CN108183877A - A kind of multi-tone frequency demodulated signal method based on FPGA - Google Patents

Abstract

The invention discloses a kind of multi-tone frequency demodulated signal methods based on FPGA, include the following steps：Multitone frequency-modulated analog letter signal is converted into digital signal；Multi-tone frequency digital medium-frequency signal down coversion is realized in FPGA；It is sampled in FPGA, reduces signal data rate；Doppler shift correction is realized in FPGA；The synchronization of symbol is realized in FPGA；The detection to pilot tone and single-tone is completed by spectrum analysis in FPGA, extracts useful signal, and then realize the demodulation of multitone FM signal.This demodulation method directly acquires amounts of frequency offset using spectrum analysis, has skipped the iteration structure of conventional phaselocked loop, and direct locking frequency offset is simultaneously corrected；The particularity of information according to entrained by multitone modulates FM signal, the Equivalent Base-Band signal got is directly carried out Fast Fourier Transform (FFT) in FPGA can extract modulated signal, need not move through phase extraction, simplify demodulation structure, reduce demodulation difficulty.

Description

A kind of multi-tone frequency demodulated signal method based on FPGA

Technical field

The invention belongs to the technical field of safe observation and control technology, specifically, being related to a kind of multitone frequency modulation based on FPGASignal demodulating method.

Background technology

The most crucial thought of multitone frequency modulation is exactly multitone coding frequency modulation, and NASA proposes 7 sinusoidal carriers of selection, and it is 2 to exportRoad-load wave and signal, so input can be binary data.Since the amplitude with signal changes over time, instantaneous power is not solidFixed, so NASA uses secondary modulation, and signal is again to radio-frequency carrier frequency modulation.Multitone frequency modulation codimg logic unit is by coded instructionsCoding is 11 letters per frame length, wherein 9 letters are address word, in addition 2 are finger as multitone frequency modulation one by oneEnable word.It is synchronous for the ease of letter, it provides to be inserted into one section of gap between letter and letter.Traditional demodulation mode is divided into：It is non-Coherent demodulation and coherent demodulation.

The flow of non-coherent demodulation is：First to the frequency-modulated wave derivation of constant amplitude, the signal of acquisition is made to become amplitude-modulation ＆ frequency-modulationwave,That is AM-FM waves, this process are referred to as frequency discrimination, and frequency discrimination causes the changing rule of modulated signal to reflect onto envelope.Then to AM-FM intoRow square root raised cosine low-pass filtering, obtains its low frequency envelope signal m (t), this process is envelope detection.

Modulated signal and DC component are extracted from wave envelope.Using low-pass filter, DC component in signal is filtered out, mostThe baseband signal before FM modulation is obtained eventually.

The flow of coherent demodulation is:FM modulated signals first pass around bandpass filter, filter out un-necessary frequency component, Ran HouliThe reference signal being concerned with all the way with carrier frequency with multiplier input is multiplied with by the modulated signal of bandpass filter, finally utilizesLow-pass filter filters out high fdrequency component, obtains original signal.

The key for realizing coherent demodulation is that receiving terminal will recover a coherent carrier with modulating carrier wave stringent synchronization, noTo then demodulated signal be distorted.In practical Safety system, the signal received is since the carrier wave of transmitting terminal is unstable or connectsThe local oscillated signal of receiving end is unstable, and the Doppler effect that aircraft high-speed flight generates can all make carrier wave that carrier frequency occurRate deviates.Receiver can not determine the carrier frequency of the modulated signal received, and it is tight can not to generate the local carrier with phase, frequency deviation with frequencyDecoding error can occur during weight, it is impossible to be properly received instruction.

Invention content

For deficiency above-mentioned in the prior art, the present invention provides a kind of multi-tone frequency demodulated signal side based on FPGAMethod, this method is simpler, improves level of integrated system and reliability.

