent stage of FFT 5 is provided with means for detecting the presence or absence of spurious with a rough accuracy by a technique for finding a singular point in the spectrum distribution. Based on the above, the spurious noise removing unit 6 starts operating.

  In Patent Document 1 described above, since the cross product detector repeats the cross product operation with the delayed signal delayed by the delay Δt to update the coarse frequency and detect the detailed spurious frequency, the delay Δt If the value is large, the frequency accuracy of detection becomes coarse instead of the detection convergence speed being fast, and conversely, if the value of the delay Δt is small, the frequency accuracy of detection is high but the detection convergence speed is slow. As described above, when the frequency accuracy is rough, there is a problem that an error occurs in removing spurious components and the received signal is deteriorated. Further, when the convergence speed is slow, there is a problem that it is impossible to follow the fluctuation of the spurious component.

  In the present embodiment, the spurious detection unit 6a in the spurious noise removal unit 6 is configured as shown in FIG. 2, and N received signals (N is N) obtained by sampling a received signal received by wireless communication a plurality of times at a predetermined time interval (Δt). The frequency of each spurious in the first to Nth sample signals based on a plurality of (natural number) sample signals (first to Nth sample signals) and the input received signal (hereinafter also referred to as input signal) In parallel, the processing to detect each of the detected frequencies is performed in parallel, and the spurious of the frequency obtained by adding the detected frequencies is removed from the input signal. Detect and remove.

Hereinafter, the configuration and operation of the spurious noise removing unit 6 will be described with reference to FIG. As shown in FIG. 2, the spurious detection unit 6 a in the spurious noise removal unit 6 includes a phase conversion unit 6 a 1, a memory 6 a 2, and N complex conjugate units (described as “complex conjugate” in the drawing) 6 a 3 _{1 to} 6 a 3 _n , N filters 6a4 _{1 to} 6a4 _n and N frequency determination units (described as “frequency determination” in the drawing) 6a5 _{1 to} 6a5 _n are provided.

A phase converter 6a1, a memory 6a2, and frequency determiners 6a5 _{1 to} 6a5 _n in the spurious detector 6a are provided as new functions according to the present invention.

  Based on the two-dimensional vector amounts of the I phase and Q phase of the received signal (sample signal) output from the mixer / filter 4, the phase converter 6a1 determines the phase of the sample signal in advance on 2π radians. The phase information is converted into phase information for identifying each area equally divided by the ratio.

  In this embodiment, 16 divisions are used, and 0, 1, 2,..., 14, 15 are associated as phase information for identifying each region, and the phase conversion unit 6a1 sets the phase of the sample signal to 0-15. Convert as phase information.

  The phase information converted by the phase converter 6a1 is stored in the memory 6a2. That is, in the memory 6a2, phase information represented by 4 bits is stored in place of the vector information of each sample signal (first to Nth delay samples).

In FIG. 2, delay 1, delay 2, delay 3, and delay N are described with respect to the memory 6 a 2. For example, the delay 1 is a received signal (input signal) received when detecting a spurious frequency. ) Means the storage area for storing the phase information of the first sample signal sampled by 1 × Δt (k ⁿ × Δt, where n = 0).

Similarly, the delay 2 means a storage area for storing phase information of the second sample signal sampled by k × Δt (k ⁿ × Δt and n = 1) time before the input signal. The delay N means a storage area for storing phase information of the Nth sample signal sampled before k ⁿ × Δt time from the input signal.

  In the present embodiment, each sampling signal obtained by sampling the received signal a plurality of times at the time interval Δt is stored in the memory 6a2, and when the spurious frequency is detected, the predetermined coefficient k is set to the time interval Δt. Each sampling signal is extracted at time intervals multiplied by a power, and the phase information of each sampling signal extracted in this way is used as delay 1, delay 2, delay 3,.

In this embodiment, the coefficient k is set to 4 in the extraction time intervals of the first to Nth sampling signals. In this way, the coefficient k is not limited to 4. However, according to the processing in the frequency determination units 6a5 _{1 to} 6a5 _n described later (spurious frequency determination is performed by dividing the frequency domain into four). “4”.

