CN102843323B - Asymmetric binary modulation signal receiver - Google Patents

Abstract

The invention discloses an asymmetric binary modulation signal receiver. A simulated impact filter is adopted by the receiver to replace a high-speed ADC (analogue to digital converter) and a digital impact filter in the current EBPSK (extended binary phase shift keying) receiver, therefore, not only are the receiving modulation performance and work efficiency of an EBPSK modulation signal obviously improved, but also the hardware cost and complexity of the EBPSK receiver are greatly reduced.

Description

A kind of asymmetric binary modulation signal receiver

Technical field

The invention belongs to the technical field of signal of communication process, be specifically related to enhancing and the demodulation of digital communication Received signal strength, in particular, being that employing is a kind of simulates the asymmetric binary modulation signal receiver that shock filter directly strengthens the asymmetric binary modulating signal with demodulation.

Background technology

1, asymmetric binary modulating signal

In digital communication system, the process baseband signal representing binary data upwards being moved given transmission frequency range is called modulation, and contrary process is then called demodulation.Binary digit modulation in communication, certain parameter (as amplitude, frequency, phase place etc.) that binary message code element " 0 " or " 1 " can be utilized directly to change (being usually referred to as " skew keying ") sinusoidal carrier realizes, and correspondingly obtains the amplitude shift keying (2-ASK) of binary (binary system), frequency shift keying (2-FSK) and phase shift keying (2-PSK) modulation signal.These binary shifted keying modulations can be collectively expressed as:

$\begin{array}{cc}{s}_0\left(t\right)=A\sin {\mathrm{\&omega;}}_{c0}t,& 0\&le;t<T\\ s_1\left(t\right)=\left\{\begin{array}{c}B\sin ({\mathrm{\&omega;}}_{c1}t+\mathrm{\&theta;}),\\ A\sin {\mathrm{\&omega;}}_{c0}t,\end{array}\right.& \begin{array}{c}0\&le;t<\mathrm{\&tau;},0\&le;\mathrm{\&theta;}\&le;\mathrm{\&pi;}\\ 0\&le;\mathrm{\&tau;}\&le;t<T\end{array}\end{array}---\left(1\right)$

In formula, s ₀(t) and s ₁t () represents the modulation waveform of code element " 0 " and " 1 " respectively; ω _c0for " 0 " is in code-element period T and " 1 " carrier angular frequencies in non-keying period T-τ, ω _c1it is then " 1 " carrier angular frequencies in keying period τ; B-A is the amplitude of carrier wave keying, and θ is the phase place of carrier wave keying.

We modulate (see " unified orthogonal binary shifted strong control modulation and demodulation method ", the patent No.: ZL200710025203.6) with (1) formula definition " unified binary shifted keying " (UBSK:UnifiedBinary Shift Keying).Why calling like this, is because if make τ=T, then (1) formula becomes:

$\begin{array}{cc}\begin{array}{c}{s}_0\left(t\right)=A\sin {\mathrm{\&omega;}}_{c0}t,\\ s_1\left(t\right)=B\sin ({\mathrm{\&omega;}}_{c1}t+\mathrm{\&theta;}),\end{array}& 0\&le;t<T,0\&le;\mathrm{\&theta;}\&le;\mathrm{\&pi;}\end{array}---\left(2\right)$

Be not difficult to find out by (2) formula:

1) if B=A and θ=0, we obtain classical 2-FSK modulation;

2) if ω _c0=ω _c1=ω, we obtain classical 2-ASK modulation, and the special case as B=0 is exactly typical on-off keying (OOK:On-Off Keying) modulation;

3) if ω _c0=ω _c1=ω, B=A and θ=π, we obtain classical 2-PSK (or BPSK) modulation.

We these classical binary modulation of being familiar be all symmetrical, namely the modulating range of code element " 0 " and " 1 " is T.But, in order to improve the availability of frequency spectrum as much as possible, namely in per unit band, transmit higher code check (taking bps/Hz as dimension), also wish more effectively to utilize emitted energy simultaneously, obtain better laser propagation effect, we make " 0 " and the modulating range of " 1 " not etc., namely in (1) formula, 0 < τ < T is fixed, so obtain large class " asymmetric binary shifted keying " (ABSK:Asymmetry Binary Shift Keying) modulation:

$\begin{array}{cc}{s}_0\left(t\right)=A\sin {\mathrm{\&omega;}}_{c0}t,& 0\&le;t<T\\ s_1\left(t\right)=\left\{\begin{array}{c}B\sin ({\mathrm{\&omega;}}_{c1}t+\mathrm{\&theta;}),\\ A\sin {\mathrm{\&omega;}}_{c0}t,\end{array}\right.& \begin{array}{c}0\&le;t<\mathrm{\&tau;},0\&le;\mathrm{\&theta;}\&le;\mathrm{\&pi;}\\ 0\&le;\mathrm{\&tau;}\&le;t<T\end{array}\end{array}---\left(3\right)$

(3) difference of formula and (2) formula, just be that code element " 0 " is different with the modulating range of " 1 ", the former is T, latter is τ < T, make thus more concentration of energy on carrier wave (reason is shown in " and Wu Lenan: the high-speed communication of ultra-narrow band be in progress. natural science be in progress; 17 (11), 2007,1467-1473 ").

2, EBPSK communication system

The broadband wireless services demand of rapid growth proposes more and more higher requirement to radio communication, directly results in aerial radio frequency more and more crowded, particularly along with the development of the third generation (3G) and forth generation (4G) wide-band mobile communication network, almost depleted compared with the continuous frequency spectrum of low-frequency range.10MHz frequency spectrum 20 years usufructuary auction prices in Europe up to 4,000,000,000 Euros, and in China, spend also to be hard to buy best frequency and bandwidth.Therefore, the same with the energy and water resources, frequency spectrum is also the grand strategy resource of country, and compress wireless transmission frequency spectrum to greatest extent and have important practical significance and direct economic benefit, the availability of frequency spectrum has become core competitive power index and the key common technology of generation information transmission system.Therefore ω is made in (3) formula _c0=ω _c1=ω _c, to get rid of the not high 2-FSK class modulation of the availability of frequency spectrum, just obtain ABSK and modulate binary phase shift keying (EBPSK:Extended BPSK) modulation that namely a most important subset expand:

$\begin{array}{cc}{s}_0\left(t\right)=A\sin {\mathrm{\&omega;}}_{c}t,& 0\&le;t<T\\ s_1\left(t\right)=\left\{\begin{array}{c}B\sin ({\mathrm{\&omega;}}_{c}t+\mathrm{\&theta;}),\\ A\sin {\mathrm{\&omega;}}_{c}t,\end{array}\right.& \begin{array}{c}0\&le;t<\mathrm{\&tau;},0\&le;\mathrm{\&theta;}\&le;\mathrm{\&pi;}\\ 0<\mathrm{\&tau;}\&le;t<T\end{array}\end{array}---\left(4\right)$

As seen from Figure 1, the main energetic of EBPSK modulation concentrates near carrier frequency, and thus its smaller bandwidth, the availability of frequency spectrum is very high.But then, the different wave shape of " 0 " and " 1 " of EBPSK modulation is very little, brings very large challenge to demodulation.

3, digital shock filter

For " 0 ", asymmetric modulation that " 1 " different wave shape is very little, the classical matched filter for symmetric modulation waveform and related detecting method are no longer best.In order to improve the demodulation performance for EBPSK modulation signal, we had once invented infinite impulse response (IIR) the digital shock filter of a class simple zero-multipole point (see " for strengthening the impact filtering method of asymmetric binary modulating signal ", patent of invention publication number: 200910029875.3), by trap-selecting frequency characteristic that its bandpass center is precipitous, the modulates information (such as phase hit) of the ABSK modulation signal taking EBPSK as representative is changed into obvious and strong parasitic amplitude modulation to impact, output signal-to-noise ratio is significantly improved, even (signal to noise ratio snr < 0) modulation intelligence of signal can be highlighted with the form of overshoot when signal is flooded completely by noise, therefore be referred to as digital shock filter (Impacting Filters) or EBPSK signal digital booster.

