阅读说明：本技术 一种应用于带通采样的三路信号混叠处理方法 (Three-path signal aliasing processing method applied to band-pass sampling ) 是由 王洪梅 王法广 李世银 樊佳恒 姚冲 于 2019-11-13 设计创作，主要内容包括：本发明提供一种应用于带通采样的三路信号混叠处理方法,用于对多频段射频带通信号进行采样、分离处理,对待处理的多频段射频带通信号通过采样流进行三阶带通采样,设置采样频率,使得采样后在同一频域内允许三个信号发生混叠；在采样流间引入延时差,形成具有相位差的采样后信号,根据相位差设计抗混叠滤波器,从而有效消除由其他两种信号引起的混叠现象。通过采用该方法,使用固定的采样频率就可以实现任意位信号的无混叠接收,在接收过程中不需要针对不同位置信号频繁的变更采样频率,简化了模拟前端。(The invention provides a three-path signal aliasing processing method applied to band-pass sampling, which is used for sampling and separating multi-band radio frequency band signals, carrying out three-order band-pass sampling on the multi-band radio frequency band signals to be processed through a sampling stream, and setting sampling frequency, so that three signals are allowed to be aliased in the same frequency domain after sampling; delay difference is introduced between sampling streams to form sampled signals with phase difference, and an anti-aliasing filter is designed according to the phase difference, so that aliasing caused by other two signals is effectively eliminated. By adopting the method, the non-aliasing reception of any bit signal can be realized by using the fixed sampling frequency, the sampling frequency does not need to be changed frequently aiming at signals at different positions in the receiving process, and the analog front end is simplified.)

1. A three-path signal aliasing processing method applied to band-pass sampling is characterized in that the method simultaneously receives a plurality of radio frequency band signals on different frequency bands, carries out sampling and post-separation processing on the multi-frequency band signals, and allows the frequency spectrums of the three paths of band-pass signals to be overlapped in the same frequency band after sampling; the method comprises the following steps:

(1) carrying out third-order band-pass sampling on the multi-band radio frequency band communication signal to be processed through a sampling stream, and setting the sampling frequency, so that only three signals are allowed to be subjected to aliasing in the same frequency domain after sampling; introducing delay difference between the first sampling streams to form three paths of sampled signals with phase difference;

(2) if the sampled signals are not subjected to aliasing, performing down-conversion on the sampled signals to convert the sampled signals into baseband signals, and separating each signal in the multi-band-pass signals one by one; if three signals are subjected to aliasing in the same frequency domain after sampling, entering the step (3);

(3) anti-aliasing filters are designed for three signals where aliasing occurs in the same frequency domain:

defining three radio frequency signals S₁，S₂And S₃The aliasing after the third-order bandpass sampling respectively has frequency spectrums R¹(f)，R²(f) And R³(f) The three sampled channels are denoted as channel a, channel B and channel C, and the frequency spectrum of the three signals after sampling by the first sample stream is R_A(f) The frequency spectrum after sampling the second sample stream is R_B(f) The frequency spectrum after sampling the third sampling flow is R_C(f)。R_A(f)，R_B(f) And R_C(f) Satisfies the relationship:

in the formula, n₁，n₂And n₃Are respectively S₁，S₂And S₃A position index value in a frequency region;

anti-aliasing filters S are respectively designed in the three channels_A(f)，S_B(f) And S_C(f) The spectrum of the recovered signal is:

to eliminate the signal S₂And S₃The anti-aliasing filter should satisfy:

and is

The negative and positive spectra are symmetric, and we only discuss the positive spectrum here; the following can be obtained:

(4) passing the three aliasing signals through S_A(f)，S_B(f) And S_C(f) Filtering, namely combining time domain data as a coefficient of an FIR filter to realize signal separation;

(5) and (4) after anti-aliasing processing is carried out on the three signals, down-conversion processing is respectively carried out on the three separated signals, and the three separated signals are converted into baseband signals.

2. The method of claim 1, wherein the multi-band bandpass signal is: a plurality of bandpass signals whose spectra do not overlap with each other before sampling.

3. The method of claim 2, wherein assuming that the bandwidth of the main bandpass signal R (f) is limited to B, the sampling rate is f_s2B. Any signal in a frequency bin of index n, denoted by n, is aliased into the first nyquist zone, where the first nyquist zone is the frequency bin | f | < B with index zero. Here, the bandwidth B means a processing bandwidth, and the sampling frequency of the third-order bandpass sampling in the step (1) satisfies the following condition:

(n-1/2)f_s＜|f|＜(n+1/2)f_s

i.e. there may be more than one signal in this frequency band.

4. The method of claim 2The three-path signal aliasing processing method applied to band-pass sampling is characterized in that S in the step (3)_A(f)、S_B(f) And S_C(f) Are respectively:

Technical Field

The invention relates to the field of software radio signal receiving, in particular to a three-path signal aliasing processing method applied to band-pass sampling.