Hereinafter, in the present embodiment, the sample time interval of each of the first to Nth sample signals is also described as a delay time. That is, the delay time of each of the first to Nth sample signals is 4 ⁿ times the delay time of the first sample signal, for example, the first delay time is “1 × Δt”, the second The delay time is “4 × Δt”, the third delay time is “16 × Δt”, and the Nth delay time is “4 ⁿ × Δt”.

Thereby, in the memory 6a2, each phase information (0 to 15) composed of the 4-bit information converted by the phase conversion unit 6a1 is delayed from 1 to delay according to each delay time (1 × Δt to 4 ⁿ × Δt). Stored as N.

  FIG. 3 schematically shows a process of converting the I-phase and Q-phase two-dimensional vector quantities of the sample signal into phase information (0 to 15) divided into 16 by the phase converter 6a1. The diagram shows the result of converting the phase obtained from the two-dimensional vector quantities of the I phase and Q phase of the sample signal as phase information “5”, that is, 4-bit information [0101].

  In this way, by converting the phase information into 16 divisions, the phase information represented by the two-dimensional vector amounts of the I-phase and Q-phase of each sample signal can be represented by only 4 bits, reducing the information amount. It is possible to reduce the used capacity of the memory 6a2 for storing the information.

When removing the spurious frequency, the 4-bit phase information stored in the memory 6a2 is converted into a unit vector having each phase information in a vector conversion unit (not shown), Each is input to the complex conjugate sections 6a3 _{1 to} 6a3 _n .

  In the conversion into the unit vector, a predetermined phase value, for example, a median value is used for each of the 16 divided phase information (0 to 15).

In each of the complex conjugate units 6a3 _{1 to} 6a3 _n , each of the first to Nth sample signals stored in the memory 6a2 is individually subjected to parallel processing, and the input output from the mixer / filter 4 By the product (inner product) of the I-phase and Q-phase two-dimensional vector quantities of the signal and the complex conjugate of the unit vectors of the first to N-th sample signals having the phase information converted by the vector conversion unit described above The vector information of each of the first to Nth sample signals is derived.

In the derivation process of the vector information in each of the complex conjugate units 6a3 _{1 to} 6a3 _n , by performing an inner product of the received signal (input signal) and the complex conjugate of the unit vector having each phase information of delays 1 to N, The spurious phase information (rotation) is obtained by rotating the input signal.

Vector information derived in each complex conjugate unit _6a3 1 _{~6a3 n} (spurious phase information (rotation)) is input to a filter _6a4 1 _{~6a4 n,} is an infinite impulse response processing in the filter _6a4 1 _{~6a4 n,} each from spurious phase information derived in complex conjugate unit _6a3 1 _{~6a3 n,} after variation in the received signal (input signal) has been removed, it is input to the frequency determining section _6a5 1 _{~6a5 n.}

The frequency determination units 6a5 _{1 to} 6a5 _n individually process the spurious frequencies related to the first to Nth sample signals in parallel from the vector information (spurious phase information) input from the filters 6a4 _{1 to} 6a4 _n. To calculate.

In the present embodiment, each frequency determination unit 6a5 _{1 to} 6a5 _n determines which region of the divided region (in this case, ¼) the spurious phase is equally divided at a predetermined ratio. judge.

For example, regarding the phase (φ ₀ ) of delay 1 of 1 × Δt, it is determined in which region of 1/4 of the entire region the phase (φ ₀ ) exists. Also, 4 with respect to the × Delta] t of the delay 2 phase (4 × φ _0), the phase (4 × φ ₀₎ is the 1/4 region of 1/4 of the entire area of which is determined by the delay 1 It is determined which is present. This is because the phase of delay 2 of 4 × Δt (4 × φ ₀ ) is 4 times delay (4 times sample time interval), and 4 × 90 degrees = 360 degrees cannot be determined. This is because the phase represented by “× φ ₀ ” corresponds to any one of the ¼ region of the entire ¼ region determined by the delay 1.

Each of the frequency determination units 6a5 _{1 to} 6a5 _n has a spurious phase value obtained from each of the delays 1 to N (or a predetermined value for each region corresponding to the phase value, in this case, a median value. ) Is divided by each delay value (4 ⁿ × Δt) to determine the spurious frequency for each sample signal.

