A method for digital separation of near frequency signals and a near frequency separation wideband digital receiver

Abstract

The application discloses a near frequency signal digital separation method and a near frequency separation wideband digital receiver, and the method comprises the following steps: a radio frequency antenna signal is directly received and sampled by an analog-to-digital converter to obtain non-square data; the signal is received and sampled by another analog-to-digital converter after adaptive gain control and square processing to obtain square data; the Gaussian white noise contained in the non-square data and the square data is respectively subjected to noise effect distribution evaluation, and the energy detection adaptive threshold of the non-square data and the square data is respectively obtained in combination with a false alarm rate; and the FFT spectrum of the non-square data and the FFT spectrum of the square data are analyzed according to the energy detection adaptive threshold in combination with a window function, so that the separation of two near frequency signals is completed. Through the introduction of the adaptive gain control technology and the adaptive threshold technology, the application can accurately and timely detect and separate two signals with very close frequencies, and greatly improves the detection accuracy of the two signals.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of wideband digital receiver signal processing, and particularly relates to a near frequency signal digital separation method and a near frequency separation wideband digital receiver. BACKGROUND

[0002] With the rapid development and wide application of communication products, the wideband digital receiver plays an unprecedented important role in signal detection and tracking due to its long-time data acquisition capability and the use of progressive flexible digital technology to extract signal information. In the application of the receiver, not only is it necessary for the digital receiver to quickly and real-timely measure the carrier frequency of the radar signal in a very wide frequency band, but also it is necessary for the digital receiver to have the performance of high probability of interception, high frequency resolution and high instantaneous dynamic range (IDR).

[0003] However, the existing wideband digital receiver has difficulty in accurately detecting the near frequency signal under high instantaneous dynamic range, and in the case of not knowing the number of received signals, the receiver may also miss the signal or even report an error signal after normal receiving processing. SUMMARY

[0004] In view of the problem that the existing wideband digital receiver has difficulty in accurately detecting the near frequency signal under high instantaneous dynamic range, the application provides a near frequency signal digital separation method and a near frequency separation wideband digital receiver. By introducing adaptive gain control technology and adaptive threshold technology, the application can accurately and real-timely detect and separate two signals with very close frequencies, greatly improving the detection accuracy of the two signals.

[0005] The application is implemented by the following technical solutions:

[0006] A near frequency signal digital separation method, which comprises the following steps:

[0007] A radio frequency antenna signal is directly received and sampled by an analog-to-digital converter to obtain a set of non-square data; the radio frequency antenna signal is received and sampled by another analog-to-digital converter after adaptive gain control and square processing to obtain a set of square data;

[0008] The Gaussian white noise contained in the non-square data and the square data is respectively evaluated for noise effect distribution to obtain two sets of Rayleigh distribution of the Gaussian white noise FFT, and the energy detection adaptive threshold of the non-square data and the square data is respectively obtained in combination with the false alarm rate;

[0009] According to the energy detection adaptive threshold of the non-square data, in combination with a window function, the frequency point with a peak value greater than the energy detection adaptive threshold in the FFT spectrum of the non-square data is analyzed to complete the separation of the first near frequency signal;

[0010] According to the energy detection adaptive threshold of the square data, a frequency point with a peak value greater than the energy detection adaptive threshold in the FFT spectrum of the square data is analyzed, and compared with the FFT spectrum of the non-square data, so as to complete separation of a second near-frequency signal.

[0011] Compared with a conventional wideband digital receiver, it is difficult to accurately detect a near-frequency signal under a high instantaneous dynamic range, the application first acquires a combined signal through setting two groups of ADCs, and performs gain control and squaring processing at the front end of one group of ADCs, so as to ensure reliable sampling of a weak signal and avoid missing detection and wrong detection, meanwhile, the application introduces an adaptive threshold technology, so as to accurately detect two signals with a very close frequency, especially when a weak signal is covered by a main lobe of a strong signal, the main lobe effect can be effectively eliminated, and the detection accuracy and reliability of two near-frequency signals are greatly improved, and the implementation of the application is simple.