In order to achieve the above object, the solution that uses of the present invention is：A kind of multitone FM signal solution based on FPGATune method, includes the following steps：

S1, multitone frequency-modulated analog letter signal is converted into digital signal；

S2, multi-tone frequency digital medium-frequency signal down coversion is carried out in FPGA；

S3, it is sampled in FPGA, reduces signal data rate；

S4, Doppler shift correction is carried out in FPGA；

S5, independent component is filtered out by low-pass filter in FPGA；

S6, the detection to pilot tone and single-tone is completed by spectrum analysis in FPGA, useful signal is extracted, so as to fulfill moreThe demodulation of sound FM signal.

Further, it in step S1, is sampled by A/D conversion chips AD9626, by the mould that carrier frequency is 70MHzThe modulated signal for intending intermediate frequency multitone frequency modulation is converted to 12bits multitone frequency-modulated digital signals, then inputs in the chip of ALTERRACarry out following digital processing.

Further, in step S2, the mixed frequency signal cos (2 π ft) that frequency is 70MHz is generated by local oscillator,Middle f=70MHz；The intermediate-freuqncy signal that carrier frequency is 70MHz and the signal that local oscillator generates are mixed, realize down coversion.

Further, in step S3, using the integral multiple of the least common multiple in all single-tone periods as a symbolLength, and as the period of this group of multitone combination signal.

Further, in step S4, frequency deviation position is obtained, then completed by way of mixing by Fast Fourier Transform (FFT)Doppler shift is corrected.

Further, the same of carrier wave is completed by the carrier wave essence synchronization combining of the thick synchronization of carrier wave in step S2 and step S4Step, the thick synchronization of step S2 realize that signal is moved from upper frequency to lower frequency in a manner of big stepping.

Further, in step S5, the synchronization of symbol is realized by using the mode of lead-lag in FPGA.

The invention has the advantages that the present invention is directed to multi-tone frequency modulation feature, it is proposed that a kind of completely new based on existingThe digital multisound FM signal demodulation method of field programmable gate array, i.e., using FPGA handle digital signal the characteristics of, processing it is moreTone frequency digital medium-frequency signal, demodulates digital baseband signal, the modules demodulating process of analog circuit is substituted, so as to improveLevel of integrated system and reliability reduce external interference and effectively reduce equipment development cost.

Description of the drawings

Fig. 1 is conventional multitone frequency modulation non-coherent demodulation procedural block diagram.

Fig. 2 is conventional multitone frequency modulation coherent demodulation procedural block diagram.

Fig. 3 is multitone fm demodulation transmitting terminal.

Fig. 4 is multitone fm demodulation receiving terminal.

Fig. 5 is the digital multisound fm demodulation functional block diagram based on FPGA.

Specific embodiment

To make the object, technical solutions and advantages of the present invention clearer, the technical solution in the present invention is carried out belowIt clearly and completely describes, it is clear that described embodiment is part of the embodiment of the present invention, instead of all the embodiments.BaseEmbodiment in the present invention, those of ordinary skill in the art obtained without creative efforts it is all itsHis embodiment, shall fall within the protection scope of the present invention.

Below in conjunction with attached drawing, the invention will be further described：

With reference to attached drawing 1- attached drawings 5, the present invention provides a kind of multi-tone frequency demodulated signal method based on FPGA, including followingStep：

S1, multitone frequency-modulated analog letter signal is converted into digital signal；It is sampled by A/D conversion chips AD9626,The modulated signal for the analog intermediate frequency multitone frequency modulation that carrier frequency is 70MHz is converted into 12bits multitone frequency-modulated digital signals, soAfterwards following digital processing is carried out in the chip of input ALTERRA.

S2, multi-tone frequency digital medium-frequency signal down coversion is carried out in FPGA；It is 70MHz to cross local oscillator and generate frequencyMixed frequency signal cos (2 π ft), wherein f=70MHz；Intermediate-freuqncy signal and local oscillator that carrier frequency is 70MHz are generatedSignal mixing, realize down coversion.