Such as, Fig. 2 (a) and Fig. 2 (b) provides overall amplitude-frequency characteristic and the local amplitude-frequency-phase-frequency characteristic broadening figure of simple zero-3 limit numeral shock filter respectively.From Fig. 2 (a), this filter has extremely narrow trap-selecting frequency characteristic near the signal(-) carrier frequency of 465kHz, and in the whole frequency band of 0-2.4MHz, present obvious overall bandpass characteristics (about 17dB, the 800kHz place that such as decays at 350kHz place decay about 55dB); And from Fig. 2 (b), the carrier frequency of signal is about in the valley point of filter amplitude-versus-frequency curve (solid line in Fig. 2 (b)) and the centre of peak value.Fig. 2 (c) is then the implementation result of this filter, and wherein the 1st road waveform is original EBPSK modulation signal, time domain waveform when the 2nd road and the 3rd road are respectively SNR=5dB after primary signal plus noise and by the output waveform after this filter.This filter visible not only filter effect is obvious, and the amplitude that input signal phase hit can be converted to amplify about 80 times output surge waveform (about from ± 2 rising to ± 160, " 1 " corresponding in figure).

But, the sample frequency f of analog to digital converter (ADC) _sdemodulation performance impact for EBPSK is very large.Such as, turn left from the right side in Fig. 3 and give the ber curve of EBPSK receiver under 40MHz, 60MHz, 80MHz and 100MHz sample frequency successively, visible ADC sample frequency is very large for the impact of EBPSK demodulation performance, and " over-sampling " gain is very high to a certain extent.Such as, for P _ethe error rate of=0.001, receiver 60MHz sampling about can save 6dB than signal to noise ratio of sampling with 40MHz, and 80MHz sampling saves 2dB than 60MHz sampling, the distribution of this overall objective for EBPSK communication system and hardware implementing, can not be ignored.Thus high performance EBPSK receiver requires that ADC has higher over-sampling rate (such as f _s> 10f _c, f _cfor signal(-) carrier frequency), this is the least favorable factor affecting the application of this system.Such as when the satellite receiver intermediate frequency of 70MHz realizes, 10 times of over-samplings also need the ADC of 700MHz, therefore need the high-speed ADC configuring 1GHz during engineering construction, although not high for ADC bit requirements, but be high price, high power consuming devices after all, be unfavorable for promoting with integrated.。

Summary of the invention

For overcoming the deficiencies in the prior art, the object of the present invention is to provide a kind of asymmetric binary modulation signal receiver, to improve the receiving demodulation performance of EBPSK modulation signal, and reducing cost and the complexity of communication system hardware realization.

Because digital shock filter is much more late than the time of occurrence of simulation shock filter, therefore current textbook all only introduces how to design digital shock filter from existing simulation shock filter, present the present invention but needs contrary process, converts simulation shock filter to by digital shock filter.Because digital shock filter designs according to the thinking of adjustment z-plane zero, pole distribution, therefore its transfer function can be write as following rational function form:

$H\left(z\right)=\frac{\underset{i=0}{\overset{I}{\mathrm{\&Sigma;}}}{a}_i\&CenterDot;z^{-i}}{1-\underset{j=1}{\overset{J}{\mathrm{\&Sigma;}}}{b}_j\&CenterDot;z^{-j}}=a_0\&CenterDot;\underset{k=1}{\overset{M}{\mathrm{\&Pi;}}}\frac{1+a_{1k}\&CenterDot;z^{-i}}{1+b_{1k}\&CenterDot;z^{-i}}---\left(5a\right)$

Because digital shock filter is simple zero-multipole point, be actually a pair conjugation zero point and at least two pairs of conjugate poles, therefore 2=I < 4≤J in (5a) formula, generate partial fraction form can be write as:

$H\left(z\right)=\underset{j=1}{\overset{J}{\mathrm{\&Sigma;}}}\frac{A_j}{1-e^{{\mathrm{\&lambda;}}_{j}T}{z}^{-1}}---\left(5b\right)$

Wherein, T=1/f _sfor the sampling period.

And the identical and transfer function being less than the simulation shock filter of limit zero point of limit number can be write as:

$H_a\left(s\right)=\underset{j=1}{\overset{J}{\mathrm{\&Sigma;}}}\frac{A_j}{s-{\mathrm{\&lambda;}}_j}---\left(6\right)$

According to the principle of Impulse invariance procedure, compare (5b) and (6) formula, the limit P of visible simulation shock filter _swith the limit p of digital shock filter _dbetween have direct corresponding relation:

P _s＝(lnp _d)×f _s(7)

But, the Z at zero point of simulation shock filter _swith the z at zero point of digital shock filter _dbetween but there is not this relation, because the shortcoming of Impulse invariance procedure is exactly because spectral aliasing causes digital shock filter to have certain deviation with the frequency response of simulation shock filter near half sample frequency, thus the filter being only suitable for designing high frequency attenuation class leads to as low pass, band, is not suitable for designing high pass, band stop filter.And corresponding to zero point just at the band stop filter at its respective frequencies place.Experiment also shows the inverse transformation directly by Impulse invariance procedure or Bilinear transformation method, be difficult to the simulation shock filter obtaining there is to digital shock filter similar amplitude-frequency characteristic, this is because shock filter to require in passband zero, the position of limit sufficiently accurately just can reach desirable impact effect, thus above-mentioned conventional method can not directly be applied mechanically.