[0012] As a preferred embodiment, the acquisition process of the square data of the application comprises:

[0013] The amplitude intensity of the radio frequency antenna signal is detected, and adaptive gain control is performed on the radio frequency antenna signal according to the detection result; the radio frequency antenna signal is defined as a combined signal composed of two signals added with Gaussian white noise, and the frequencies of the two signals are and ;

[0014] The signal after adaptive gain control is subjected to squaring operation, and the frequency components of the squared signal are , , , , and ;

[0015] The analog signal after squaring operation is received and sampled by an analog-to-digital converter, so as to obtain the square data.

[0016] Considering that when the amplitudes of the two signals are small, the amplitude of the signal after squaring processing is smaller, at this time, the analog-to-digital converter may not accurately sample the squared signal, therefore, the application adds adaptive gain control amplification at the back end of the antenna for amplifying the received signal, so as to improve the accuracy and reliability of data acquisition.

[0017] As a preferred embodiment, the application performs adaptive gain control on the radio frequency antenna signal according to the detection result, and specifically comprises:

[0018] If the sum of the amplitudes of the two signals is less than or equal to 0.01V, the gain is set to 50;

[0019] If the sum of the amplitudes of the two signals is greater than 0.01V and less than or equal to 0.05V, then the gain is set to 10.

[0020] If the sum of the amplitudes of the two signals is greater than 0.05V and less than or equal to 0.1V, then the gain is set to 5;

[0021] If the sum of the amplitudes of the two signals is greater than 0.1V, no gain amplification is performed, and the gain is set to 1.

[0022] In a preferred embodiment, the noise effect distribution evaluation process of the present invention specifically includes:

[0023] The Gaussian white noise signal in the data sampled by the analog-to-digital converter is windowed;

[0024] Perform an FFT transform on the windowed signal to obtain the first noise distribution result;

[0025] The first noise distribution result is stored, and the process is returned to the windowing step for a second loop until the prediction count is reached, at which point the Rayleigh distribution of the Gaussian white noise FFT is obtained.

[0026] In a preferred embodiment, the energy detection adaptive threshold acquisition process of the present invention includes:

[0027] The desired false alarm rate is obtained by comparing the area from the threshold to infinity on the horizontal axis to the area from 0 to infinity on the horizontal axis in the Rayleigh distribution of the Gaussian white noise FFT.

[0028] Based on the Rayleigh distribution results of the Gaussian white noise FFT, the false alarm rate is determined. Thus, the adaptive threshold for energy detection is calculated.

[0029] In a preferred embodiment, the separation process of the first near-frequency signal of the present invention includes:

[0030] The non-square data is subjected to windowing and FFT transformation sequentially to obtain the FFT spectrum of the non-square data;

[0031] The fine portion of the first near-frequency signal frequency is obtained from the FFT spectrum of the non-square data;

[0032] Analyze the points in the FFT spectrum of the non-squared data where the peak value is greater than its energy detection adaptive threshold, find the frequency of the point with the largest peak value, and sum it with the fine part to obtain the first near-frequency signal.

[0033] In a preferred embodiment, the process for obtaining the fine portion of the first near-frequency signal frequency of the present invention includes:

[0034] Take an integer frequency point from the FFT spectrum of the non-square data, and denot its amplitude as . Take another one with Adjacent and located On the other side, the amplitude is higher The frequency point is denoted as its amplitude. ;

[0035] definition arrive The distance between them is , The location is ,therefore and The amplitude of the position can be expressed as:

[0036]

[0037]

[0038] and The ratios are as follows:

[0039]

[0040] According to the above formula, we get The expression is as follows:

[0041]

[0042] Therefore, fine frequency It can be represented as

[0043]

[0044] in, for The indicators This represents the frequency resolution of the FFT.

