A method for generating broadband complex background signals based on frequency domain processing

Abstract

The present invention discloses a method for generating a broadband complex background signal based on frequency domain processing, comprising the following steps: S1, generating broadband noise time domain data with flat frequency domain and designated broadband noise spectrum envelope data; S2, generating several narrowband signals, performing sampling rate conversion and frequency fine-tuning on the narrowband signals; S3, performing partially overlapping segmented FFT transformation on the broadband noise time domain data and the narrowband time domain data, respectively, to obtain broadband flat noise spectrum and narrowband spectrum data, and merging the broadband flat noise spectrum and the narrowband spectrum data after amplitude weighting; S4, performing IFFT transformation on the merged broadband spectrum data, to obtain broadband time domain data; S5, extracting and splicing multiple segments of broadband time domain data, to obtain continuous broadband time domain data. Through the above-mentioned method, the present invention can simulate and generate a broadband complex electromagnetic environment containing designated frequency domain envelope noise and numerous background signals with relatively low complexity.

Description

Technical Field

[0001] The invention relates to the fields of radio electronics and signal processing technology, and in particular to a method for generating a broadband complex background signal based on frequency domain processing. Background Art

[0002] In the performance inspection of radio monitoring equipment, in order to evaluate the adaptability of the equipment to the harsh background signal environment, it is usually involved in performance tests carried out in a complex electromagnetic environment. The typical characteristics of a complex electromagnetic environment are background electromagnetic signals with a large number of frequencies and different system styles, as well as background noise signals with fluctuating frequency domain envelopes. For third-party non-cooperative background signals in the electromagnetic space, they often appear in a third-party, non-cooperative, intermittent and sudden manner. The time of occurrence, frequency of occurrence, and signal power are difficult to determine in advance and difficult to reproduce. In order to more accurately evaluate the performance of equipment in a complex electromagnetic environment, the background signals required by the complex electromagnetic environment are often generated through simulation.

[0003] The broadband background signal includes two parts: the background electromagnetic signal with specified parameters and the broadband background noise signal. The traditional background noise signal is generally generated by time domain method, which makes it difficult to realize a broadband noise environment in which the envelope fluctuates with the frequency domain. It can only simulate flat broadband background noise. The generation of background electromagnetic signal can be divided into two steps: generating several baseband signals and broadband merging of multiple baseband signals. Among them, the baseband signal generation mainly generates the required narrowband waveform according to the specified modulation and coding parameters of the information bits, and the broadband merging of multiple baseband signals mainly merges multiple narrowband signals into unified broadband data through a series of digital processing as an excitation signal for external transmission. Among them, the traditional method of merging broadband multi-baseband signals is to transform each baseband signal into broadband data with high sampling rate and corresponding radio frequency (or intermediate frequency) by time domain interpolation and frequency conversion, and then merge multiple broadband data containing a single narrowband signal in time domain. When constructing complex electromagnetic background signals, the number and types of signals that can be generated simultaneously are important capability indicators to measure the complexity of the constructed background. At the same time, the bandwidth of the broadband signal containing multiple narrowband signals is often much higher than the baseband signal bandwidth. Conventional broadband multi-signal generation methods require high-multiple time domain interpolation and high sampling rate frequency conversion of a large number of baseband signals, which has a large computational complexity. Under relatively limited computing resources, the number, diversity, or frequency distribution of the baseband signals generated is limited, making it difficult to meet the broadband multi-signal generation requirements of some complex electromagnetic environments. Summary of the invention

[0004] The main technical problem solved by the present invention is to provide a method for generating broadband complex background signals based on frequency domain processing, which can generate broadband complex background signals with a large number of electromagnetic signal styles, wide frequency distribution, and background noise fluctuating with frequency with low computational complexity, and meet the background signal simulation requirements in complex electromagnetic environment testing.