Through a large amount of Computer Simulations and numerical analysis, the present invention draws the final transformation relation between digital shock filter with simulation shock filter coefficient: directly by zero of digital shock filter, pole data undertakies corresponding by following formula:

$\begin{array}{c}{Z}_s=\left(\ln z_d\right)\&times;f_s=10\&times;\left(\ln z_d\right)\&times;f_c,(z_d\&NotEqual;0)\\ P_s=\left(\ln p_d\right)\&times;f_s=10\&times;\left(\ln p_d\right)\&times;f_c\end{array}---\left(8\right)$

The molecule A of each fraction in coefficient (6) formula can be obtained thus _j, and then obtain simulation shock filter H _as coefficient that () is corresponding and whole zero, limit.Although complete the conversion from digital shock filter to simulation shock filter thus, but performance may not best (mainly the last simulation shock filter kind for physics realization and material can have multiple choices, especially its Q value possibility difference is larger), now only need prescribed poles constant, zero frequency is finely tuned.Owing to only having a zero point, thus workload is little, and is the batch production of simulation shock filter in the future, leaves the means of fine setting, because from principle, the consistency of simulation shock filter product can not exceed digital shock filter.

After obtaining the design parameter of simulation shock filter thus, can according to the working frequency range of EBPSK receiver used, customization is applicable to the simulation shock filter of corresponding band, both cavity body filter can be selected, medium (as sapphire) filter, the microwave filter on-line operations such as microstrip transmission line are at microwave frequency band, also quartz-crystal filter is utilized, piezoelectric ceramic filter, Surface Acoustic Wave Filter, mechanical filter, even MEMS (micro electro mechanical system) (MEMS:Micro-Electro-Mechanical Systems) filter etc. realize (as long as having high quality factor and high q-factor) at lower carrier frequency or intermediate frequency.Be reflected in EBPSK receiver, the direct exactly simulation shock filter with customizing according to design parameter of the present invention, replace original ADC and digital shock filter, and judgement and the basic demodulation process such as bit synchronization constant, remain and the analog signal envelope of shock filter output carried out.But now directly can carry out sampling thresholding compare and adjudicate demodulation and without the need to ADC, and sample rate can from f _s>=10f _cbe down to f _s=f _c, be even down to f _s=r _b(after bit synchronization is set up, r _bbit rate for information rate and signal transmission).

The present invention has following beneficial effect:

1) cheap and simple is realized.In EBPSK receiver, ADC and digital shock filter is instead of with simulation shock filter, thus make sample frequency be down to signal carrier frequency even transmission code rate (being generally 1/tens of signal carrier frequency) from the integral multiple (such as 10 times) of carrier frequency, significantly reduce the hard-wired cost of receiver, complexity and power consumption.

2) performance is obviously improved.Simulation shock filter is equivalent to digital shock filter when sample rate is tending towards infinity, theoretical and experiment all shows the demodulation performance that significantly can improve receiver, and the operating frequency of effective elevator system and transmission code rate (because eliminating the process bottleneck of ADC to EBPSK receiver).

3) design easy.Give with carrier frequency the coefficient expressions of the simulation shock filter being parameter.As can be seen from (8) formula, as long as determine carrier frequency just can be simulated shock filter accordingly by digital shock filter.

Above-mentioned explanation is only the general introduction of technical solution of the present invention, in order to better understand technological means of the present invention, and can be implemented according to the content of specification, coordinates accompanying drawing to be described in detail as follows below with preferred embodiment of the present invention.The specific embodiment of the present invention is provided in detail by following examples and accompanying drawing thereof.

Accompanying drawing explanation

Fig. 1 is the power spectrum measured value of the EBPSK modulation in 430MHz tranmitting frequency.Be modulated on 30MHz carrier frequency and carry out, then upconvert to 430MHz, in (4) formula, get θ=π, B=A and modulation duty cycle K:N=2:300, thus code check is 100kbps.