[0045] In a preferred embodiment, the separation process of the second near-frequency signal of the present invention includes:

[0046] Perform an FFT transform on the squared data to obtain the FFT spectrum of the squared data;

[0047] Analyze the points in the FFT spectrum of the squared data where the peak value is greater than its energy detection adaptive threshold, find the two points with the largest peak values, and compare them with the FFT spectrum of the non-squared data to separate the second near-frequency signal.

[0048] In a preferred embodiment, the present invention analyzes the points in the FFT spectrum of the squared data whose peak values ​​are greater than its energy detection adaptive threshold, finds the two points with the largest peak values, and compares them with the FFT spectrum of the non-squared data to separate the second near-frequency signal, specifically including:

[0049] Find the points in the FFT spectrum of the squared data that are greater than its energy detection adaptive threshold, and denote the frequency of the point with the largest peak as . The frequency of the second peak is denoted as ;

[0050] Then the point frequency with the largest peak and , , and Comparison, if the point frequency If it equals one of the four values, then the analysis of the second largest peak is performed; if it does not equal one of the four values, then the analysis of whether it equals the frequency resolution of the FFT is performed. The specific process includes:

[0051] 1) If the frequency of this point is equal to the frequency resolution of the FFT, then based on the FFT spectrum of the non-square data, determine the relative position of the second near-frequency signal and the first near-frequency signal, and then determine the frequency of the second near-frequency signal based on this relative position:

[0052]

[0053] 2) If the frequency at this point is not equal to the frequency resolution of the FFT, then define the difference between the first and second near-frequency signals. Sum And twice the second signal If one of the following three conditions is met: It is a local peak in the FFT spectrum of the non-squared data; the second larger peak or ; or The second near-frequency signal can be obtained by identifying local peaks in the FFT spectrum of the squared data that are above the adaptive threshold.

[0054] ;

[0055] If the frequency at that point equal , , and At one of these times, continue with the second, larger peak. Perform the analysis; similarly, compare it with... , , and Comparison, if If it equals one of them, then we can directly obtain ;

[0056] if If one of the conditions is not met, proceed according to 1)-2). Determining whether it equals the frequency resolution of the FFT.

[0057] On the other hand, the present invention proposes a near-frequency separation broadband digital receiver, the digital receiver comprising: an adaptive gain controller, a squarer, two analog-to-digital converters and a signal detection unit;

[0058] The adaptive gain controller automatically performs adaptive gain control on the radio frequency antenna signal based on the strength of the radio frequency antenna signal;

[0059] The squarer performs square processing on the signal output by the adaptive gain controller;

[0060] One of the analog-to-digital converters directly receives the radio frequency antenna signal and samples it to obtain a set of non-square data.

[0061] Another analog-to-digital converter receives the signal output by the squarer and samples it to obtain a set of squared data;

[0062] The signal detection unit evaluates the noise effect distribution of the Gaussian white noise contained in the non-square data and the square data respectively, obtains the Rayleigh distribution of the two sets of Gaussian white noise FFT, and obtains the energy detection adaptive threshold of the non-square data and the square data respectively by combining the false alarm rate.

[0063] The signal detection unit analyzes the frequency points in the FFT spectrum of the non-square data whose peak values ​​are greater than its energy detection adaptive threshold, based on the energy detection adaptive threshold of the non-square data and a window function, thereby completing the separation of the first near-frequency signal.

[0064] The signal detection unit analyzes the frequency points in the FFT spectrum of the squared data whose peak values ​​are greater than the energy detection adaptive threshold based on the energy detection adaptive threshold of the squared data, and compares them with the FFT spectrum of the non-squared data to complete the separation of the second near-frequency signal.

[0065] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0066] 1. This invention solves the problem that existing broadband digital receivers are unable to accurately detect near-frequency signals under high transient dynamic range by adopting adaptive gain control technology and adaptive threshold technology, thereby improving the detection accuracy and reliability of near-frequency signals;

[0067] 2. This invention uses a two-channel analog-to-digital converter to acquire radio frequency signals, obtaining a set of non-square data and a set of square data. At the same time, it combines gain adaptive control amplification technology to ensure reliable sampling of weaker signals. Then, the first near-frequency signal (stronger signal) is detected from the FFT spectrum of the non-square data, and the second near-frequency signal (weaker signal) is detected from the FFT spectrum of the square data. By comparing the two signals, the invention avoids the situation of false alarms and false negatives caused by setting a threshold, and greatly improves the detection accuracy of the two signals.