In this way, a value obtained by adding the spurious frequencies related to the first to Nth sample signals obtained in the frequency determination units 6a5 _{1 to} 6a5 _n is detected as a spurious frequency to be removed from the input signal.

FIG. 4 shows processing by the frequency determination units 6a5 _{1 to} 6a5 _n .

As shown in FIG. 4, the frequency determining unit 6a5 ₁ determines spurious frequency of the first sample signal is a signal sampled at the first delay time (1 × Delta] t), the frequency of the angle information phi _0.

Frequency determining unit 6a5 ₂ may determine the spurious frequency of the second sample signal is a signal sampled at a second delay time (4 × Delta] t), the frequency of the angle information _{(φ M = 4 × φ 0} mod 2π) To do. The determination of the frequency in the frequency determination section 6a5 _2, in the frequency band of the input signal, it is possible to specify the quarter of the region including a frequency determined by the frequency determining unit 6a5 _1.

Moreover, the frequency determining portion 6a5 ₃ is a spurious frequency of the third third of the sample signal delayed by the time (16 × Δt) is sampled signals, the angle information _{(φ M + 1 = 4 ×} 4 × φ 0 mod 2π) The frequency is determined. The determination of the frequency in the frequency determination section 6a5 _3, including in the frequency band of the input signal, a quarter of the region containing the frequency determined by the frequency determining unit 6a5 _2, i.e., a frequency determined by the frequency determining unit 6a5 ₁ An area of 1/16 can be specified.

FIG. 5 shows the processing contents for the first to N-th sample signals (delay 1 to delay N) by the frequency determination units 6a5 _{1 to} 6a5 _n . The detection frequency band shown in FIG. 5 is a band region of a reception signal to be detected by each frequency determination unit 6a5 _{1 to} 6a5 _n . Here, for the first sample signal (delay 1), the input signal band BW All of _sys are targeted, and for the second sample signal (delay 2), ¼ of the band BW _sys of the input signal is targeted, and for the Nth sample signal (delay N), the band BW _sys of the input signal is targeted. 1/4 ^N is the target.

The determination phase Ai indicates a corresponding phase region of the specified frequency, and here, there are four regions of π / 4, 3π / 4, −π / 4, and −3π / 4. The determination frequency output unit BWi represents an input signal area calculated by each of the frequency determination units 6a5 _{1 to} 6a5 _n , that is, a narrowed frequency area. For example, the first sample signal (delay 1) is narrowed down to ¼ of the band BW _sys of the input signal, and the second sample signal (delay 2) is narrowed down to 1/16 of the band BW _sys of the input signal. Thus, the Nth sample signal (delay N) is narrowed down to 1/4 ^{N + 1} of the bandwidth BW _sys of the input signal.

According to this frequency determining unit _6a5 1 _{~6a5 n,} the processing content for the sample signal of the N from the first is sequentially added at each frequency determination unit _6a5 1 _{~6a5 n,} wherein in FIG. 5 "spurious frequencies : Fspr = Σ ((Ai / (π / 2)) × BWi) "frequency determination unit 6a5 _n of addition result obtained in is detected as finally sought spurious frequencies.

FIG. 6 shows a state where the narrowing-down process of the determination frequency output unit BWi for each of the first to Nth sample signals by the frequency determination units 6a5 _{1 to} 6a5 _n is developed with the frequency as the horizontal axis.

In FIG. 6, in order to search for and specify the spurious frequency to be specified (the spurious frequency being searched for) in the band region (spurious frequency search region) of the input signal, first, the frequency determination unit 6a5 ₁ for the first sample signal. Determines any region in the spurious frequency search region divided into four equal parts. Here, the third region (1-3) is determined.

Frequency determining unit 6a5 ₂ for the second sample signal determines any region in the spurious frequency search region third region determined by the frequency determining unit 6a5 ₁ is further divided into four equal parts. Here, the third region (2-3) is determined. In the same manner, to narrow down the spurious frequency search region to the frequency determination unit 6a5 _n.

Each of the frequency determination units 6a5 _{1 to} 6a5 _n specifies a phase value predetermined for the narrowed-down region as a spurious phase related to the sample signal. Here, the median value in the band of each region is a phase value determined in advance for the region.