Fig. 2 (a) and Fig. 2 (b) provides overall amplitude-frequency characteristic and the local amplitude-frequency-phase-frequency characteristic broadening figure of simple zero-3 limit digital filter respectively; Fig. 2 (c) is then the implementation result of this filter, wherein from top to bottom: the 1st road waveform is original EBPSK modulation signal; 2nd road and the 4th road are the waveform after primary signal adds noise, and signal to noise ratio is respectively SNR=5dB and SNR=0dB; 3rd road and the 5th road waveform are then respectively this twice plus noise signal by the output waveform after this filter.

Fig. 3 is the demodulation performance comparison diagram of EBPSK digital receiver under different ADC sample frequencys.

Fig. 4 is the asymmetric binary modulation signal receiver block diagram of one embodiment of the invention.

Fig. 5 (a) is the amplitude-frequency response of the simulation shock filter of one embodiment of the invention; Fig. 5 (b) is then the amplitude-frequency response of corresponding prototype numeral shock filter.

Fig. 6 (a) is the Simulink simplified model of the whole EBPSK modulator-demodulator of one embodiment of the invention; Fig. 6 (b) detects judgement and the Simulink module corresponding to bit synchronization, corresponding to the EBPSK_Demodulator in Fig. 6 (a) in EBPSK receiver;

Fig. 7 (a) is the EBPSK modulation signal waveform after the partial enlargement of one embodiment of the invention; When Fig. 7 (b) is the noiseless of one embodiment of the invention, EBPSK modulation signal is by the impact filtering output waveform (partial enlargement) after simulation shock filter, and in figure, ordinate is amplitude, and abscissa is the time; Fig. 7 (c) be one embodiment of the invention low-pass filtering is carried out to impact filtering output waveform absolute value after the impact envelope waveform that obtains, in figure, ordinate is amplitude, and abscissa is the time; Fig. 7 (d) be one embodiment of the invention to impacting the NRZ obtained after envelope shaping, in figure, ordinate is amplitude, and abscissa is the time.

Embodiment

Below with reference to the accompanying drawings and in conjunction with the embodiments, describe the present invention in detail.

A kind of asymmetric binary modulation signal receiver, its Received signal strength be according to (3) formula produce and the ABSK modulation signal launched, particularly according to its most important subset i.e. conventional special case (4) formula produce and the EBPSK modulation signal launched.

Asymmetric binary modulation signal receiver of the present invention mainly comprises low-converter and EBPSK demodulator, wherein low-converter be classical communication receiver commonly use, object is by EBPSK frequency modulating signal from higher rf conversion to lower intermediate frequency, so that demodulator processes.Described EBPSK demodulator comprises a simulation shock filter, carrys out the phase modulation information of outstanding Received signal strength, promotes demodulation performance and save expensive high-speed ADC; Described simulation shock filter, is characterized in that:

1) design parameter of described simulation shock filter is converted by the design parameter of the corresponding digital shock filter as prototype.

2) the prototype numeral shock filter of described simulation shock filter, it is iir digital filter, be made up of a pair conjugation zero point and at least two pairs of conjugate poles, signal carrier frequency is higher than zero frequency but lower than all pole frequencies, and the close degree of zero frequency and pole frequency, at least to reach 10 of signal carrier frequency ^-3magnitude.

3) Z at zero point of described simulation shock filter transfer function _swith limit P _s, first by the z at zero point of prototype numeral shock filter transfer function _dwith limit p _d, try to achieve according to (8) formula, that is:

$\begin{array}{c}{Z}_s=\left(\ln z_d\right)\&times;f_s=10\&times;\left(\ln z_d\right)\&times;f_c,(z_d\&NotEqual;0)\\ P_s=\left(\ln p_d\right)\&times;f_s=10\&times;\left(\ln p_d\right)\&times;f_c\end{array}---\left(8\right)$

Again at limit P _swhen fixing, to Z at zero point _sfinely tune.

4) Physical realization of described simulation shock filter is not limit, but the Q value of this filter in designed resonance frequency is higher, and the performance of impact filtering is better.Both the microwave filter on-line operations such as cavity body filter, dielectric filter, microstrip transmission line can have been selected thus at microwave frequency band, also utilized quartz-crystal filter, piezoelectric ceramic filter, Surface Acoustic Wave Filter, mechanical filter, even MEMS filter etc. to realize at lower carrier frequency or intermediate frequency.