[0068] 3. Compared with the multiple threshold settings of traditional methods, this application only needs to set one threshold for the same set of data, namely a single threshold. By comparing the FFT spectra of non-square data and square data, weak signals can be detected, while still maintaining a low false alarm rate and a high dynamic range of dual signal amplitudes.

[0069] 4. By employing a window function, this invention effectively suppresses the side lobes and spurious components of strong signals, reduces FFT leakage, and after adding the window function, the frequency resolution is mainly determined by the main lobe of the windowed signal, which can accurately detect two signals with very close frequencies, breaking the limitation of detecting two near-frequency signals in a single FFT process.

[0070] 5. This invention is implemented using a hardware structure, which greatly satisfies the current requirements of broadband digital receivers for real-time response (hundreds of nanoseconds). Attached Figure Description

[0071] The accompanying drawings, which are included to provide a further understanding of embodiments of the invention and form part of this application, do not constitute a limitation thereof. In the drawings:

[0072] Figure 1 This is a flowchart of a method according to an embodiment of the present invention.

[0073] Figure 2 This is a flowchart of the adaptive gain control according to an embodiment of the present invention.

[0074] Figure 3 This is a flowchart of the AWGN distribution evaluation process according to an embodiment of the present invention.

[0075] Figure 4 The Rayleigh distribution of AWGN FFT under different gain conditions is shown in the embodiments of the present invention.

[0076] Figure 5This is a flowchart illustrating the signal separation process according to an embodiment of the present invention.

[0077] Figure 6 This is a Hanning window spectrum diagram according to an embodiment of the present invention.

[0078] Figure 7 This is a block diagram of the near-frequency separation broadband digital receiver according to an embodiment of the present invention. Detailed Implementation

[0079] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the embodiments and accompanying drawings. The illustrative embodiments and descriptions of the present invention are only used to explain the present invention and are not intended to limit the present invention.

[0080] Example:

[0081] Existing broadband digital receivers struggle to accurately detect near-frequency signals within a high instantaneous dynamic range. Furthermore, they are prone to issues such as missed or false alarms. To address these shortcomings, this embodiment proposes a digital near-frequency signal separation method. This method utilizes a dual-channel analog-to-digital converter and adaptive gain control to acquire radio frequency signals. By comparing signals, it avoids missed or false alarms. Furthermore, by employing adaptive thresholding technology, it can quickly and accurately separate two signals with very similar frequencies, significantly improving the detection accuracy of near-frequency signals.

[0082] Specifically, such as Figure 1 As shown, the method proposed in this embodiment includes the following steps:

[0083] Step 100: The radio frequency antenna signal is directly received and sampled by an analog-to-digital converter to obtain non-square data; the radio frequency antenna signal is then received and sampled by another analog-to-digital converter after gain control and squaring to obtain square data; the two analog-to-digital converters are identical.

[0084] Step 200: Evaluate the noise effect distribution of Gaussian white noise contained in non-square data and square data respectively, obtain Rayleigh distribution of Gaussian white noise FFT for two sets, and obtain the energy detection adaptive threshold for non-square data and square data respectively by combining the false alarm rate.

[0085] Step 300: Based on the energy detection adaptive threshold of the non-square data and combined with the window function, analyze the frequency points in the FFT spectrum of the non-square data whose peak value is greater than its energy detection adaptive threshold, and complete the separation of the first near-frequency signal.

[0086] Step 400: Based on the adaptive threshold for energy detection of the squared data, analyze the frequency points in the FFT spectrum of the squared data whose peak values ​​are greater than the adaptive threshold for energy detection, and compare them with the FFT spectrum of the non-squared data to complete the separation of the second near-frequency signal.

