CN104735007A - Direct center frequency channelizing method free of constraint to center frequencies - Google Patents

Abstract

The invention discloses a direct center frequency channelizing method free of the constraint to center frequencies, belongs to the technical field of digital signal processing, and particularly relates to the digital channelizing technology. Firstly, the data rate of signals is reduced through multiple branches and a D multiple extraction mode, the integrity of data is guaranteed, and the data are not lost; the branches are filtered, inverse discrete Fourier transformation is conducted, multiple output signals are modulated, and output signals of channels are acquired. K sub-channel signals are acquired at the same time only through one filter; center frequency signals are directly channelized, and a multi-channel parallel DDC and a channelizing module behind the DDC adopted in a traditional bandwidth digital channelizing method are omitted; the positions of the center frequencies of sub-channels are not specially limited, and when the center frequencies of the sub-channels are wholly moved leftwards or rightwards, only the coefficients of multiple filters and re-modulated signals acquired after inverse discrete Fourier transformation need to be changed.

Description

A kind of to the unconfined direct intermediate-frequency channel method of centre frequency

Technical field

The invention belongs to digital signal processing technique field, be specifically related to digital channelizing technology.

Background technology

Electronic warfare is the important component part in modern battlefield, has significant impact to the result of war.EW receiver is the important system in electronic warfare, has important researching value.

On modern battlefield, in time domain or on frequency domain, be all flooded with highdensity electromagnetic signal, new challenge is proposed to EW receiver.These signals have following characteristics: frequency range is wide, and the operating frequency range of current most of radar systems covers 2 ~ 18GHz; Instant bandwidth is large, and the bandwidth of operation of some wideband radar is more than 1GHz; Signal density is large, and at present for the full frequency band receiver of 0.1 ~ 40GHz frequency range, signal density can reach 100 ~ 5,000,000 pulses per second.These features require modern electronic warfare receiver to have large instant bandwidth, arriving signal can process multiple while, the performance such as real-time.

In the developing history of EW receiver, there is eurypalynous receiver perhaps, as crystal video receiver, instantaneous frequency measurement receiver, superheterodyne receiver, compressive receiver, Prague receiver, channelized receiver.Various types of receiver all defectiveness, receiver without any type ideally can meet the requirement of electronic warfare, comparatively speaking, channelized receiver because of its to while arriving signal there is best sorting capability, in electronic warfare, there is using value most, be subject to the research that various countries are extensive and deep.

In traditional digital channelized receiver, usually DDC (Digital Down Converter is first utilized, digital down converter) intermediate frequency real signal is converted to baseband complex signal, then utilize highly effective algorithm (as discrete Fourier transform, discrete cosine transform etc.) to realize channelizing.When the bandwidth inputting intermediate frequency real signal is very large, because the operational capability of digital signal processor is limited, need multiple DDC that large bandwidth Channel division is become multiple little bandwidth channel, then respectively to the further channelizing of multiple little bandwidth channels.This process is equivalent to two-stage channelizing, and the first order utilizes DDC to carry out coarse channel, and the second level utilizes highly effective algorithm to carry out thin channelizing.The operation efficiency of this method is low, cost is high, and becomes along with the increase of input signal bandwidth and become increasingly complex.

In order to the efficiency solving multi-stage channel is low, cost is high, baroque problem, the present invention proposes a kind of method of directly intermediate-freuqncy signal being carried out to channelizing, the method only needs one-level computing to realize channelizing, and the method utilizes multiphase filtering and IDFT (inverse discrete Fourier transform) to optimize operation efficiency dexterously.Another advantage of the method is, does not have special requirement to the centre frequency of input intermediate-freuqncy signal.Relative to traditional channelization method, resource utilization, the operation efficiency of the method increase substantially.

Summary of the invention

Technical problem to be solved by this invention is, provides a kind of method to wideband IF signal direct channels, thus reaches the object that required computational resource is few, cost is low, efficiency is high.

The present invention for solving the problems of the technologies described above adopted technical scheme is, a kind of to the unconfined direct intermediate-frequency channel method of centre frequency.

