CN103701492B - The underwater acoustic array method of linear FM signal modulation /demodulation - Google Patents

Abstract

The underwater acoustic array method of linear FM signal modulation /demodulation, relates to underwater acoustic array.Raw information is carried out the digital signal after message sink coding must compress, then carries out chnnel coding, using the data after chnnel coding as linear FM signal control code, control to generate linear FM signal by frequency hop sequences and be modulated obtaining modulated signal as carrier wave；To modulated signal through D/A conversion and power amplification, and eventually through transmitting transducer, signal after modulation is converted into sound wave and propagates in ocean water Acoustic channel；By receiving transducer, the acoustical signal propagated in ocean water Acoustic channel reception is converted into the signal of telecommunication, after preposition amplification and bandpass filtering, obtains analogue signal；The analogue signal received is changed through A/D；After the digital signal detection after A/D conversion to synchronization, carry out linear FM signal solution mid frequency and frequency, demodulation frequency, obtain demodulating data；Demodulating data is made channel-decoding；Gained digital signal is carried out source coding, obtains information.

Description

The underwater acoustic array method of linear FM signal modulation /demodulation

Technical field

The present invention relates to underwater acoustic array, particularly relate to a kind of employing reversion and be multiplied solution linear FM signal center frequently Rate method, refinement discrete fractional Brownian random field (Zoom-DFRFT) solve the linear FM signal of linear FM signal frequency modulation rate and adjust The underwater acoustic array method of system demodulation.

Background technology

In traditional underwater acoustic array, often utilize the frequency of carrier wave to the information of carrying.Compared to MPSK etc. other Modulation system, MFSK modulation band utilization rate is more low, and under same band, its rate of information throughput is the most backward, but it Steadily and surely, reliable performance makes it occupy important one seat in underwater sound communication.Therefore, More Secure Underwater Acoustic Frequency-Hopping Communication is improved Band efficiency is to ensureing that underwater information is stable, high efficiency of transmission is significant.

The conventional method improving underwater acoustic array band efficiency is that multi-carrier modulation (MCM) technology is multiple with frequency division Combining by (FDM) technology, the serial data stream of script high-speed transfer is i.e. passed parallel by its core concept by multiple subcarriers Defeated but the method requires that system has bigger bandwidth.

Beaujean of U.S., Atlantic university in 2009 et al. devises frequency hopping frequency division multiplexing (FH-FDM) system.This is Unite depth of water 3m, depth of receiver 1m, horizontal range 120m simulated environment under, traffic rate can reach 5272bps, and the bit error rate reaches To less than 0.14% (sound source level 170dB re1 μ Pa at1m, reverberation 9.31ms, doppler spread 1.5Hz), but needed for this system Bandwidth be up to 27kHz([1] Beaujean P P, Pajovic M, Carlson E, et al.Frequency-hopped frequency division multiplexed signaling for underwater acoustic communications between60and90kHz in ports and very shallow waters[C]// OCEANS2009,MTS/IEEE Biloxi-Marine Technology for Our Future:Global and Local Challenges.IEEE, 2009:1-7.).

Sun little Dong et al. devises a kind of OFDM (OFDM) frequency hopping underwater sound communication system.This system uses difference Technology is demodulated, and eliminates and inserts pilot tone and carry out the expense of channel estimation, but its time domain differential ference spiral and Frequency domain differential demodulation The performance loss of 2～about 3dB is had than coherent demodulation.([2] Sun little Dong, Hou Chaohuan. a kind of differential ference spiral OFDM frequency hopping underwater sound Communication system [J]. electroacoustic techniques, 2007,31 (10): 71-74.)

Another kind of method is to be modulated as carrier wave by linear FM signal.Owing to linear FM signal possesses mid frequency Fc and two parameters of frequency modulation rate μ, under same symbol length with bandwidth condition, relative to the carrier wave of single-frequency, linear frequency modulation Signal can carry the information of its twice.Therefore use linear FM signal modulation to replace original MFSK modulation, can be original in guarantee On the basis of hopping scheme anti-multi-path capability so that the rate of information throughput is improved largely.Assume minimum frequency sweep model Enclosing identical with MFSK modulation minimum frequency space, the multiple that the rate of information throughput improves the most in theory is

$k=\frac{{\log }_{2}2(1+2+\·\·\·+S)}{{\log }_{2}S}=\frac{{\log }_{2}S(S+1)}{{\log }_{2}S}=1+{\log }_S(S+1)$

S is the system number of MFSK modulation.When S is bigger, k levels off to 2.