S3, it is sampled in FPGA, reduces signal data rate；By the integer of the least common multiple in all single-tone periodsLength as a symbol again, and as the period of this group of multitone combination signal.Sample frequency and fast FourierPoints codetermined spectral resolution, that is to say and the sampling period is determined, affect signal demodulation when fast Fourier becomeThe spectral leakage degree changed.In order to inhibit spectral leakage, the period in one group of tone signal is fixed, and selection is adoptedSample frequency and the points of Fast Fourier Transform (FFT) need to meet certain relationship, simultaneously because sample frequency and fast Fourier becomeThe points changed directly determine the resolution ratio of frequency spectrum, usually need to consider the size of resolution ratio in Practical Project.By above-mentionedIllustrate understand digital intermediate frequency signal is sampled not only consider Practical Project in processing hardware performance limitation andMore accurately to obtain information entrained in modulated signal.

S4, Doppler shift correction is carried out in FPGA；Since the frequency in the influence and transmission process of physical device is inclinedIt moves so that the centre frequency of the signal by down coversion does not move to zero-frequency, and cadre enrolled among workers is solved caused by Doppler effect to avoidIt disturbs, therefore selects to carry out Doppler shift correction after completing to sample.The precision of Doppler's correction is by FFT spectrum resolution ratio systemAbout.Frequency deviation position is obtained by Fast Fourier Transform (FFT) first, then Doppler shift correction is completed by way of mixing.

S5, independent component is filtered out by low-pass filter in FPGA；According to the single audio frequency selected by multitone FM signalThe cut off band width of low-pass filter is arranged on a rational range by rate range, while effective information is retained as far as possibleFilter out high frequency and image component.

S6, the detection to pilot tone and single-tone is completed by spectrum analysis in FPGA, useful signal is extracted, so as to fulfill moreThe demodulation of sound FM signal；The Equivalent Base-Band signal that have passed through Doppler shift correction can be into after by low-pass filterThe detection of row pilot tone and single-tone, since the frequency of pilot tone is fixed, the frequency of single-tone is also known range, can by calculatingThe points of the Fast Fourier Transform (FFT) result corresponding to pilot tone are obtained, can similarly obtain the Fast Fourier Transform (FFT) corresponding to all single-tonesAs a result points.It is compared by the range value corresponding to the points to a certain range of Fast Fourier Transform (FFT) resultThe Frequency point that signal is included is obtained, data source is provided for subsequent comparison judgement.

In the present embodiment, the thick synchronization of carrier wave of step 2 completes the synchronous of carrier wave with the carrier wave essence synchronization combining of step 4,Realization is more simple compared with traditional carrier synchronization mode, as a result more accurate.The thick of step 2 is synchronized in a manner of big steppingRealize that signal is moved from upper frequency to lower frequency, upper frequency is near 70MHz, and lower frequency is near zero-frequency.It is passing throughAfter crossing filtering extraction, data sampling rate reduces, and sets relatively small number of FFT points that can obtain higher resolution ratio, higherResolution ratio ensure that the effect of Doppler frequency spectrum correction.Calculating step is：

(1) if input signal S is IF input signals, i.e.,

S=A cos (2 π (f_c+Δf)t+β₁sin(2πf₁t)+β₂sin(2πf₂t)...+β_n-1sin(2πf_n-1t)+β_nsin(2πf_nt))

Wherein | Δ f | ＜ f_d, f_dMaximum value for Doppler shift；

The mixed frequency signal of down coversion is：

S_h=cos (2 π f_ct)

Low frequency component after down coversion：

S_l=cos (2 π Δ ft+ β₁sin(2πf₁t)+β₂sin(2πf₂t)...+β_n-1sin(2πf_n-1t)+β_nsin(2πf_nt))

Low frequency component after Doppler shift is corrected：

S_{f_cos}=cos (β₁sin(2πf₁t)+β₂sin(2πf₂t)...+β_n-1sin(2πf_n-1t)+β_nsin(2πf_nt))

S_{f_sin}=sin (β₁sin(2πf₁t)+β₂sin(2πf₂t)...+β_n-1sin(2πf_n-1t)+β_nsin(2πf_nt))

So far become the synchronization for completing carrier wave, obtain Equivalent Base-Band signal.

