CN102694563A - Digital demodulation down-conversion system and method for acoustical signals - Google Patents

Abstract

The invention provides a technical scheme of a digital demodulation down-conversion system and a digital demodulation down-conversion method for acoustical signals, which is implemented through that: after being subjected to AD (analog-to-digital) conversion, down-conversion, extraction and filtering on acoustical signals, finally output digital baseband demodulation signals are subjected to DA conversion, so that at the demodulation end, clear acoustical signals can be heard. The technical scheme is used for restoring acoustical signals, has great flexibility and expansibility, and can achieve an effect of much lower power consumption, much higher operating frequency, and more reliable performance.

Description

A kind of digital demodulation down conversion system and method that is used for voice signal

Technical field:

What the present invention relates to is a kind of digital demodulation down conversion system and method that is used for voice signal.

Background technology:

In the prior art; When recovering voice signal, technique known uses FM signal receiver analog demodulator mostly; Analog demodulator has four bigger shortcomings on signal processing: at first; The non-thread phase characteristic of intermediate-frequency filter influences its linear FM characteristic greatly, and The model generally can only reach 10dB to the inhibition ability of monkey chatter; The second, for the accent spoke of eliminating in the FM signal disturbs, can be impaired greatly through the signal amplitude of amplitude limiter, make the output signal of frequency discriminator that big distortion arranged; Its three, this analogy method is not suitable for doing on computers real-time simulation, is difficult to through emulation it to be done further research and improvement, therefore can only do experiment with side circuit, the design cycle is longer; The communication system that last this analogy method constructs does not have expandability and programmability, and it defines repertoire by hardware, and life cycle is shorter, and this is the existing in prior technology weak point.

Summary of the invention:

The object of the invention is exactly not enough to existing in prior technology; And a kind of digital demodulation down conversion system of voice signal and technical scheme of method of being used for is provided; Be used to recover voice signal, have very big flexibility and extensibility, much lower power consumption can be arranged; Much higher operating frequency, performance are also more reliable.

This programme is realized through following technical measures: a kind of digital demodulation down conversion system that is used for voice signal is characterized in that comprising like lower module:

Analog-to-digital conversion module is used for the analog if signal sampling that receives is obtained digital medium-frequency signal; The digital mixing module multiplies each other digital medium-frequency signal respectively with two local oscillation signals, quadrature demodulation is an I/Q two-way digital baseband signal, and two local oscillation signals during quadrature demodulation and digital medium-frequency signal are with homophase frequently, and the complete quadrature of described local oscillation signal; The digital demodulation down conversion module is used for I/Q two-way baseband signal is carried out the filtering extraction of integral multiple, carries out shaping filter then, has realized smothing filtering, has eliminated the band external noise, and I/Q two-way baseband signal is reduced to digital audio and video signals; D/A converter module converts digital audio and video signals into analog signal; The bandpass filtering amplification module carries out filtering to analog signal and amplifies output sound signal.

The sample frequency of described digital-to-analogue conversion is 50MHz.

Described digital demodulation down conversion module comprises cic filter, HB filter, 255 rank FIR filters; Cic filter can realize extracting the filtering extraction that factor D is not 2 power power, and the extraction yield of CIC filtering is 4-32, and half-band filter can realize that the D=2M multiple extracts; The maximum extraction yield of half-band filter is 32; Cic filter and half-band filter are united use, realize arbitrary integer extraction doubly, and the FIR filter carries out shaping filter to the signal after extracting.

A kind of digital demodulation down conversion method that is used for voice signal is characterized in that comprising the steps:

1) convert the analog if signal that receives into digital medium-frequency signal through analog to digital converter,

2) use the local oscillation signal of two quadratures that digital medium-frequency signal is carried out quadrature demodulation, obtain I/Q two-way digital baseband signal,

3) to I/Q two-way digital baseband signal extract, filtering, realize smothing filtering, eliminate the band external noise, I/Q two-way baseband signal is reduced to digital audio and video signals;

4) convert digital audio and video signals into analog signal,

5) analog signal is carried out filtering and amplify output sound signal.

The sample frequency of described digital-to-analogue conversion is 50MHz.