[0087] In one alternative implementation, the radio frequency antenna signal can be defined as a combination of signals with added Gaussian white noise, specifically:

[0088]

[0089] Where A and B are the amplitudes of the two signals, respectively. and These correspond to the frequencies of the two signals, and These are the phases of the two signals, It is Gaussian white noise.

[0090] Based on this, the above signal is directly received and sampled by an analog-to-digital converter, thus obtaining a set of non-square data. The specific steps for obtaining square data include the following:

[0091] Step 101: Perform adaptive gain control on the radio frequency antenna signal based on the amplitude intensity of the radio frequency antenna signal.

[0092] When the amplitudes of two signals are much less than 1, directly squaring the signals will result in amplitudes of the squared signals that are smaller than the original signal amplitudes. This may cause the analog-to-digital converter (ADC) to be unable to accurately sample the squared signals, especially when the amplitude of the squared signal is less than the ADC's 1-bit resolution. Therefore, this embodiment adds an adaptive gain controller to the back end of the antenna to adaptively control the gain of the RF antenna signal. According to the working principle of the ADC, the amplitude of the frequency component of the input signal that can be sampled can only be within the dynamic range of the ADC. Therefore, this embodiment completes the design of adaptive gain control based on the amplitude intensity of the input signal. The specific operation is as follows: First, peak detection is performed. By detecting the sum of the amplitudes of the RF antenna signals, three adaptive amplitude levels of 0.01V, 0.05V, and 0.1V are set. Then, according to the different amplitude levels, four adaptive gain control levels of 50, 10, 5, and 1 are set to achieve precise and controllable gain of the combined signal.

[0093] Specifically, the adaptive gain control process in this embodiment is as follows: Figure 2 As shown, it includes:

[0094] If the sum of the amplitudes of the two signals is less than or equal to 0.01V, then the gain is set to 50.

[0095] If the sum of the amplitudes of the two signals is greater than 0.01V and less than or equal to 0.05V, then the gain is set to 10.

[0096] If the sum of the amplitudes of the two signals is greater than 0.05V and less than or equal to 0.1V, then the gain is set to 5;

[0097] If the sum of the amplitudes of the two signals is greater than 0.1V, no gain amplification is performed, and the gain is set to 1.

[0098] For example, in this embodiment, if the sum of the amplitudes of the two signals is set between 0.05V and 0.1V, the gain in the adaptive gain controller is 5.

[0099] Step 102: Squaring the signal after adaptive gain control yields a set of squared data, namely:

[0100]

[0101]

[0102]

[0103] Wherein, the squared signal frequency component is , , , , and When the Gaussian white noise energy is very small, only a few frequency components of the analog signal after passing through the squarer can be detected. , , and .

[0104] In an optional implementation, step 200 further includes the following sub-steps:

[0105] Step 201: Evaluate the noise effect distribution of the Gaussian white noise contained in the two sampled signals (i.e., non-square data and square data), such as... Figure 3 As shown, the specific operation is as follows:

[0106] Use a window function to window the Gaussian white noise signal in the data sampled by the ADC;

[0107] A 256-point FFT transformation was performed on the windowed data to obtain the first noise distribution result;

[0108] The first noise distribution result is stored, and the process returns to the windowing step for the next iteration. This loop continues until a preset number of iterations is reached (e.g., if the total length of the ADC sampling data is 131072, then each 256-point FFT requires 512 iterations). At this point, the Rayleigh distribution of the Gaussian white noise FFT is obtained, as shown below. Figure 4 As shown.

[0109] Step 202: Based on the Rayleigh distribution results of the Gaussian white noise FFT, the expected false alarm rate is obtained, which is expressed as:

[0110]

[0111] in, It is an adaptive threshold. It is the result of FFT. yes The corresponding distribution Depend on The area from the threshold to infinity The ratio of the areas from 0 to infinity is determined by this.