Background

Software defined radio has been widely used in wireless communication, especially 4G communication, as a method and means for implementing wireless communication. The core idea of the software radio is to realize the application software with functions as much as possible and simplify the analog part as much as possible, so the processing requirement of the software radio on the radio frequency signal has higher sampling rate and precision, the application of the band-pass sampling theorem can greatly reduce the required radio frequency sampling rate, and lays the foundation for the subsequent real-time processing. The bandpass sampling theory was originally proposed by r.g. vaughan in 1991, and in recent years, due to the development of AD sampling technology, the sampling rate and accuracy are continuously improved, and the bandpass sampling becomes a powerful theoretical support for realizing software radio. However, the conventional software radio platform applies a bandpass sampling theory a little, taking a conventional software radio universal peripheral (USRP) as an example, a zero intermediate frequency sampling mode is adopted, and the hardware design part of the sampler is complex. And the digital signal processing part only performs the traditional digital down-conversion and other works, and does not perform the anti-aliasing filter design, so that the multi-band signal with aliasing cannot be received, and the universality of software radio is limited.

Software radio was originally proposed to solve the problem of intercommunication of military communication during three-military combined operations, and has been developed from the military field to various fields such as electronic warfare, radar, satellite and modern mobile. In military and commercial applications, it is often necessary to process multiple radio frequency signals on different frequency bands simultaneously, and selecting an appropriate sampling frequency is a difficulty in receiving multi-band signals. Most scholars consider mainly selecting as low a sampling frequency as possible to reduce the burden of back-end digital processing on the premise of no aliasing of frequency spectrum when processing multi-band signals. Many scholars also strive to find new algorithms to simplify the frequency selection process, however, these methods inevitably limit the selection of sampling frequency to avoid aliasing, and the complicated calculation process also increases the difficulty of implementation, and the lower the sampling frequency, the higher the sampling precision is required for the pre-analog RF band-pass filter, so that the pursuit of low sampling frequency is not the optimal method. With the development of hardware such as an ADC and the enhancement of computer processing capacity, higher sampling frequency can be tolerated.

Disclosure of Invention

The purpose of the invention is as follows: in order to solve the technical problem, the invention provides a three-path signal aliasing processing method applied to band-pass sampling, which meets the requirement of software radio on aliasing-free reception of multi-band signals. Specifically, three-order bandpass sampling with adjustable time delay is used for receiving three paths of aliasing signals.

The technical scheme is as follows: in order to achieve the technical effects, the technical scheme provided by the invention is as follows: a three-path signal aliasing processing method applied to band-pass sampling is characterized in that a plurality of radio frequency band signals on different frequency bands are received simultaneously, multi-band signals are subjected to sampling first and then separation processing, and frequency spectrums of three band-pass signals are allowed to be overlapped in the same frequency band after sampling;

the method comprises the following steps:

(1) carrying out third-order band-pass sampling on the multi-band radio frequency band communication signal to be processed through a sampling stream, and setting the sampling frequency to allow three paths of signals to be subjected to aliasing in the same frequency domain after sampling;

(2) if the sampled signals are not subjected to aliasing, performing down-conversion on the sampled signals to convert the sampled signals into baseband signals, and separating each signal in the multi-band-pass signals one by one; if three paths of signals are subjected to aliasing in the same frequency domain after sampling, entering the step (3);

(3) anti-aliasing filters are designed for three signals where aliasing occurs in the same frequency domain:

defining three radio frequency signals S₁，S₂And S₃The aliasing after the third-order bandpass sampling respectively has frequency spectrums R¹(f)，R²(f) And R³(f) The three sampled channels are denoted as channel a, channel B and channel C, and the frequency spectrum of the three signals after sampling by the first sample stream is R_A(f) The frequency spectrum after sampling the second sample stream is R_B(f) The frequency spectrum after sampling the third sampling flow is R_C(f)。R_A(f)，R_B(f) And R_C(f) Satisfies the relationship:

in the formula, n₁，n₂And n₃Are respectively S₁，S₂And S₃A position index value in a frequency region;

anti-aliasing filters S are respectively designed in the three channels_A(f)，S_B(f) And S_C(f) The spectrum of the recovered signal is:

to eliminate the signal S₂And S₃The anti-aliasing filter should satisfy:

and is

The negative and positive spectra are symmetrical and we only discuss the positive spectrum here. According to (2) (3), the equations (5) (6) can be simplified as follows:

(4) passing the three aliasing signals through S_A(f)，S_B(f) And S_C(f) Filtering, namely combining time domain data as a coefficient of an FIR filter to realize signal separation;

(5) and (4) after anti-aliasing processing is carried out on the three signals, down-conversion processing is respectively carried out on the three separated signals, and the three separated signals are converted into baseband signals.

Specifically, the multi-band-pass signal is: a plurality of bandpass signals whose spectra do not overlap with each other before sampling.