Solve linear FM signal to include solving mid frequency fc and two steps of frequency, demodulation frequency μ.Note receives signal

$r\left(t\right)=A\left(t\right)\sin \left[2\mathrm{\π}(f_{c}t+\frac{1}{2}{\mathrm{\μt}}^2)\right]+n\left(t\right),-\frac{T}{2}\≤t\≤\frac{T}{2},$

A (t) is signal envelope, and n (t) is noise, and T is signal duration, is multiplied with himself after being inverted by r (t),

$\begin{array}{c}x\left(t\right)=r\left(t\right)\·r(-t)\\ =\left\{A\left(t\right)\sin \right[2\mathrm{\π}(f_{c}t+\frac{1}{2}{\mathrm{Mt}}^2)]+n\left(t\right)\}\·\left\{A(-t)\sin \right[2\mathrm{\π}(-f_{c}t+\frac{1}{2}{\mathrm{Mt}}^2)]+n(-t)\}\\ =\frac{1}{2}A\left(t\right)A(-t)[\cos \left(4\mathrm{\π}{f}_{c}t\right)-\cos \left(2\mathrm{\π}{\mathrm{\μt}}^2\right)]+n^{\′}\left(t\right)\end{array}$

Wherein,

$n^{\′}\left(t\right)=n\left(t\right)A(-t)\sin \left[2\mathrm{\π}(-f_{c}t+\frac{1}{2}{\mathrm{\μt}}^2)\right]+n(-t)A\left(t\right)\sin \left[2\mathrm{\π}(f_{c}t+\frac{1}{2}{\mathrm{\μt}}^2)\right]+n\left(t\right)n(-t)$

In x (t) expression formula bracket, centered by Section 2, frequency is 0, frequency modulation rate is the linear FM signal of 2 μ, therefore only needs Meet and select modulated signal low-limit frequency i.e. to can use high pass filter to incite somebody to action equal to 2 times of linear FM signal maximum swept frequency range greatly It filters, and the single-frequency components after filtering is 2 times of required linear FM signal mid frequency.

Frequency, demodulation frequency has used Fourier Transform of Fractional Order (Fractional Fourier Transform, following shorthand For FRFT), it is that classical Fourier transformation one is promoted.FRFT can represent the Time And Frequency characteristic of signal simultaneously, due to Being linear transformation, therefore it avoids the cross term problem of tradition time-frequency distributions.The most important thing is, multi-signal is being divided by FRFT Number all has energy accumulating characteristic on territory, rank, and owing to integral kernel has linear FM characteristic, therefore, this energy accumulating characteristic Best to linear FM signal, select suitable conversion exponent number p, just linear FM signal can be transformed to pulse signal.

Fourier Transform of Fractional Order is a line integral computing, and the fractional number order Fourier after this conversion is designated as u territory, then The p rank Fourier Transform of Fractional Order of signal s (t) is

$\left(F^{p}s\right)\left(u\right)={\∫}_{-\∞}^{+\∞}{K}_p(t,u)s\left(t\right)\mathrm{dt}$

Wherein,

$K_p(t,u)=\left\{\begin{array}{c}\sqrt{\frac{1-j\cot \mathrm{\α}}{2\mathrm{\π}}}\mathrm{exp\; j}(\frac{t^2+u^2}{2}\cot \mathrm{\α}-\frac{\mathrm{tu}}{\sin \mathrm{\α}}),\mathrm{\α}\≠\mathrm{n\π}\\ s\left(t\right),\mathrm{\α}=2\mathrm{n\π}\\ s(-t),\mathrm{\α}=(2n\±1)\mathrm{\π}\end{array}\right.$

It is referred to as the integral kernel function of Fourier Transform of Fractional Order, α=p pi/2.([3] are happy and carefree, Deng Bing, Wang Yue. fractional order Fu In leaf transformation and application [M] thereof. publishing house of Tsing-Hua University, 2009.)