In the present invention, suppose that receiving intermediate frequency signal as shown in Figure 1.In Fig. 1, the bandwidth of intermediate-freuqncy signal is B, occupies K sub-channels, and the bandwidth sum centre frequency interval of subchannel is all △ B=2 π/M=B/K, and wherein M is positive integer, is called channel factors.The centre frequency of kth sub-channels is ω _k=ω ₀+ 2 π k/M, wherein ω ₀be the centre frequency of the 0th sub-channels, ω ₀can optional position in (0, π) scope, as long as meet the condition of frequency spectrum not aliasing.

DDC is utilized to obtain the structured flowchart of kth sub-channels signal as shown in Figure 2.If the length of filter h [n] is N, and N and channel factors M meets relational expression N=LM, wherein L is positive integer.Then the output signal of kth sub-channels can be expressed as formula 1.

$\begin{array}{c}{y}_k\left[n\right]=\underset{m=0}{\overset{N-1}{\mathrm{\Σ}}}h\left[m\right]x[\mathrm{nD}-m]e^{-j({\mathrm{\ω}}_0+2\mathrm{\πk}/M)(\mathrm{nD}-m)}\\ =e^{-j{\mathrm{\ω}}_0\mathrm{nD}}\underset{m=0}{\overset{N-1}{\mathrm{\Σ}}}{e}^{j{\mathrm{\ω}}_{0}m}h\left[m\right]x[\mathrm{nD}-m]e^{-j2\mathrm{\πk}(\mathrm{nD}-m)/M}\end{array}$ Formula 1

When channel factors M and extraction factor D meet relational expression M=FD (F is positive integer), make in formula 1 if regard w [m] as new complex filter, and carry out M phase decomposition to it, formula 2 can be obtained.

$\begin{array}{c}{y}_k\left[n\right]=e^{-j{\mathrm{\ω}}_0\mathrm{nD}}\underset{m=0}{\overset{N-1}{\mathrm{\Σ}}}w\left[m\right]x[\mathrm{nD}-m]e^{-j2\mathrm{\πk}(\mathrm{nD}-m)/M}\\ =e^{-j{\mathrm{\ω}}_0\mathrm{nD}}\underset{i=0}{\overset{N/M-1}{\mathrm{\Σ}}}\underset{p=0}{\overset{M-1}{\mathrm{\Σ}}}w[\mathrm{iM}+p]x[\mathrm{nD}-\mathrm{iM}-p]e^{-j2\mathrm{\πk}(\mathrm{nD}-\mathrm{iM}-p)/M}\\ =e^{-j{\mathrm{\ω}}_0\mathrm{nD}}\underset{i=0}{\overset{N/M-1}{\mathrm{\Σ}}}\underset{p=0}{\overset{M-1}{\mathrm{\Σ}}}w[\mathrm{iM}+p]x[(n-\mathrm{iF})D-p]e^{-j2\mathrm{\πk}[(n-\mathrm{iF})D-p]/M}\end{array}$ Formula 2

Make x _p[i]=x [iD-p], g _p[i]=w [iM+p], c _p[i]=g _p[i/F] (i.e. c _p[i] is g _pthe F times of interpolation of [i]), then formula 2 can be equivalent to formula 3, and wherein " * " represents convolution algorithm, MIDFT _k{ x} represents kth output x being carried out to M point IDFT computing.

$\begin{array}{c}{y}_k\left[n\right]=e^{-j{\mathrm{\ω}}_0\mathrm{nD}}\underset{i=0}{\overset{L-1}{\mathrm{\Σ}}}\underset{p=0}{\overset{M-1}{\mathrm{\Σ}}}{g}_p\left[i\right]x_p[n-\mathrm{iF}]e^{-j2\mathrm{\πk}(n-\mathrm{iF})D/M}{e}^{j2\mathrm{\πkp}/M}\\ =e^{-j{\mathrm{\ω}}_0\mathrm{nD}}\underset{l=0}{\overset{(L-1)F}{\mathrm{\Σ}}}\underset{p=0}{\overset{M-1}{\mathrm{\Σ}}}{c}_p\left[l\right]x_p[n-l]e^{-j2\mathrm{\πk}(n-l)D/M}{e}^{j2\mathrm{\πkp}/M}\\ =e^{-j{\mathrm{\ω}}_0\mathrm{nD}}\underset{p=0}{\overset{M-1}{\mathrm{\Σ}}}\{\left(x_p\right[n\left]e^{-j2\mathrm{\πknD}/M}\right)*c_p[n\left]\right\}{e}^{j2\mathrm{\πkp}/M}\\ =e^{-j{\mathrm{\ω}}_0\mathrm{nD}}\×M\·{\mathrm{IDFT}}_k\{\left(x_p\right[n\left]e^{-j2\mathrm{\πkn}/F}\right)*c_p[n\left]\right\}\end{array}$ Formula 3