Further, assume that the bandwidth of the main band-pass signal R (f) is limited to B and the sampling rate is f_s2B. Any signal in a frequency bin of index n, denoted by n, is aliased into the first nyquist zone, where the first nyquist zone is the frequency bin | f | < B with index zero. Note that the bandwidth B means a processing bandwidth, and the sampling frequency of the third-order bandpass sampling in the step (1) satisfies the following condition:

(n-1/2)f_s＜|f|＜(n+1/2)f_s(8)

i.e. there may be more than one signal in this frequency band.

Further, S in the step (3)_A(f)、S_B(f) And S_C(f) The expression of (a) is:

here, the first and second liquid crystal display panels are,

has the advantages that: compared with the existing software radio signal receiving technology, the invention has the following advantages: due to the adoption of a third-order band-pass sampling technology, the non-aliasing reception of the multi-band signal in the software radio is realized. The fixed sampling frequency is used to realize the non-aliasing reception of any bit signal, the sampling frequency does not need to be changed frequently aiming at signals at different positions in the receiving process, and the analog front end is simplified. The signal separation can be realized by only adjusting the parameters of the anti-aliasing filter in real time aiming at signals at different positions, hardware does not need to be changed, and the universality and the flexibility of software radio are improved.

Drawings

FIG. 1 is an internal schematic diagram of an embodiment of the present invention;

FIG. 2 is a block diagram of functional modules of a software radio signal receiving apparatus according to an embodiment of the present invention;

FIG. 3 is a signal spectrum diagram of a third order bandpass sampling channel A according to an embodiment of the present invention;

FIG. 4 is a signal spectrum diagram of a third order bandpass sampling channel B according to an embodiment of the present invention;

FIG. 5 is a signal spectrum diagram of a third order bandpass sampling channel C according to an embodiment of the present invention;

Detailed Description

Let the radio frequency band pass signal to be sampled be r (f), and its bandwidth be B. Using the sampling frequency f_sIs f_s2B. All signals in the following frequency regions are defined as signals with an index n, which we define as the position index:

(n-1/2)f_s＜|f|＜(n+1/2)f_s

after bandpass sampling, all signals with index n are mapped into the frequency range-B < f < B. According to the band-pass sampling principle, the band-pass signal at any position can be recovered within 0 < f < B. However, for multi-bandpass signals at different index positions, overlap may occur within 0 < f < B after bandpass sampling. In the prior art, when a multi-band signal is processed, the sampling frequency is selected as low as possible under the premise that the frequency spectrum is not subjected to aliasing so as to reduce the burden of back-end digital processing. Many scholars also aim to find new algorithms to simplify the frequency selection process, however, on the premise of avoiding aliasing, the methods inevitably limit the selection of sampling frequency, and meanwhile, the complicated calculation process also increases the difficulty of implementation, and the lower the sampling frequency, the higher the requirement of sampling precision on the preposed analog RF band-pass filter.

In order to solve the above technical problems, the present invention provides a software radio multi-band signal receiving method, which receives a plurality of radio frequency band signals on different frequency bands at the same time, performs sampling and post-separation processing on the multi-band signals, allows the frequency spectrums of three band-pass signals to overlap in the same frequency band after sampling, and can reduce the burden of an analog front end, the method including the steps of:

(1) carrying out third-order band-pass sampling on the multi-band radio frequency band communication signal to be processed through a sampling stream, and setting the sampling frequency to allow three signals to be subjected to aliasing in the same frequency domain after sampling;

(2) if the sampled signals are not subjected to aliasing, performing down-conversion on the sampled signals to convert the sampled signals into baseband signals, and separating each signal in the multi-band-pass signals one by one; if two signals are subjected to aliasing in the same frequency domain after sampling, entering the step (3);

(3) designing an anti-aliasing filter aiming at three signals with aliasing in the same frequency domain:

defining three radio frequency signals S₁，S₂And S₃The aliasing after the third-order bandpass sampling respectively has frequency spectrums R¹(f)，R²(f) And R³(f) The three sampled channels are denoted as channel a, channel B and channel C, and the frequency spectrum of the three signals after sampling by the first sample stream is R_A(f) The frequency spectrum after sampling the second sample stream is R_B(f) The frequency spectrum after sampling the third sampling flow is R_C(f)。R_A(f)，R_B(f) And R_C(f) Satisfies the relationship:

in the formula, n₁，n₂And n₃Are respectively S₁，S₂And S₃A position index value in a frequency region;

anti-aliasing filters S are respectively designed in the three channels_A(f)，S_B(f) And S_C(f) The spectrum of the recovered signal is:

to eliminate the signal S₂And S₃The anti-aliasing filter should satisfy:

and is

The negative and positive spectra are symmetrical and we only discuss the positive spectrum here. The following can be obtained:

(4) passing the three aliasing signals through S_A(f)，S_B(f) And S_C(f) Filtering, namely combining time domain data as a coefficient of an FIR filter to realize signal separation;

(5) and (4) after anti-aliasing processing is carried out on the three signals, down-conversion processing is respectively carried out on the three separated signals, and the three separated signals are converted into baseband signals.

In order to fully and clearly explain the technical scheme of the invention, the invention is further described by combining the specific embodiment and the attached drawings.