[0005] In order to solve the above technical problems, a technical solution adopted by the present invention is: to provide a method for generating a broadband complex background signal based on frequency domain processing, comprising the following steps:

[0006] S1, generate frequency-domain flat broadband noise time domain data n and specify broadband noise spectrum envelope data A n ;

[0007] S2, generate several narrowband signals s i , and based on the broadband spectrum resolution f d and the frequency of the narrowband signal in the broadband f wi , for narrowband signal s i Perform sampling rate conversion and frequency fine-tuning to obtain narrowband time domain data x i ;

[0008] S3, broadband noise time domain data n and narrowband time domain data x i Perform partially overlapping segmented FFT transforms respectively to obtain the first k The broadband flat noise spectrum of segment N 0k and narrowband spectrum data Y ik The broadband flat noise spectrum and narrowband spectrum data are amplitude-weighted and then merged to obtain the broadband spectrum data Z k ;

[0009] S4. For broadband spectrum data Z k Perform IFFT transformation to obtain k The broadband time domain data of segment z k ;

[0010] S5. Extract and splice multiple segments of broadband time domain data to remove z k The total of the first and last L w Length data, and k-1 The segment data are spliced ​​to obtain continuous broadband time domain data.

[0011] Preferably, the broadband noise time domain data n with flat frequency domain in step S1 can be generated by using Gaussian noise or other methods; in order to improve the operation efficiency of FFT and IFFT, the broadband noise spectrum envelope data A n Data length Nw The frequency resolution should be selected as 2 to the power of m to reduce the complexity of time domain data sampling rate conversion and carrier frequency fine-tuning. f d Should be selected as an integer, corresponding to the broadband spectrum data Z k The sampling rate is F w =N w f d .

[0012] Preferably, the sampling rate conversion and frequency fine-tuning in step S2 are mainly used to achieve spectral line adaptation of the narrowband spectrum and the broadband noise spectrum, and should follow the following principles: i Narrowband time domain data x i Sampling rate F i Should meet F i= N i f d ( N i is an integer representing the i At the same time, if the i-th signal has a symbol rate of R si The digital signal should satisfy F i >R si , if the bandwidth is B i The analog signal should satisfy F i >B i , the carrier frequency at baseband f ni Should satisfy m=( f wi - f ni ) / f d The value of is an integer.

[0013] Preferably, in steps S3 and S5, the partially overlapping FFT transformation and the segmented splicing of the broadband time domain data are performed according to the following principles:

[0014] Y1) When performing FFT operation on broadband noise time domain data and i-th narrowband time domain data, the overlapping data lengths are L w and L i, the overlap length is greater than 0 and less than the FFT length. The typical value can be selected as half of the FFT length;

[0015] Y2) When extracting broadband time domain data, keep z k Data Center N w - L w Length data, remove the first and last L w data, when L w When it is an even number, remove the first and last L w / 2 data; when L w When / 2 is an odd number, remove the first and last ( L w -1) / 2+1 and ( L w -1) / 2, or remove the first and last ( L w -1) / 2 and ( L w -1) / 2+1 data.

[0016] Preferably, the amplitude weighting of the broadband flat noise spectrum in step S3 refers to weighting the broadband flat noise spectrum N 0k With A n The data is multiplied point by point to obtain a broadband background noise spectrum with a specified frequency domain envelope and random noise characteristics.

[0017] Preferably, the amplitude weighting of the narrowband spectrum data in step S3 refers to weighting the narrowband spectrum data Y ik With the specified signal amplitude A i To multiply, A i The value of is obtained according to the signal-to-noise ratio requirements (or amplitude requirements) of different signals. In order to avoid the jump of the narrowband spectrum edge on the broadband spectrum, it is also necessary to add a narrowband filter. ik The corresponding filter spectrum H i It is implemented as a dot product in the frequency domain.

[0018] Preferably, the merging of the weighted broadband flat noise spectrum and the narrowband spectrum data in step S3 refers to combining each narrowband spectrum with its corresponding position in the broadband spectrum ( f wi - f ni ) / f dThe data are added to realize the up-conversion operation of the narrowband signal in the frequency domain.

[0019] Compared with the prior art, the present invention has the following beneficial effects:

[0020] It can generate broadband background noise that meets the specified frequency domain envelope requirements, which is more in line with the actual broadband noise environment that fluctuates with frequency;

[0021] The ability to generate several narrowband signals within a broadband with lower complexity helps to generate a more complex background signal environment with more signals under limited resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 This is a block diagram of the implementation principle of a method for generating broadband complex background signals based on frequency domain processing.