The channelization structure of formula 3 correspondence as shown in Figure 3.In Fig. 3, each computing can only obtain the output y of a sub-channels _k[n], efficiency is lower.After complex mixer before Fig. 3 median filter is moved on to IDFT module, obtain the equivalent structure as Fig. 4.In Fig. 4, the output of IDFT is all multiplied with same complex exponential signal, and the result be multiplied only has a branch road to be effective.If each output signal of IDFT is multiplied with the complex exponential signal of correspondence, then can export K sub-channels signal, the efficient channel structure obtained thus as shown in Figure 5 simultaneously.

The present invention is a kind of to the unconfined direct intermediate-frequency channel method of centre frequency, and the method comprises the following steps:

Step 1: receiving intermediate frequency signal, the bandwidth of this intermediate-freuqncy signal is B, occupies K sub-channels, Received signal strength is inputted successively M branch road, sampling period of time delay successively during every bar branch input signal; Carry out D to M branch road more doubly to extract, data transfer rate has been dropped to the 1/D of receiving data rate; Wherein M is integer, and its value is by formula f _s/ M=B/K determines, f _sfor the sample rate of Received signal strength;

Step 2: plural multiphase filtering is carried out to the extracted data of M branch road;

Step 3: carry out leaf inverse transformation in M point discrete Fourier to the output signal of M phase filtering, obtains M intermediate output signal;

Step 4: before obtaining step 3, K output signal carries out complex-exponential-modulation, obtains final K sub-channels signal.

A described step 2 couple M branch road is according to complex filter carry out multiphase filtering, wherein h [m] is the prototype lowpass filter corresponding with subchannel bandwidth, ω ₀represent the centre frequency of the 0th sub-channels.

Described step 4 adopts signal before obtaining step 3, K output signal carries out complex-exponential-modulation, wherein n round numbers.

The present invention has taken into full account the feature of IF input signals, utilizes multiphase filtering and IDFT computing to realize direct intermediate-frequency channel.The invention has the beneficial effects as follows:

1, the 1st step first extracts, and the data transfer rate of subsequent treatment has dropped to the 1/D of input data transfer rate, and operation efficiency is high;

2, utilize multiphase filtering and IDFT computing, only need 1 filter just to obtain K sub-channels signal simultaneously;

3, directly carry out channelizing to intermediate-freuqncy signal, save the channelization block after multidiameter delay DDC and DDC of Conventional wide band digital channelizing method, resource requirement significantly reduces;

4, the position of the IF-FRE of sub-channel does not have special restriction, when the centre frequency entirety of subchannel moves to left or moves to right, only need change the complex modulated signal after the coefficient w [n] of complex filter in Fig. 5 and IDFT .

Accompanying drawing explanation

Fig. 1 is the division of input intermediate-freuqncy signal and subchannel;

Fig. 2 is the structured flowchart utilizing DDC to obtain kth sub-channels signal;

Fig. 3 is direct intermediate-frequency channel structure of the present invention;

Fig. 4 is the equivalent structure of direct intermediate-frequency channel of the present invention;

Fig. 5 is the efficient configuration of direct intermediate-frequency channel of the present invention;

Fig. 6 (a) is the channel division method of simulation example and the amplitude-frequency characteristic of filter;

Fig. 6 (b) is the amplitude-frequency characteristic of the input signal of simulation example;

Fig. 7 is the amplitude-frequency characteristic of the output signal of the subchannel of simulation example;

Fig. 8 is the output signal real part waveform of the subchannel of simulation example;

Fig. 9 is the output signal imaginary part waveform of the subchannel of simulation example;

Figure 10 is the instantaneous amplitude of the output signal of the subchannel of simulation example.