In the present embodiment, in step s 5, symbol is realized by using the mode of lead-lag in FPGAIt is synchronous.According to the modulation principle of multitone FM signal and the mode of carrying information, when carrying out symbol synchronization without preciselyIt is synchronous, but require control errors by analysis, emulation show that the worst error of permission is accorded with for 1/4 symbol in rational rangeNumber, therefore the mode of lead-lag is taken to make to carry out symbol synchronization, it is sufficient to ensure the correctness of single-tone detection.

In the present embodiment, in step sl, in digital communication systems, according to Nyquist (Niquist) criterion, samplingFrequency will generally be more than 2 times of signal bandwidth.The sample frequency of A/D is set as fs, meets following condition：

Wherein, M=1,2 ... ..., fs >=2B；

It is sampled by A/D conversion chips AD9626, carrier frequency is modulated for the analog intermediate frequency multitone frequency modulation of 70MHzSignal is converted to 12bits multitone frequency-modulated digital signals, then inputs and following digital processing is carried out in the chip of ALTERRA.

In the present embodiment, in step s 2, it is as follows：

Input signal S is intermediate-freuqncy signal, i.e.,

S=A cos (2 π (f_c+Δf)t+β₁sin(2πf₁t)+β₂sin(2πf₂t)...+β_n-1sin(2πf_n-1t)+β_nsin(2πf_nt))

Wherein, | Δ f | ＜ f_d, f_dMaximum value for Doppler shift；

The mixed frequency signal of down coversion is：

S_m=cos (2 π f_ct)

With multiplier by S and S_mIt is multiplied, realizes down coversion.

Low frequency component after down coversion is：

S_l=cos (2 π Δ ft+ β₁sin(2πf₁t)+β₂sin(2πf₂t)...+β_n-1sin(2πf_n-1t)+β_nsin(2πf_nt))

Signal by down coversion is there are Doppler frequency spectrum, in order to which the amounts of frequency offset for obtaining degree of precision needs first to carry out reduction of speedProcessing.

In the present embodiment, in step s3, sampling rate/extracting multiple before post-sampling rate=extraction is extracted.

The type of decimation filter mainly has integral comb filter and half-band filter, may be used in combination according to demand,If desired for 25 times of extractions, the CIC cascades that two 5 times of extractions may be used are realized；If desired for 32 times of extractions, two 5 are may be usedFive HBF cascade realizations are realized in the CIC cascades extracted again；If desired for 40 times extraction, may be used one 5 times extraction CIC andThree HBF cascades are realized.

In the present embodiment, in step s 4, specifically：

S_l=cos (2 π Δ ft+ β₁sin(2πf₁t)+β₂sin(2πf₂t)...+β_n-1sin(2πf_n-1t)+β_nsin(2πf_nt))

△ f in above formula represent Doppler shift amount, its offset can be obtained by FFT.

Doppler shift is corrected by the way of mixing：

S_Q=cos (2 π Δ ft)

S_Q=sin (2 π Δ ft)

S_IR=S_l·S_I

S_QI=S_l·S_Q

By S_IR, S_QILow-pass filter is respectively fed to, filters out out-of-band noise and high frequency images component.

The cutoff frequency of low-frequency filter should be greater than maximum sidetone frequency, while should ensure that and farthest filter out unrelated pointAmount and image component.

There are two types of ways of realization for the low-pass filter of number, that is, have limit for length's unit impulse response (Finite ImpulseResponse, FIR) digital filter and infinite impulse response (Infinite Impulse Response, IIR) digital filteringDevice, Finite Impulse Response filter are compared with iir digital filter, are had many good qualities, such as the features such as linear phase and stability andDesign is relatively ripe, therefore realizes low-pass filtering using Finite Impulse Response filter.