[0023] Figure 2 It is the main step of a method for generating broadband complex background signals based on frequency domain processing.

[0024] Figure 3 It is the broadband noise spectrum envelope required to be generated in the implementation example.

[0025] Figure 4 This is an implementation example that requires a narrowband signal contained within a wideband.

[0026] Figure 5 This is a broadband complex background signal generated by the method of the present invention in the implementation example.

[0027] Figure 6 It is the original narrowband signal in the implementation example and the corresponding narrowband signal extracted from the generated broadband signal. DETAILED DESCRIPTION

[0028] The preferred embodiments of the present invention are described in detail below in conjunction with the accompanying drawings so that the advantages and features of the invention can be more easily understood by those skilled in the art, thereby making a clearer and more definite definition of the protection scope of the present invention.

[0029] See also Figures 1 to 6 , the embodiment of the present invention includes:

[0030] Complex background signal generation requirements:

[0031] The test of a certain device requires a complex background signal of shortwave 1.5MHz~30MHz, and the background signal generated is required to be as Figure 3 The background noise spectrum envelope shown in the figure generates the following five narrowband signals (the baseband signal is shown in the time-frequency domain as follows Figure 4 shown):

[0032] Signal 1: QPSK, symbol rate 1600Bd, RF signal frequency 1.8MHz, signal-to-noise ratio 20dB;

[0033] Signal 2: 2FSK, symbol rate 6400Bd, RF signal frequency 10MHz, signal-to-noise ratio 25dB;

[0034] Signal 3: 16QAM, symbol rate 3200Bd, RF signal frequency 16.4MHz, signal-to-noise ratio 20dB;

[0035] Signal 4: single tone, signal frequency 24.66MHz, RF signal-to-noise ratio 10dB;

[0036] Signal 5: AM, bandwidth 12kHz, signal frequency 29.6MHz, signal-to-noise ratio 15dB.

[0037] The present invention is used to generate complex background signals:

[0038] A method for generating a broadband complex background signal based on frequency domain processing comprises the following steps:

[0039] S1, use Gaussian random noise to generate frequency domain flat broadband noise time domain data n, and generate frequency resolution f d =320Hz, resulting in data length N w =2 17 Broadband noise spectrum envelope data A n , the corresponding broadband data sampling rate is F w =N w f d =41.943040MHz, which can cover the required bandwidth of 1.5MHz~30MHz;

[0040] S2. Generate the corresponding narrowband signal s according to the required 5 signal modulation parameters i ( i =1,...,5), and according to the broadband frequency resolution f d and the required signal carrier frequency and signal symbol rate, the narrowband sampling rates used were converted to 6.4kHz, 25.6kHz, 12.8kHz, 3.2kHz, and 25.6kHz, respectively, and the corresponding FFT lengths were 20, 80, 20, 10, and 80, respectively. The frequencies of the five signals were fine-tuned to 0Hz, 0Hz, 0Hz, 160Hz, and 0Hz, respectively;

[0041] S3, set the overlap size of the FFT segment to 1 / 2 of the corresponding FFT length, and perform segmented FFT calculation to obtain the first k The broadband flat noise spectrum of segment N 0k and narrowband spectrum data Y ik The broadband flat noise spectrum and narrowband spectrum data are amplitude-weighted and then merged to obtain the broadband spectrum data Z k like Figure 5 As shown, we can see that the background noise spectrum envelope meets the set value, and the narrowband signal has been moved to the corresponding frequency;

[0042] S4. For broadband spectrum data Z k Perform IFFT transformation to obtain k The broadband time domain data of segment z k ;

[0043] S5. Extract and splice multiple segments of broadband time domain data to remove z k The total of the first and last L w =2 16 Length data, and k-1 By splicing the segment data, the required broadband time domain data can be obtained.

[0044] From the generated broadband time domain data, the corresponding five narrowband signals are extracted respectively and compared with the original narrowband signals in step S2, such as Figure 6 As shown, it can be seen that the generated signal has good consistency with the original narrowband signal except for the signal fluctuation caused by the added noise in the process, which can verify that the method of the present invention can generate broadband complex background signal data containing multiple signals.

[0045] The invention discloses a method for generating a broadband complex background signal based on frequency domain processing, which can simulate and generate a broadband complex electromagnetic environment containing specified frequency domain envelope noise and numerous background signals with relatively low complexity.