Owing to linear FM signal frequency band range used is limited, it focuses on order FRFT peak value and is also only limitted to certain on u territory One scope, therefore only need to make discrete fractional Brownian random field (DFRFT), that is Zoom-on specific interval DFRFT.Assuming that FRFT to be calculated is at the interval [u of local spectra₁,u₂M point periodic sampling value on], becomes fractional number order Fourier Measure and discrete turn to u=u₀+ m Δ I ,-M/2≤m≤M/2, wherein u₀=(u₂+u₁)/2 represent interval midpoint, Δ I=(u₂-u₁)/ (M-1) represent resolution, then the expression formula of Zoom-DFRFT is:

$X_{p,\mathrm{\&lambda;},P}\left(u\right)=\frac{\sqrt{1-\mathrm{j\&gamma;}}{e}^{\mathrm{j\&pi;\&gamma;}{N}_0{\mathrm{\&lambda;}}^2}}{2\mathrm{\&Delta;x}}{e}^{\mathrm{j\&pi;}\left(\frac{\mathrm{\&gamma;\&lambda;}}{P}\right)m}{e}^{\mathrm{j\&pi;}\left(\frac{\mathrm{\&gamma;}-\mathrm{P\&beta;}}{4P^2{N}_0}\right)m^2}\underset{n=-N_0}{\overset{N_0}{\mathrm{\&Sigma;}}}{e}^{\mathrm{j\&pi;}\left(\frac{\mathrm{\&beta;}}{4P{N}_0}\right){(m-n)}^2}{e}^{\mathrm{j\&pi;}\left(\frac{\mathrm{P\&gamma;}-\mathrm{\&beta;}}{4P{N}_0}\right)n^2}{e}^{-\mathrm{j\&pi;\&beta;\&lambda;m}}x\left(\frac{n}{2\mathrm{\&Delta;x}}\right),0\&le;n<N_0$

Wherein, γ=cot α, β=csc α,N₀Counting for input signal, λ=u₀/ Δ x be " translate because of Son ", represent local spectra center u₀Relative position in whole spectral limit [-Δ x/2, Δ x/2], (2 Δ I Δ x) are P=1/ " zoom factor ", represents that resolution ax I of local spectra is relative to the standard resolution 1/ (amplification of 2 Δ x).Asking in above formula It is discrete convolution form with part, therefore can realize Fast numerical with FFT and calculate.([4] Zhao Xinghao, happy and carefree, Deng Bing etc. The quick New calculating method [J] of Fourier Transform of Fractional Order. electronic letters, vol, 2007,35 (6): 1089-1093.)

When there is Doppler frequency shift f_dTime, peak moves, and causes the sampled value of sampling location fractional order Fourier spectrum Reduce.Definition deviation ratio is σ=2 π | f_d| T, T are Baud Length, then focus on declining of fractional-order fractional order Fourier spectral amplitude ratio Subtracting coefficient is

$\mathrm{\&thetav;}={\left(\sin c\frac{\mathrm{\&gamma;}}{2}\right)}^2$

Summary of the invention

It is an object of the invention to provide a kind of underwater acoustic array method of linear FM signal modulation /demodulation.

The present invention comprises the following steps:

1) raw information is carried out message sink coding, the digital signal after being compressed；

2) digital signal after the compression of step 1) gained is carried out chnnel coding；

3) using step 2) data after gained chnnel coding as linear FM signal control code, control raw by frequency hop sequences Linear FM signal is modulated as carrier wave, obtains modulated signal；

4) modulated signal of step 3) gained is changed and power amplification through D/A, and will eventually through transmitting transducer After modulation, signal is converted into sound wave and propagates in ocean water Acoustic channel；

5) by reception transducer, the acoustical signal propagated in ocean water Acoustic channel reception is converted into the signal of telecommunication, through preposition Amplify and obtain analogue signal after bandpass filtering；

6) analogue signal receiving step 5) is changed through A/D；

7), after to the digital signal detection after the conversion of step 6) A/D to synchronization, linear FM signal solution mid frequency is carried out And frequency, demodulation frequency, obtain demodulating data；

8) demodulating data of step 7) gained is made channel-decoding；

9) digital signal of step 8) gained is carried out source coding, obtain information.