Then, each of the frequency determination units 6a5 _{1 to} 6a5 _n divides the spurious phase thus identified by each delay value (4Mn × Δt) to obtain the spurious frequency related to the sample signal.

Furthermore, as described above with reference to FIG. 5, the processing result for each of the first to Nth sample signals by each of the frequency determination units 6a5 _{1 to} 6a5 _n is expressed by the expression “spurious frequency: Fspr = Σ (( Ai / (π / 2)) × BWi) ”is sequentially added in each of the frequency determination units 6a5 _{1 to} 6a5 _n , and the addition result of the frequency determination unit 6a5 _n is finally obtained. It is detected as a desired spurious frequency.

In this way, spurious frequency is the addition result of the frequency determination section 6a5 _n is input is output from the spurious detection unit 6a to the noise suppression unit 6b.

  As shown in FIG. 7, the noise suppression unit 6b includes a mixer 6b1, a DC offset removal unit 6b2, a DC offset detection unit 6b3, and a mixer 6b4. The mixer 6b1 mixes and combines the spurious frequency input from the spurious detector 6a and the input signal input from the mixer / filter 4, and outputs the mixed signal. Here, the mixer 6b1 removes the spurious frequency component input from the spurious detection unit 6a from the input signal input from the mixer / filter 4.

  The signal output from the mixer 6b1 is input to the DC offset removal unit 6b2 and the DC offset detection unit 6b3, and the DC offset detection unit 6b3 obtains the amplitude of the spurious and outputs it to the DC offset removal unit 6b2.

  The DC offset removal unit 6b2 removes the spurious amplitude value input from the DC offset detection unit 6b3 from the signal input from the mixer 6b1, and outputs the result to the mixer 6b4.

  The mixer 6b4 mixes and combines the signal input from the DC offset removal unit 6b2 and the spurious frequency input from the spurious detection unit 6a, and outputs the resultant to the FFT 5. Here, the mixer 6b4 removes spurious from the input signal by frequency-converting the signal input from the DC offset removal unit 6b2 using the spurious frequency input from the spurious detection unit 6a.

  8 to 10 show the noise removal processing operation from the received signal according to the present embodiment by the OFDM demodulator in the receiver 10 shown in FIG. 1, and FIG. 8 shows the OFDM signal to be received and the spurious signal. (Spurious noise) is shown, and it is shown that the amplitude of the OFDM signal varies depending on the frequency component of the spurious included in the OFDM signal to be received.

  Each signal shown in FIG. 9 corresponds to the spurious frequency conversion processing of the noise suppression unit 6b having the configuration shown in FIG. 7, and the mixer 6b1 receives an OFDM signal including spurious noise input from the mixer / filter 4. Then, by mixing and synthesizing with the spurious frequency input from the spurious detection unit 6a, an (× (−Fspr)) OFDM signal from which the spurious frequency component has been removed is obtained.

  In FIG. 10, the signal after removing the spurious amplitude value detected by the DC offset detection unit 6b3 and the mixer 6b4 in the DC offset removal unit 6b2 with respect to the OFDM signal from which the spurious frequency component in FIG. 9 has been removed. 2 shows a signal obtained by frequency-converting (× (+ Fspr)) the signal input from the DC offset removal unit 6b2 using the spurious frequency input from the spurious detection unit 6a.

  Next, a noise removal processing example according to the present embodiment will be described with reference to FIG. First, a received signal received by wireless communication is sampled a plurality of times at a predetermined time interval (Δt) and stored in a memory (step 1101).

Next, by parallel processing, the first to Nth sample signals are extracted at a time interval (4 ⁿ × Δt) obtained by multiplying the time interval (Δt) by a power of a predetermined coefficient (k, here 4). Then, a spurious frequency for each sample signal is derived, and a spurious frequency derived by parallel processing is combined to detect a desired spurious frequency (step 1102). Then, the detected spurious frequency is removed from the input signal (step 1103).