Further, shown in Figure 4, asymmetric binary modulation signal receiver of the present invention comprises: one for receiving the antenna 1 of EBPSK modulation signal, described antenna 1 connects a pre-amplifying module 2, described pre-amplifying module 2 connects a frequency mixer 3, described frequency mixer 3 connects an intermediate frequency amplification module 4, described intermediate frequency amplification module 4 connects a phase discriminator 5, described phase discriminator 5 connects a low-pass filtering mode block 6, described low-pass filtering mode block 6 connects one for generation of the voltage controlled oscillator 7 of local oscillation signal, described voltage controlled oscillator 7 connects described frequency mixer 3, a low-converter is formed with this, described intermediate frequency amplification module 4 also connects a simulation shock filter 8, and described simulation shock filter 8 connects a detection judging module 9, and described detection judging module 9 connects one for the bit synchronization module 10 of output information sequence, also comprise a 1GHz with reference to crystal oscillator 11 and a clock generator 12, described 1GHz connects described phase discriminator 5 and described clock generator 12 respectively with reference to crystal oscillator 11, and described clock generator 12 connects described detection judging module 9 and described bit synchronization module 10 respectively.

Further shown in Figure 4, the operation principle that a kind of asymmetric binary modulating signal of the present invention receives is as follows:

1, the asymmetric binary modulating signal received from antenna is after enlarge leadingly, to be multiplied by frequency mixer 3 with the local oscillation signal carrying out voltage controlled oscillator 7 and to carry out down-conversion, be divided into two-way to export after obtaining 1GHz intermediate frequency: a road is directly supplied to asymmetric binary modulating signal demodulator after intermediate frequency amplifies; Another road is given phase discriminator 5 and is carried out phase compare with the signal of 1GHz reference crystal oscillator, its error signal controls the frequency of voltage controlled oscillator after low-pass filtering 6, and the 1GHz intermediate frequency finally making down-conversion obtain strictly is locked in 1GHz with reference on crystal oscillator frequency, thus this 1GHz is used as system clock with reference to crystal oscillation signal after clock generator 12 shaping, can directly for the detection judging module of asymmetric binary modulating signal demodulator and bit synchronization module provide sampling synchronization clock.

2, for the asymmetric binary modulation analog if signal giving demodulator:

1) simulation shock filter 8 is utilized to carry out signal to noise ratio enhancing and give after phase hit is converted to parasitic amplitude modulation detecting judging module 9;

2) by described detection judging module 9 take out EBPSK modulation signal simulation impact filtering output signal envelope (by take absolute value and low-pass filtering two steps come);

3) directly adjudicate " 0 " and " 1 ", without the need to being transformed into Base-Band Processing again at the synchronization point of system clock.Now both can adopt the simplest threshold judgement, integration also can be adopted to adjudicate;

4) judgement exports after the bit synchronization adjustment of bit synchronization module 10 is (under the control at system clock, and under certain error rate) recovering transmission information sequence, described bit synchronization module is then basic link and the mature technology of digital communication receiver.

It is worthy of note: if adopt classical threshold judgement (or first adjudicate accumulate again), then ADC also can save, namely the direct analog signal that exported by shock filter in sampling instant and a threshold value presetting (or self adaptation calculates) compare, exceed this threshold value to be judged to " 1 ", otherwise be judged to " 0 ", and this process is exactly the standard feature of an analog comparator.

3, simulation shock filter of the present invention

Shock filter is a kind of narrow-band digital band pass filter of special infinite impulse response (IIR), the a pair conjugation zero point very close by resonance frequency and at least two pairs of conjugate poles (in the present embodiment, a pair conjugation zero point or limit being all called a zero point or limit) are formed, an extremely narrow trap-selecting frequency characteristic is presented, as shown in Fig. 2 (a) in its passband.Trap characteristic depends on zero point, and its trap effect can perform to ultimate attainment by the unit circle being taken at z-plane this zero point; Selecting frequency characteristic depends on the comprehensive function of all limits, due to limit is taken on unit circle can be unstable, therefore in order to obtain more sharp-pointed frequency-selecting effect, the frequency of all limits should be selected all very near even overlapping, unimodal with what formed on amplitude-versus-frequency curve.Require the frequency at zero point near and lower than the frequency of all limits, the carrier frequency of signal, then between the zero frequency and pole frequency of filter, is approximately in the centre of filter amplitude-versus-frequency curve (solid line in Fig. 2 (b)) valley point and peak value.Fig. 2 (b) upper valley dot frequency (i.e. zero frequency) and the close degree of crest frequency (be pole frequency for multiple pole, non-multiple pole be then approximately to the frequency of its vector), at least will reach 10 of signal carrier frequency ^-3magnitude.