In step 3), 2 times of the low-limit frequency of described modulated signal >=linear FM signal maximum swept frequency range.

In step 7), described solution mid frequency is the signal reversion after synchronizing and is multiplied with himself, filters low frequency Making Fourier transformation after component, the half of calculated rate is mid frequency；The center frequency tried to achieve according to described frequency, demodulation frequency Rate determines possible tune frequency values, and does refinement discrete fractional Brownian random field respectively on corresponding focusing fractional-order, The frequency modulation rate receiving signal is determined by the maximum of modulus value after converting.

Owing to linear FM signal possesses mid frequency fc and two parameters of frequency modulation rate μ, in same symbol length and bandwidth Under the conditions of, relative to the carrier wave of single-frequency, linear FM signal can carry the information of its twice.Therefore employing linear FM signal Modulation replaces original MFSK modulation, on the basis of ensureing original hopping scheme anti-multi-path capability so that the rate of information throughput Improved largely.Compared with existing underwater sound hopping scheme, the present invention has the following advantages:

(1) tradition underwater acoustic array only uses the frequency of carrier wave as carrier transmission information, and transfer rate is relatively low, therefore frequency band Under utilization rate is relatively low, and the present invention utilizes linear FM signal as the carrier of information, the load in each element duration Ripple has mid frequency and two parameters of frequency modulation rate, under same symbol length with bandwidth condition, relative to the load of single-frequency Ripple, linear FM signal can be carried the information of its twice, be substantially increased the band efficiency of frequency-hopping system.

(2) linear FM signal modulation has the feature such as anti-multipath jamming, anti-channel fading, be suitably applied in the most by force, In the complicated underwater sound communication channel that the time space frequency rising and falling big becomes.

(3) the method for the invention calculates process simply, will not increase too much amount of calculation, can guarantee that the existing underwater sound simultaneously The anti-multi-path capability of hopping scheme.

Accompanying drawing explanation

Fig. 1 is the schematic block diagram of the embodiment of the present invention.

Fig. 2 is the upper sweep linearity FM signal modulation pattern in the modulation of 4FH-LFM shown in Fig. 1.

Fig. 3 is the lower sweep linearity FM signal modulation pattern in the modulation of 4FH-LFM shown in Fig. 1.

Fig. 4 is the time-frequency figure of certain section of modulated signal of the modulation output of 4FH-LFM shown in Fig. 1.

Fig. 5 is the flow chart of frequency, demodulation frequency in demodulation shown in Fig. 1.For solving the mid frequency of linear FM signal, f₀ For the low-limit frequency in each frequency hopping band limits, Δ f_LFM=300Hz, for minimum swept frequency scope, [] expression takes nearby Whole, c₀Represent that the mid frequency estimated is at whole frequency hopping group [f₀,f₀+8Δf_LFMRelative position in],Represent and take downwards Whole, rindex{ } represent and take off mark, rumax{ represent and take the u axial coordinate that spectral peak is corresponding, TABLE [i] [j] is demodulation letter Breath inquiry table.

Fig. 6 be frequency hopping-linear frequency modulation (4FH-LFM) system built with Fig. 1 under Gaussian white noise channel with frequency hopping-many The simulation performance of carrier wave-frequency shift keying (FH-MC-MFSK) system compares.

Fig. 7 is that frequency hopping-linear frequency modulation (4FH-LFM) system with Fig. 1 structure is under the superposition Gaussian white noise channels of many ways Compare with the simulation performance of frequency hopping-multicarrier-frequency shift keying (FH-MC-MFSK) system.

Detailed description of the invention

Below in conjunction with the accompanying drawings invention is described in further detail.

Fig. 1 shows the theory diagram of the embodiment of the present invention.First original source is obtained by communication transmitting terminal as message sink coding Digital signal after compression, carries out chnnel coding to the digital signal after compression, and coding can use convolutional code, Turbo code etc. Encode.