Described step 3) is used cic filter, HB filter, 255 rank FIR filters, and cic filter can realize extracting the filtering extraction that factor D is not 2 power power, and the extraction yield of CIC filtering is 4-32, and half-band filter can be realized D=2 ^MMultiple extracts, and the maximum extraction yield of half-band filter is 32,, cic filter and half-band filter are united use, realize arbitrary integer extraction doubly, and the FIR filter carries out shaping filter to the signal after extracting.

Integral comb filter (CIC) be fit to be done when extracting the decimation filter that factor D is not 2 power power.It usually joins together to use with half-band filter, thereby realizes arbitrary integer extraction doubly.

Integral comb filter is meant that the impulse response of this filter has following form:

$h\left(n\right)=\left\{\begin{array}{c}\mathrm{1,0}\≤n\≤D-1\\ 0,\mathrm{else}\end{array}\right.$

Can find out that from following formula cic filter is that a kind of nothing is taken advantage of filter, only need just can realize many times of extractions of signal when using it through addition, this has just reduced operand greatly.Wherein D is the exponent number of filter, also is simultaneously to extract the factor.According to the definition of Z-transformation, the Z-transformation of cic filter is:

$H\left(z\right)=\underset{n=0}{\overset{N-1}{\mathrm{\Σ}}}h\left(n\right)z^{-n}$

$=\frac{1-z^{-N}}{1-z^{-1}}$

$=\frac{1}{1-z^{-1}}(1-z^{-N})$

$=H_1\left(z\right)H_2\left(z\right)$

In the formula,

$H_1\left(z\right)=\frac{1}{1-z^{-1}}$

H ₂(z)=1-z ^-N

It is thus clear that cic filter is made up of two parts: integrator H ₁(z) and comb filter H ₂(z) cascade.The amplitude-frequency characteristic of cic filter.

Z=e ^{J ω}Substitution can get H ₂(z) frequency response is:

$H_2\left(e^{\mathrm{j\ω}}\right)=1-e^{-\mathrm{j\ωN}}$

$=e^{-\mathrm{j\ωN}/2}\·\left[\frac{e^{\mathrm{j\ωN}/2}-e^{-\mathrm{j\ωN}/2}}{2}\right]$

$=e^{-\mathrm{j\ωN}/2}\sin (\mathrm{\ωN}/2)$

Its amplitude-frequency characteristic is:

$\left|H_2\left(e^{\mathrm{j\ω}}\right)\right|=2\left|\sin \left(\frac{\mathrm{\ωN}}{2}\right)\right|$

| H ₂(e ^{J ω}) | shape just as a comb, so be referred to as comb filter to its image.Equally can be in the hope of integrator H ₁(z) frequency response is:

$H_1\left(e^{\mathrm{j\ω}}\right)=\frac{1}{1-e^{-\mathrm{j\ω}}}$

$=\frac{e^{\mathrm{j\ω}/2}}{2}{\left[\frac{e^{\mathrm{j\ω}/2}-e^{-\mathrm{j\ω}/2}}{2}\right]}^{-1}$

$=\frac{e^{\mathrm{j\ω}/2}}{2}\·{\left(\sin \frac{\mathrm{\ω}}{2}\right)}^{-1}$

So the overall frequency response of cic filter is:

$\left|H\left(e^{\mathrm{j\ω}}\right)\right|=|H_1\left(e^{\mathrm{j\ω}}\right)\·H_2\left(e^{\mathrm{j\ω}}\right)|$

$=|\sin (\mathrm{\ωN}/2)/\sin (\mathrm{\ω}/2)|$

$=|N\·\mathrm{Sa}\left(\frac{\mathrm{\ωN}}{2}\right)\·{\mathrm{Sa}}^{-1}\left(\frac{\mathrm{\ω}}{2}\right)|$

In the formula,

Be shan, and Sa (0)=1, so the range value of cic filter at ω=0 place is N, that is: | H (e ^J0) |=N.