Realize Fig. 2 (a), in (b) and (c) the digital shock filter of effect be one according to simple zero-3 limit numeral shock filter designed by mentioned above principle, have 1 pair of conjugation zero point and 3 pairs of conjugate poles, its particular location is:

z＝[0 0 0 0 e ^0.62832ie ^-0.62832i]

p＝[0.99995e ^0.62855i0.99995e ^-0.62855i0.92e ^0.62855i0.92e ^-0.62855i0.90e ^0.62855i0.90e ^-0.62855i]

The zeros and poles that (8) formula of substitution can obtain corresponding simulation shock filter is as follows:

Z _s＝10×(lnz)×f _c＝[6.282i -6.282i]×f _c

P _s＝10×(lnp)×f _c

＝[-0.0005000125004166272+6.2855i -0.0005000125004166272-6.2855i

-0.833816089390510+6.2855i -0.833816089390510-6.2855i

-1.053605156578263+6.2855i -1.053605156578263-6.2855i]×f _c

The transfer function shape of this filter is:

$H_a\left(s\right)=k\&CenterDot;\frac{(s-{\mathrm{zs}}_1)(s-{\mathrm{zs}}_2)(s-{\mathrm{zs}}_3)(s-{\mathrm{zs}}_4)(s-{\mathrm{zs}}_5)(s-{\mathrm{zs}}_6)}{(s-{\mathrm{ps}}_1)(s-{\mathrm{ps}}_2)(s-{\mathrm{ps}}_3)(s-{\mathrm{ps}}_4)(s-{\mathrm{ps}}_5)(s-{\mathrm{ps}}_6)}$

But now filter gain k is with f _cdifference and change.Therefore by as follows for zero data correction, gain k=1 can be made:

Z _s＝10×(lnz)×f _c＝[6.282i -6.282i 10 10 10 10]×f _c

At this with f _c=40GHz is example, by f _c=4 × 10 ¹⁰after substituting into above-mentioned zero pole point expression formula, obtain the amplitude-frequency response of designed simulation shock filter as shown in Fig. 5 (a), Fig. 5 (b) is the amplitude-frequency response of corresponding digital shock filter, both visible, there is very high similitude, especially zero in passband, pole location are completely corresponding, this guarantees when to zero, limit require high, the effect of anticipation can be reached according to the simulation shock filter designed by the present invention.

4, performance simulation

According to existing EBPSK communication system digital model, build the Simulink simulation model that receiver adopts the EBPSK modulator-demodulator of simulation shock filter, as shown in Fig. 6 (a).Specifically describe as follows:

1) need not any chnnel coding, under 40GHz carrier frequency, additive white Gaussian noise (AWGN) channel, the EBPSK modulation that (4) formula of selection defines, gets θ=π, f _c=ω _c/ 2 π=40GHz, τ=2 × 1/f _c=5 × 10 ^-11s, T=5 × 10 ^-10s (namely code check is 2Gbps) exemplarily (emulated data is added up by 100,000 code elements).

2) contrast Fig. 4 and build Simulink model as Fig. 6 (b), corresponding to the ebpsk_Demodulator module in Fig. 6 (a).Obviously, the effect of simulation shock filter of the present invention is identical with digital shock filter, namely carries out signal to noise ratio enhancing to modulation signal and its phase hit is converted to parasitic amplitude modulation.

3) because Simulink can not realize real analogue system emulation, therefore we limit maximum step-length (Max step size) used in simulation process.Because this analogue system is based on analog signal, there is not Sampling, is therefore rational by the simulation result limiting maximum step-length to obtain close to actual conditions.And use the Simulink model of digital filter can not to limit maximum step-length, because the sample rate of ADC is determined, if limit maximum step-length, ADC is by nonsensical.

4) above-mentioned impact filtering output waveform is first taken absolute value, then obtain its impact envelope output, as shown in Fig. 7 (c) by low pass filter.