4FH-LFM modulator approach is used to be modulated the digital signal after chnnel coding, the linear tune of each frequency hopping group Frequently signal modulation pattern is shown in Fig. 2, Fig. 3, and the decimal scale information correspondence represented by each linear FM signal marks below pattern. Systematic sampling rate fs is set to 96kHz, and frequency range is 10～20kHz, and frequency hopping number is 4, and frequency hopping pattern is incremented by successively for order, Symbol width is 10ms, and traffic rate is 600bps.Binary data after chnnel coding is grouped as linear frequency modulation by 6 Signal control code, searches modulation pattern and presses the linear FM signal that frequency hop sequences generation is corresponding.

For example, it is assumed that the digital signal after chnnel coding be 010101,010110,111001,100100}, corresponding ten Binary information is that { 21,22,57,36}, lookup Fig. 2, Fig. 3 modulation pattern obtains (fc, μ) parameter group sequence of 4 sections of linear FM signals Be classified as (10.30kHz, 60kHz/s), (12.95kHz, 90kHz/s), (16.95kHz ,-30kHz/s), (18.55kHz ,- 21kHz/s) }, its time-frequency figure such as Fig. 4.

Modulated signal can be converted into by digital to analog converter (DAC) plus after synchronous head signal and protection interval framing Analogue signal, the signal after being amplified by transmitting transducer after power amplification is converted to sound wave and passes in ocean water Acoustic channel Broadcast.

At receiving terminal, receive transducer and acoustical signal is converted to the signal of telecommunication and gathers entrance processing system, by front storing It is made bandpass filtering after amplifying in the frequency band range used by senior general's signal, is then changed by analog-digital converter (ADC) Become digital signal.

Detect and synchronize and judge that the start stop bit of synchronous head signal postpones by synchronous head to start demodulation, first solve Mid frequency, will be multiplied with himself after receiving signal r (t) reversion, makees Fourier transformation, calculated rate after filtering low frequency component Half be mid frequency.Try to achieve after mid frequency by the flow process frequency, demodulation frequency of Fig. 5, i.e. first pass through Fig. 2, Fig. 3 and judge at this Mid frequency raises the probable value { μ of frequency_i, (0≤i ＜ N, N≤4), to there being N number of focusing fractional order { p_i, can be to it at this Making n times Zoom-DFRFT in several focusing fractional orders, the refinement interval range of Zoom-DFRFT need to comprise linear FM signal and exist Focus on the peak coordinate point u after fractional order carries out DFRFT₀, this point can basisDetermine, its InThe mid frequency of the linear FM signal for solving, f_sFor sample rate, B is swept frequency range, N₀Count for symbol sample.If The frequency modulation rate of signal is μ_iMAX, (0≤i_MAX＜ N), then have

$\max \left\{\right|\left\{F^{p_{\mathrm{iMAX}}}r\right\}\left(u\right)\left|\right\}>\max \left\{\right|\left\{F^{p_i}r\right\}\left(u\right)\left|\right\},(i\&NotEqual;i_{\mathrm{MAX}},0\&le;i<N)$

The most just can determine that frequency modulation rate μ_iMAX.According to (the c tried to achieve₀,i_MAX) inquiry table 1 to be to determine linear FM signal institute Represent decimal scale information, if μ_iMAX＜ 0 and the decimal scale information that checks in are less than 56, then this decimal scale information must be added 28 and is Last decimal scale information.Corresponding channel-decoding is made with message sink coding the most available with source coding according to channel coding method The stay of two nights.

Demodulating information inquiry is as follows:

$\mathrm{TABLE}164=\left[\begin{array}{cccc}63& 0& 0& 0\\ 21& 0& 0& 0\\ 22& 62& 0& 0\\ 23& 14& 0& 0\\ 24& 15& 61& 0\\ 16& 25& 7& 0\\ 8& 17& 26& 60\\ 0& 9& 18& 27\\ 1& 10& 19& 59\\ 2& 11& 20& 0\\ 3& 12& 58& 0\\ 4& 13& 0& 0\\ 5& 57& 0& 0\\ 6& 0& 0& 0\\ 56& 0& 0& 0\\ 56& 0& 0& 0\end{array}\right]$

Fig. 6 show traffic rate be 600bps, bandwidth 10～20kHz, frequency hopping number be 4, simulated environment be white Gaussian Under the setting of noisy communication channel, frequency hopping-linear frequency modulation (4FH-LFM) system and frequency hopping-multicarrier-frequency shift keying (FH-MC-MFSK) The simulation performance of system compares.By contrast: when signal to noise ratio is more than-9.4dB, the former is better than the latter at performance；And when letter When ratio of making an uproar is less than-9.4dB, the latter's performance is better than the former.