Half-band filter can be realized D=2 ^MMultiple extracts.Its computational efficiency is high, and real-time, half-band filter is meant frequency response H (e ^{J ω}) the satisfied following FIR filter that concerns:

ω _A=π-ω _C

δ _s=δ _p=δ

Be the resistance band (π-ω of half-band filter _A) and passband width (ω _C) equate, and the ripple of passband stopband also equates

Half-band filter has following character:

H(e ^jω)=1-H(e ^j(π-ω))

H(e ^jπ/2)=0.5

$h\left(k\right)=\left\{\begin{array}{c}1,k=0\\ 0,k=\±2,\±4,...\end{array}\right.$

The impulse response h (k) that is to say half-band filter is zero at all the other even number points except zero point is non-vanishing entirely, so when adopting half-band filter to realize the conversion of sampling rate, only need general amount of calculation, very high calculation rate is arranged.

Mainly to act on be that whole channel is carried out shaping filter to the FIR filter in the digital demodulation low-converter; Do not make extract function; Behind signal process CIC, the HB filter, the sampling rate that is input to the FIR filter is very low comparatively speaking, therefore under certain processing clock speed; The FIR filtering of higher-order can be arranged, and it is fine to make that the index such as passband fluctuation, transition band bandwidth, stopband minimal attenuation of filter can design.

The unit impact response h (n) of FIR filter, list entries x (n), output sequence y (n), then the input/output relation formula of filter is:

$y\left(n\right)=\underset{m=0}{\overset{N-1}{\mathrm{\Σ}}}h\left(m\right)x(n-m)$

FIR Filter Design task mainly is to select time-limited h (n), approaches transfer function H (e as much as possible ^{J ω}).

This shows that the present invention compared with prior art has outstanding substantive distinguishing features and obvious improvement, the beneficial effect of its enforcement also is conspicuous.

Description of drawings

Fig. 1 is the structural representation of the specific embodiment of the invention

Fig. 2 is the flow chart of the specific embodiment of the invention.

Embodiment

For clearly demonstrating the technical characterstic of this programme, below through an embodiment, and combine its accompanying drawing, this programme is set forth.

Can find out that through accompanying drawing the digital demodulation down conversion system that is used for voice signal of this programme comprises like lower module:

Analog-to-digital conversion module is used for the analog if signal sampling that receives is obtained digital medium-frequency signal; The digital mixing module multiplies each other digital medium-frequency signal respectively with two local oscillation signals, quadrature demodulation is an I/Q two-way digital baseband signal, and two local oscillation signals during quadrature demodulation and digital medium-frequency signal are with homophase frequently, and the complete quadrature of described local oscillation signal; The digital demodulation down conversion module is used for I/Q two-way baseband signal is carried out the filtering extraction of integral multiple, carries out shaping filter then, has realized smothing filtering, has eliminated the band external noise; I/Q two-way baseband signal is reduced to digital audio and video signals; D/A converter module converts digital audio and video signals into analog signal; The bandpass filtering amplification module carries out filtering to analog signal and amplifies output sound signal.

The sample frequency of described digital-to-analogue conversion is 50MHz.

Described digital demodulation down conversion module is mainly accomplished by PDC chip and FPGA; Wherein: the groundwork that the PDC chip is accomplished has: the firstth, and the digital quadrature down-conversion; Digital mixer multiplies each other the orthogonal local oscillation signal of digital medium-frequency signal and digital controlled oscillator (NCO) generation; Generate I/Q two-way mixed frequency signal, interested signal is downconverted to zero intermediate frequency; The secondth, LPF, the filtering out of band signal extracts useful signal; The 3rd is sample rate conversion, reduces sampling rate through digital decimation, extracts factor range greatly the ability that can be designed to broadband narrow-band digital channel alive is provided.The groundwork that FPGA accomplishes has: the control word that configuration down-conversion chip needs; The speed of control figure signal; Synchronised clock and control signal when generating the DA conversion; The function that adds digital filtering in case of necessity.

Described digital demodulation down conversion module comprises cic filter, HB filter, 255 rank FIR filters; Cic filter can realize extracting the filtering extraction that factor D is not 2 power power; Half-band filter can realize that the D=2M multiple extracts; Cic filter and half-band filter are united use, realize arbitrary integer extraction doubly, and the FIR filter carries out shaping filter to the signal after extracting.