[0046] The above descriptions are merely embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for generating a broadband complex background signal based on frequency domain processing, characterized in that: The following steps are involved: S1, generate frequency-domain flat broadband noise time domain data n and specify broadband noise spectrum envelope data A n ; S2, generate several narrowband signals s i , and based on the broadband spectrum resolution f d and the frequency of the narrowband signal in the broadband f wi , for narrowband signal s i Perform sampling rate conversion and frequency fine-tuning to obtain narrowband time domain data x i ; S3, broadband noise time domain data n and narrowband time domain data x i Perform partially overlapping segmented FFT transforms respectively to obtain the first k The broadband flat noise spectrum of segment N 0k and narrowband spectrum data Y ik The broadband flat noise spectrum and narrowband spectrum data are amplitude-weighted and then merged to obtain the broadband spectrum data Z k ; S4. For broadband spectrum data Z k Perform IFFT transformation to obtain k The broadband time domain data of segment z k ; S5. Extract and splice multiple segments of broadband time domain data to remove z k The total of the first and last L w Length data, and k-1 By splicing the segment data, continuous broadband time domain data can be obtained; The amplitude weighting of the broadband flat noise spectrum in step S3 refers to the broadband flat noise spectrum N 0k With A n The data is multiplied point by point to obtain a broadband background noise spectrum with a specified frequency domain envelope and random noise characteristics; The amplitude weighting of the narrowband spectrum data in step S3 refers to the narrowband spectrum data Y ik With the specified signal amplitude A i To multiply, A i The value of is obtained according to the signal-to-noise ratio requirement or amplitude requirement of different signals. In order to avoid the jump of the edge of the narrowband spectrum on the broadband spectrum, it is also necessary to add a narrowband filter. ik The corresponding filter spectrum H i It is implemented as a dot product in the frequency domain.

2. The method for generating a broadband complex background signal based on frequency domain processing according to claim 1, characterized in that: In step S1, the broadband noise time domain data n with flat frequency domain is generated by using Gaussian noise. To improve the operation efficiency of FFT and IFFT, the broadband noise spectrum envelope data A n Data length N w The frequency resolution is selected as 2 to the power of m to reduce the complexity of time domain data sampling rate conversion and frequency fine-tuning. f d Select as an integer, the corresponding broadband spectrum data Z k The sampling rate is F w =N w f d .

3. The method for generating a broadband complex background signal based on frequency domain processing according to claim 1, characterized in that: The sampling rate conversion and frequency fine-tuning in step S2 are used to achieve spectral line adaptation of the narrowband spectrum and the broadband noise spectrum, following the following principles: i Narrowband time domain data x i Sampling rate F i Should meet F i= N i f d ,in N i is an integer representing the i The data length of the narrowband FFT is R si The digital signal should satisfy F i >R si , if the bandwidth is B i The analog signal should satisfy F i >B i , the carrier frequency at baseband f ni Should satisfy m=( f wi - f ni ) / f d The value of is an integer.

4. The method for generating a broadband complex background signal based on frequency domain processing according to claim 1, characterized in that: In steps S3 and S5, partially overlapping FFT transformation and segmented splicing of broadband time domain data are performed according to the following principles: Y1) When performing FFT operation on broadband noise time domain data and i-th narrowband time domain data, the overlapping data lengths are L w and L i , the overlap length is greater than 0 and less than the FFT length; Y2) When extracting broadband time domain data, keep z k Data Center N w - L w Length data, remove the first and last L w data, when L w When it is an even number, remove the first and last L w / 2 data; when L w When / 2 is an odd number, remove the first and last ( L w -1) / 2+1 and ( L w -1) / 2, or remove the first and last ( L w -1) / 2 and ( L w -1) / 2+1 data.

5. The method for generating a broadband complex background signal based on frequency domain processing according to claim 1, characterized in that: The merging of the weighted broadband flat noise spectrum and the narrowband spectrum data in step S3 refers to the merging of each narrowband spectrum with its corresponding position in the broadband spectrum ( f wi - f ni ) / f d The data are added to realize the up-conversion operation of the narrowband signal in the frequency domain.

Images