5), after obtaining impacting envelope, signal enters EBPSK_Demodulator module, refers to Fig. 6 (b).Utilize self-synchronizing method to extract bit synchronization information from the impact envelope of EBPSK Received signal strength at this, bit synchronization is carried out to the demodulation court verdict of EBPSK signal.

6) error rate under different maximum simulation step length is obtained, in table 1.Can find out, although Simulink analogue system can not realize simulated completely, by the maximum step-length of restriction, still can reflect that simulation shock filter of the present invention reaches expection object, improve bit error rate performance.

Table 1 reduces EBPSK demodulation performance change during simulation step length approaching simulation shock filter

The foregoing is only the preferred embodiments of the present invention, be not limited to the present invention, for a person skilled in the art, the present invention can have various modifications and variations.Within the spirit and principles in the present invention all, any amendment done, equivalent replacement, improvement etc., all should be included within protection scope of the present invention.

Claims (2)

1. an asymmetric binary modulation signal receiver, comprise one for will expansion bpsk modulated signal frequency from higher rf conversion to the low-converter of lower intermediate frequency and one expansion binary phase shift keying demodulator, it is characterized in that: the binary phase shift keying demodulator of described expansion comprises a simulation shock filter:

1) described simulation shock filter is band pass filter, is made up of a pair conjugation zero point and at least two pairs of conjugate poles, signal carrier frequency f during work _chigher than zero frequency but lower than all pole frequencies, and the close degree of zero frequency and pole frequency, at least to reach 10 of signal carrier frequency ^-3magnitude, thus in passband, present again an extremely narrow trap-selecting frequency characteristic;

2) Z at zero point of described simulation shock filter transfer function _swith limit P _s, the z at zero point of filter transfer function can be impacted by the corresponding numeral as prototype _dwith limit p _dand the signal carrier frequency f of working frequency range _c, directly convert according to (1) formula, that is:

Z _s＝10×(lnz _d)×f _c，(z _d≠0)

(1)P _s＝10×(lnp _d)×f _c

3) described simulation shock filter is for the output response of the bpsk modulated signal of expansion, its modulates information is changed into obvious and strong parasitic amplitude modulation to impact, the binary phase shift keying modulation of described expansion as (2) formula define:

$\begin{array}{c}{s}_0\left(t\right)=A\sin {\mathrm{\&omega;}}_{c}t,0\&le;t<T\\ s_1\left(t\right)=\left\{\begin{array}{c}B\sin ({\mathrm{\&omega;}}_{c}t+\mathrm{\&theta;}),0\&le;t<\mathrm{\&tau;},0<\mathrm{\&theta;}\&le;\mathrm{\&pi;}---\left(2\right)\\ A\sin {\mathrm{\&omega;}}_{c}t,0<\mathrm{\&tau;}\&le;t<T\end{array}\right.\end{array}$

In formula, s ₀(t) and s ₁t () represents the modulation waveform of code element " 0 " and " 1 " respectively; ω _c=2 π f _cfor carrier angular frequencies, θ is " 1 " phase place at keying period τ intercarrier keying, and B-A is the amplitude of carrier wave keying, and T is code-element period;

4) described simulation shock filter allows to adopt different Physical realization, when there is deviation because of material, technique, then and only need at limit P _swhen fixing, to only Z at zero point _sfinely tune.

2. asymmetric binary modulation signal receiver according to claim 1, it is characterized in that: described asymmetric binary modulation signal receiver comprises one for receiving the antenna (1) of EBPSK modulation signal, described antenna (1) connects a pre-amplifying module (2), described pre-amplifying module (2) connects a frequency mixer (3), described frequency mixer (3) connects an intermediate frequency amplification module (4), described intermediate frequency amplification module (4) connects a phase discriminator (5), described phase discriminator (5) connects a low-pass filtering mode block (6), described low-pass filtering mode block (6) connection one is for generation of the voltage controlled oscillator (7) of local oscillation signal, described voltage controlled oscillator (7) connects described frequency mixer (3), described intermediate frequency amplification module (4) also connects described simulation shock filter (8), described simulation shock filter (8) connects one and detects judging module (9), and described detection judging module (9) connection one is for the bit synchronization module (10) of output information sequence, one clock generator (12) connects described detection judging module (9) and described bit synchronization module (10) respectively, one 1GHz connects described phase discriminator (5) and described clock generator (12) respectively with reference to crystal oscillator (11).