This design is to voice signal, so the passband design is at 200-3000Hz.The FIR filter adopts Blackman window to realize LPF, uses MATLAB to produce the prototype low-pass filter coefficients of respective taps, deposits among the RAM before DDC starts working through parameter configuration module.Adopt special-purpose D/A conversion chip to accomplish the recovery of audio signal at demodulating end, according to international standard, the speed of output sound signal is located at 48kHz, exports voice signal and synchronised clock simultaneously through FPGA.

A kind of digital demodulation down conversion method that is used for voice signal comprises the steps: 1) convert the analog if signal that receives into digital medium-frequency signal through analog to digital converter, sample frequency is 50mhz; 2) use the local oscillation signal of two quadratures that digital medium-frequency signal is carried out quadrature demodulation, obtain I/Q two-way digital baseband signal, 3) I/Q two-way digital baseband signal is used cic filter, HB filter, 255 rank FIR filters extract, filtering; Realize smothing filtering, eliminate the band external noise; I/Q two-way baseband signal is reduced to digital audio and video signals, and cic filter can realize extracting the filtering extraction that factor D is not 2 power power, and half-band filter can realize that the D=2M multiple extracts; Cic filter and half-band filter are united use, realize arbitrary integer extraction doubly, and the FIR filter carries out shaping filter to the signal after extracting.4) convert digital audio and video signals into analog signal, 5) analog signal is carried out filtering amplification, output sound signal.

The present invention is not limited in above-mentioned embodiment, and the variation that those of ordinary skills make in essential scope of the present invention, remodeling, interpolation or replacement also should belong to protection scope of the present invention.

Claims (6)

1. digital demodulation down conversion system that is used for voice signal is characterized in that comprising like lower module:

Analog-to-digital conversion module is used for the analog if signal sampling that receives is obtained digital medium-frequency signal;

The digital mixing module multiplies each other digital medium-frequency signal respectively with two local oscillation signals, quadrature demodulation is an I/Q two-way digital baseband signal, and two local oscillation signals during quadrature demodulation and digital medium-frequency signal are with homophase frequently, and the complete quadrature of described local oscillation signal;

The digital demodulation down conversion module is used for I/Q two-way baseband signal is carried out the filtering extraction of integral multiple, carries out shaping filter then, has realized smothing filtering, has eliminated the band external noise, and I/Q two-way baseband signal is reduced to digital audio and video signals;

D/A converter module converts digital audio and video signals into analog signal;

The bandpass filtering amplification module carries out filtering to analog signal and amplifies output sound signal.

2. the digital demodulation down conversion system that is used for voice signal according to claim 1 is characterized in that: the sample frequency of described digital-to-analogue conversion is 50MHz.

3. the digital demodulation down conversion system that is used for voice signal according to claim 1; It is characterized in that: described digital demodulation down conversion module comprises cic filter, HB filter, 255 rank FIR filters; Cic filter can realize extracting the filtering extraction that factor D is not 2 power power, and half-band filter can be realized D=2 ^MMultiple extracts, and cic filter and half-band filter are united use, realizes arbitrary integer extraction doubly, and the FIR filter carries out shaping filter to the signal after extracting.

4. a digital demodulation down conversion method that is used for voice signal is characterized in that comprising the steps:

1) convert the analog if signal that receives into digital medium-frequency signal through analog to digital converter,

2) use the local oscillation signal of two quadratures that digital medium-frequency signal is carried out quadrature demodulation, obtain I/Q two-way digital baseband signal,

3) to I/Q two-way digital baseband signal extract, filtering, realize smothing filtering, eliminate the band external noise, I/Q two-way baseband signal is reduced to digital audio and video signals

4) convert digital audio and video signals into analog signal,

5) analog signal is carried out filtering and amplify output sound signal.

5. the digital demodulation down conversion method that is used for voice signal according to claim 4 is characterized in that: the sample frequency of described digital-to-analogue conversion is 50MHz.

6. the digital demodulation down conversion method that is used for voice signal according to claim 4; It is characterized in that: described step 3) is used cic filter, HB filter, 255 rank FIR filters; Cic filter can realize extracting the filtering extraction that factor D is not 2 power power, and half-band filter can be realized D=2 ^MMultiple extracts, and cic filter and half-band filter are united use, realizes arbitrary integer extraction doubly, and the FIR filter carries out shaping filter to the signal after extracting.

Images