Complete Doppler shift and correct and pass through the signal of wave digital lowpass filter be:

S_{f_cos}=cos (β₁sin(2πf₁t)+β₂sin(2πf₂t)...+β_n-1sin(2πf_n-1t)+β_nsin(2πf_nt))

S_{f_sin}=sin (β₁sin(2πf₁t)+β₂sin(2πf₂t)...+β_n-1sin(2πf_n-1t)+β_nsin(2πf_nt))

Respectively to S_{f_cos}And S_{f_sin}Lead-lag synchronization is carried out, for the points foundation of synchronous sum of series lead-lagRequired precision is configured.

By S_{f_cos}And S_{f_sin}Spectrum analysis is carried out to carry out single-tone detection.

It is the FFT frequency points corresponding to single-tone by the result that FFT is detected, being converted to it, it is corresponding to can obtainMonotony frequency, reduction formula are as follows：

Monotony frequency=FFT frequency points/FFT always counts * sample rates.

Such as detect acquisition FFT frequency points be 38, FFT always points be 16384, sample rate 1.6384MHz, then correspondenceMonotony frequency be 38/16384*1638400=3800Hz.

In demodulating process, local oscillator, multiplier, low-pass filter, data delay caching and FFT pass through FPGAMiddle IP kernel generation is realized.

Finally it should be noted that：The above embodiments are merely illustrative of the technical solutions of the present invention, rather than its limitations；AlthoughThe present invention is described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that：It still may be usedTo modify to the technical solution recorded in foregoing embodiments or carry out equivalent replacement to which part technical characteristic；And these modification or replace, various embodiments of the present invention technical solution that it does not separate the essence of the corresponding technical solution spirit andRange.

Claims (7)

1. a kind of multi-tone frequency demodulated signal method based on FPGA, it is characterized in that, include the following steps：

S1, multitone frequency-modulated analog letter signal is converted into digital signal；

S2, multi-tone frequency digital medium-frequency signal down coversion is carried out in FPGA；

S3, it is sampled in FPGA, reduces signal data rate；

S4, Doppler shift correction is carried out in FPGA；

S5, independent component is filtered out by low-pass filter in FPGA；

S6, the detection to pilot tone and single-tone is completed by spectrum analysis in FPGA, useful signal is extracted, so as to fulfill multi-toneThe demodulation of frequency signal.

2. the multi-tone frequency demodulated signal method according to claim 1 based on FPGA, it is characterized in that, the step S1In, it is sampled by A/D conversion chips AD9626, by the modulated letter of analog intermediate frequency multitone frequency modulation that carrier frequency is 70MHzNumber 12bits multitone frequency-modulated digital signals are converted to, then input and following digital processing is carried out in the chip of ALTERRA.

3. the multi-tone frequency demodulated signal method according to claim 2 based on FPGA, it is characterized in that, the step S2In, the mixed frequency signal cos (2 π ft), wherein f=70MHz that frequency is 70MHz are generated by local oscillator；It is by carrier frequencyThe signal mixing that the intermediate-freuqncy signal and local oscillator of 70MHz generates, realizes down coversion.

4. the multi-tone frequency demodulated signal method according to claim 1 based on FPGA, it is characterized in that, the step S3In, using the integral multiple of the least common multiple in all single-tone periods as the length of a symbol, and it is used as this group of multitoneCombine the period of signal.

5. the multi-tone frequency demodulated signal method according to claim 1 based on FPGA, it is characterized in that, the step S4In, frequency deviation position is obtained by Fast Fourier Transform (FFT), then Doppler shift correction is completed by way of mixing.

6. the multi-tone frequency demodulated signal method according to claim 1 based on FPGA, it is characterized in that, pass through the stepCarrier wave slightly synchronous synchronous, the thick synchronization of the step S2 that carrier wave is completed with the carrier wave essence synchronization combining of the step S4 in S2Realize that signal is moved from upper frequency to lower frequency in a manner of big stepping.

7. the multi-tone frequency demodulated signal method according to claim 1 based on FPGA, it is characterized in that, the step S5In, the synchronization of symbol is realized by using the mode of lead-lag in FPGA.