Fig. 7 show traffic rate be 600bps, bandwidth 10～20kHz, frequency hopping number be 4, simulated environment folded for many ways Add under the setting of Gaussian white noise channel, frequency hopping-linear frequency modulation (4FH-LFM) system and frequency hopping-multicarrier-frequency shift keying (FH- MC-MFSK) simulation performance of system compares.By contrast: when signal to noise ratio is higher than-6dB, the former is better than the latter at performance.

The present invention is on the basis of tradition underwater acoustic array, at transmitting terminal, with binary information source as linear frequency modulation Signal control code, and press frequency hop sequences control generation different center frequency, the linear FM signal of different frequency modulation rate；Receiving End, is multiplied with himself after being inverted by reception signal, makees Fourier transformation after filtering low frequency component, and the half of calculated rate is Mid frequency, determines possible tune frequency values according to the mid frequency tried to achieve, and does respectively on corresponding focusing fractional-order Refinement discrete fractional Brownian random field, is determined the frequency modulation rate receiving signal by the maximum of modulus value after converting.Utilize this communication Method, can be on the basis of ensureing hopping scheme anti-multi-path capability so that the rate of information throughput is improved largely.

Claims (4)

1. the underwater acoustic array method of linear FM signal modulation /demodulation, it is characterised in that comprise the following steps:

1) raw information is carried out message sink coding, the digital signal after being compressed；

2) to step 1) digital signal after the compression of gained carries out chnnel coding；

3) using step 2) data after gained chnnel coding as linear FM signal control code, control to generate line by frequency hop sequences Property FM signal is modulated as carrier wave, obtains modulated signal；Described modulation method particularly includes: by two after chnnel coding Binary data as linear FM signal control code by 6 packets, is searched modulation pattern and presses the line that frequency hop sequences generation is corresponding Property FM signal, modulation pattern divides according to (fc, μ) parameter group sequence of linear FM signal, and wherein fc is linear tune Frequently the mid frequency of signal, μ is the frequency modulation rate of linear FM signal；

4) to step 3) modulated signal of gained through D/A conversion and power amplification, and will modulation eventually through transmitting transducer Rear signal is converted into sound wave and propagates in ocean water Acoustic channel；

5) by reception transducer, the acoustical signal propagated in ocean water Acoustic channel reception is converted into the signal of telecommunication, through preposition amplification With obtain analogue signal after bandpass filtering；

6) to step 5) analogue signal that receives changes through A/D；

7) to step 6) A/D conversion after digital signal detection to synchronize after, carry out linear FM signal solution mid frequency reconciliation Frequency modulation rate, obtains demodulating data；

8) to step 7) demodulating data of gained makees channel-decoding；

9) to step 8) digital signal of gained carries out source coding, obtains information.

2. the underwater acoustic array method of linear FM signal modulation /demodulation as claimed in claim 1, it is characterised in that in step 3) in, 2 times of the low-limit frequency of described modulated signal >=linear FM signal maximum swept frequency range.

3. the underwater acoustic array method of linear FM signal modulation /demodulation as claimed in claim 1, it is characterised in that in step 7), in, described solution mid frequency is the signal reversion after synchronizing and is multiplied with himself, makees Fourier after filtering low frequency component Conversion, the half of calculated rate is mid frequency.

4. the underwater acoustic array method of linear FM signal modulation /demodulation as claimed in claim 1, it is characterised in that in step 7), in, the mid frequency tried to achieve according to described frequency, demodulation frequency determines possible tune frequency values, and in corresponding focusing fractional order Do refinement discrete fractional Brownian random field on secondary respectively, the maximum of modulus value after converting determine the frequency modulation rate receiving signal.