  As described above with reference to each drawing, in this embodiment, the frequency of spurious (periodic noise) included in the received signal is sampled at time intervals from the received signal and stored in the memory. A spurious is removed from a received signal (input signal) based on a frequency obtained by adding each calculated frequency separately in parallel processing based on the signal. Here, based on the result obtained based on the first sample signal sampled by 1 × Δt time before the input signal, it is determined in which quarter region of the unit circle the spurious exists, Based on the result obtained based on the second sample signal sampled by 4 × Δt time before the input signal, the ¼ region obtained based on the first sample signal is further reduced to ¼ region. By determining in which region the spurious frequency exists, the region where the spurious frequency exists is narrowed down. This makes it possible to detect and remove unknown spurious frequencies at high speed and with high accuracy.

  In the present embodiment, the information of each sample signal stored in the memory is only phase information (4-bit information representing 0 to 15) obtained from the two-dimensional vector quantities of the I-phase and Q-phase of the received signal. . Thereby, the memory capacity of the memory used for storing each sample signal can be reduced.

  The present invention is not limited to the embodiments described with reference to the drawings, and various modifications can be made without departing from the scope of the invention. For example, in the present embodiment, each phase information of each sample signal stored in the memory (storage device) is 4-bit information indicating one of 16 divided areas consisting of 0 to 15, but divided into 8 pieces. It may be 3-bit information indicating any one of the areas or 5-bit information indicating any of the 32 divided areas.

  In the present embodiment, when extracting the first to Nth sample signals to be processed, the sampling time interval Δt is multiplied by a power of a predetermined coefficient (k = 4). This is a configuration in which the spurious frequency is determined in a region that is narrowed down by a quarter for each of the first to Nth sample signals by dividing the region determined by the frequency determination unit into four according to the coefficient 4 However, the area may be narrowed down by another ratio such as 1/8. In this way, when the area is narrowed down by 1/8, the coefficient k may be set to 8.

  In this way, by making the coefficient k and the area narrowing ratio the same, it is possible to avoid overlapping of spurious frequencies in adjacent sampling signals, so that more efficient processing is achieved. However, the configuration may be such that the coefficient k and the area narrowing ratio are not the same.

  Also, when extracting the first to Nth sample signals, not the time interval obtained by multiplying the sampling time interval Δt by a power of a predetermined coefficient (k = 4), for example, the first sample signal The extraction time interval may be sequentially increased from the first to the Nth sample signal. For example, the extraction time interval may be sequentially increased from the first sample signal to the Nth sample signal as 1 × Δt, 2 × Δt, 3 × Δt,..., N × Δt.

  In the description of the present embodiment, the spurious noise removing unit 6 determines the spurious frequency by means of detecting the presence / absence of spurious noise with a rough accuracy (not shown) provided in the subsequent stage of the FFT 5 described above. The number of the Nth sample signal may be designated from the first sample signal, and the Nth sample signal may be designated as a processing target from the third sample signal instead of the first sample signal. Thereby, spurious noise removal processing according to desired accuracy can be executed, and the amount of memory 6a2 used can be reduced.

  In the present embodiment, the detection of the spurious frequency by the spurious detection unit 6a may be performed once when the receiver 10 is activated, for example, or the spurious frequency changes due to a change in the external environment or the like. Alternatively, the spurious frequency may be periodically detected by the spurious detector 6a.

  Further, the spurious noise removing unit 6 and the OFDM demodulator including the spurious noise removing unit 6 according to the present embodiment can be applied to a television receiver, a mobile phone, a game machine, a car navigation system, and the like.

1 Antenna 2 RF
3 ADC
4 Mixer / Filter 5 FFT
6 spurious noise removing unit 6a spurious detecting unit 6b noise suppressing unit 6a1 phase converting unit 6a2 memory 6a3 _{1 to} 6a3 _n complex conjugate unit 6a4 _{1 to} 6a4 _n filter 6a5 _{1 to} 6a5 _n frequency judging unit 6b1 mixer 6b2 DC offset removing unit 6b3 DC Offset detector 6b4 Mixer 10 Receiver

Claims (10)