Embodiment

Those of ordinary skill in the art will appreciate that, embodiment described here is to help reader understanding's implementation method of the present invention, should be understood to that protection scope of the present invention is not limited to so special statement and embodiment.Those of ordinary skill in the art can make various other various concrete distortion and combination of not departing from essence of the present invention according to these technology enlightenment disclosed by the invention, and these distortion and combination are still in protection scope of the present invention.

In following simulation example, sample rate is f _s=160MHz, channel factors M=16, subchannel bandwidth △ B=10MHz, number of subchannels K=4, the extraction factor is D=8, and the output sampling rate of subchannel is 20MHz.Centre frequency and the input signal of subchannel are as shown in table 1, and input signal is the summation of 3 kinds of signals in table 1.Simulation result is as shown in Fig. 6 ~ 10.

Fig. 6 (a) is the amplitude-frequency characteristic of channel division method and filter.The passband corner frequency of lowpass prototype filter is 5MHz, and stopband corner frequency is 5.5MHz.Here in order to draw conveniently, the amplitude-versus-frequency curve of the filter of each channel is moved on in respective centre frequency, in fact only need with a low pass filter.

Fig. 6 (b) is the amplitude-frequency characteristic of input signal, and input signal is the summation of 3 kinds of signals in table 1.

Fig. 7 is the amplitude-frequency characteristic of the output signal of subchannel, and Fig. 8 is corresponding real part waveform, and Fig. 9 is corresponding imaginary part waveform.There is no signal in channel 2, output corresponding input signal in channel 1,3,4, demonstrate the correctness of channelization method of the present invention.Each channel ± 5MHz beyond signal be subject to the suppression of filter, little within Amplitude Ratio ± 5MHz.

Figure 10 is the instantaneous amplitude of the output signal of subchannel.Instantaneous amplitude in channel 1,3,4 obviously non-vanishing time is identical with the duration of input signal.In channel 2, there are some irregular spikes, this be due to the transition band of filter can not be zero, stopband can not infinite attenuation, the signal of other channel can be come in by aliasing.The duration of these spikes is very short and irregular, can judge to reject.

The simulated conditions of the unconfined direct intermediate-frequency channel method of table 1 pair centre frequency

Channel number	Channel center frequency	Input signal
			1	33MHz	Real cosine continuous signal: f 0＝34MHz
2	43MHz	Nothing
			3	53MHz	Solid linear chirp signal: f 0＝51MHz,B＝2MHz,τ＝20μs
4	63MHz	Solid linear chirp signal: f 0＝64MHz,B＝5MHz,τ＝10μs

Claims (3)

1., to the unconfined direct intermediate-frequency channel method of centre frequency, the method comprises the following steps:

Step 1: receiving intermediate frequency signal, the bandwidth of this intermediate-freuqncy signal is B, occupies K sub-channels, Received signal strength is inputted successively M branch road, sampling period of time delay successively during every bar branch input signal; Carry out D to M branch road more doubly to extract, data transfer rate has been dropped to the 1/D of receiving data rate; Wherein M is integer, and its value is by formula f _s/ M=B/K determines, f _sfor the sample rate of Received signal strength;

Step 2: plural multiphase filtering is carried out to the extracted data of M branch road;

Step 3: carry out leaf inverse transformation in M point discrete Fourier to the output signal of M phase filtering, obtains M intermediate output signal;

Step 4: before obtaining step 3, K output signal carries out complex-exponential-modulation, obtains final K sub-channels signal.

2. as claimed in claim 1 a kind of to the unconfined direct intermediate-frequency channel method of centre frequency, it is characterized in that a step 2 couple M branch road is according to complex filter w [m]=h [m] e ^{j ω om}carry out multiphase filtering, wherein h [m] is the prototype lowpass filter corresponding with subchannel bandwidth, ω ₀represent the centre frequency of the 0th sub-channels.

3. as claimed in claim 1 a kind of to the unconfined direct intermediate-frequency channel method of centre frequency, it is characterized in that described step 4 adopts signal e ^{-j ω 0nD}before obtaining step 3, K output signal carries out complex-exponential-modulation, wherein n round numbers.