A spurious frequency included in the received signal is detected based on each of a plurality of sample signals stored by sampling the received signal a plurality of times at predetermined time intervals and stored. A frequency detector to perform,
A noise suppression unit that removes the spurious frequency detected by the frequency detection unit from the received signal;
Equipped with a,
The frequency detection unit extracts the plurality of sample signals at a time interval obtained by multiplying the time interval by a power of a predetermined coefficient, and uses the extracted plurality of sample signals to detect a spurious frequency included in the reception signal. Noise removal circuit that detects noise. A spurious frequency included in the received signal is detected based on each of a plurality of sample signals stored by sampling the received signal a plurality of times at predetermined time intervals and stored. A frequency detector to perform,
A noise suppression unit that removes the spurious frequency detected by the frequency detection unit from the received signal;
Equipped with a,
The frequency detector
A phase converter for deriving phase information of the sample signal from two-dimensional vector quantities of the I phase and Q phase of each of the plurality of sample signals;
A storage unit for storing the phase information of each of the plurality of sample signals;
A plurality of complex conjugates that individually calculate the product of the complex conjugate of a unit vector having phase information of each sample signal stored in the storage unit and the two-dimensional vector quantities of the I-phase and Q-phase of the received signal, respectively, in parallel And
A plurality of frequency determination units that individually determine in parallel each of the spurious frequencies in each sample signal based on the phase of each vector information individually calculated by the plurality of complex conjugate units;
With
A noise removal circuit that detects a frequency obtained by adding the frequencies detected by the plurality of frequency determination units as the spurious frequency . The frequency detection unit is different time intervals to extract a plurality of sample signals, extracted by using a plurality of sample signals to detect the frequency of the spurious contained in the received signal according to claim 1 or 2 noise removal according circuit. The frequency detection unit extracts the plurality of sample signals at a time interval obtained by multiplying the time interval by a power of a predetermined coefficient, and uses the extracted plurality of sample signals to detect a spurious frequency included in the reception signal. The noise removal circuit according to any one of claims 1 to 3 , wherein detection of noise is performed. Wherein the frequency detecting section, the detection of the spurious frequency based on the respective said received signals of said plurality of sample signals, each comprising a plurality of frequency determining section for performing individually parallel for each of the plurality of sample signals, The frequency obtained by adding the respective frequencies detected by the plurality of frequency determination units is detected as the spurious frequency. Claim 3 dependent on claim 1, claim 3 dependent on claim 1, or claim 4 dependent on claim 1 Noise removal circuit. 3. The noise removal circuit according to claim 2 , wherein the phase information derived by the phase conversion unit includes bit information that identifies each divided region equally divided at a predetermined ratio on 2π radians. The infinite impulse response processing is individually performed on the vector information calculated by the product of the complex conjugate of the unit vector and the two-dimensional vector quantities of the I phase and Q phase of the received signal by the plurality of complex conjugate units. It has a plurality of filter parts to be performed in parallel
7. The noise removal circuit according to claim 2, wherein the plurality of frequency determination units determine a spurious frequency in each sample signal based on a phase in the vector information subjected to infinite impulse response processing by the plurality of filter units. . The noise removal circuit according to any one of claims 1 to 7,
A receiving unit that receives a signal transmitted by a carrier wave and generates the received signal to be input to the noise removing circuit;
With receiver. A spurious frequency included in the received signal is detected based on each of a plurality of sample signals stored by sampling the received signal a plurality of times at predetermined time intervals and stored. A first step to:
A second step of removing from the received signal the spurious at the frequency detected in the first step;
Only including,
The first step extracts the plurality of sample signals at a time interval obtained by multiplying the time interval by a power of a predetermined coefficient, and uses the extracted plurality of sample signals to detect spurious signals included in the received signal. A noise removal method for frequency detection . A spurious frequency included in the received signal is detected based on each of a plurality of sample signals stored by sampling the received signal a plurality of times at predetermined time intervals and stored. A first step to:
A second step of removing from the received signal the spurious at the frequency detected in the first step;
Only including,
The first step includes
Deriving phase information of the sample signal from two-dimensional vector quantities of the I phase and Q phase of each of the plurality of sample signals;
A complex conjugate of a unit vector having phase information of each sample signal stored in a storage unit that stores the phase information of each of the plurality of sample signals, and a two-dimensional vector amount of I phase and Q phase of the received signal Calculating each product individually in parallel;
Determining in parallel each of the spurious frequencies in each sample signal based on the phase of each vector information individually calculated;
With
A noise removal method for detecting a frequency obtained by adding the frequencies detected in the first step as the spurious frequency .