[0112] Step 203: Based on the Rayleigh distribution of Gaussian white noise, determine the false alarm rate. for Therefore, from the false alarm rate expression, the adaptive threshold thr can be obtained, and its energy expression is as follows:

[0113]

[0114] It should be noted that: using the above process, the adaptive thresholds for energy detection of the two sets of data, non-square data and square data, can be obtained respectively. Specifically, step 201 can process the Gaussian white noise (i.e., the original white noise) in the non-square data and the Gaussian white noise (i.e., the white noise amplified and squared by adaptive gain control) in the square data to obtain the Rayleigh distribution curves of the two sets of Gaussian white noise. Then, through steps 202 and 203, the adaptive thresholds for energy detection of the two sets of data can be obtained.

[0115] An optional implementation method is as follows: Figure 5 As shown, step 300 includes the following sub-steps:

[0116] Step 301: Windowing and FFT transformation are sequentially performed on the non-square data obtained after ADC sampling to obtain the FFT spectrum of the non-square data, such as... Figure 6 As shown.

[0117] In this embodiment, the Hanning window is used for windowing operations. The Hanning window function is expressed as follows:

[0118]

[0119] Where T represents the window width or the time of the window function. Based on the Hanning window function, its Fourier transform form is:

[0120]

[0121] in, .

[0122] Step 302: Obtain the fine portion of the first near-frequency signal frequency from the FFT spectrum of the non-square data.

[0123] Specifically, an integer frequency point is selected from the FFT spectrum of the non-square data, and its amplitude is denoted as... Take another one with Adjacent and located On the other side, and the amplitude is lower The frequency point is denoted as its amplitude. Then define arrive The distance between them is (Frequency difference) The location is ,therefore and The amplitude of the position can be expressed as:

[0124]

[0125]

[0126] and The ratios are as follows:

[0127]

[0128] Based on the above formula, we can obtain The expression is as follows:

[0129]

[0130] Define the frequency resolution of the FFT 5MHz, fine frequency It can be represented as

[0131]

[0132] in, for The indicators.

[0133] Step 303: Analyze the points in the FFT spectrum of the non-square data where the peak value is greater than its energy detection adaptive threshold, find the point with the largest peak value, and sum it with the fine part to obtain the accurate first near-frequency signal (stronger signal).

[0134] Specifically, from the FFT spectrum of the non-squared data, find the points that are greater than the adaptive energy detection threshold, and record the frequency of the point with the largest peak as . The frequency of the point with the second highest peak value that is greater than the adaptive threshold is denoted as... (When the frequencies of two signals are very close, only the point of maximum peak can be detected). Then... With fine details Summing the frequencies yields the accurate value of the first near-frequency signal (stronger signal).

[0135] .

[0136] Specifically, such as Figure 5 As shown, step 300 also includes the following sub-steps:

[0137] Step 401: Perform an FFT transform on the squared data to obtain the FFT spectrum of the squared data.

[0138] Step 402: Analyze the points in the FFT spectrum of the squared data where the peak value is greater than its energy detection adaptive threshold, find the two points with the largest peak values, and compare them with the FFT result of the non-squared data to separate the second near-frequency signal (weaker signal).

[0139] Specifically, first, find the points in the FFT spectrum of the squared data that are greater than its energy detection adaptive threshold, and denote the frequency of the point with the largest peak as . The frequency of the second peak is denoted as Then, the frequency of the point with the largest peak value. and , , and If the value equals one of the four values, the analysis proceeds to the second larger peak. If it does not equal one of the four values, the analysis continues to determine if the value equals the frequency resolution of the FFT (5MHz). The specific process is as follows:

[0140] 1) This means that the frequencies of the two signals in the combined signal are very close, and then the point of maximum peak value is analyzed based on the FFT plot of the non-square data. The left nearest point ( ), right nearest point ( The amplitude of the left neighboring point is greater than the amplitude of the right neighboring point, i.e. exist On the left, and vice versa, on the right. Therefore, the frequency of the second signal is as follows:

[0141]

[0142] 2) At this point, the difference between the two near-frequency signals in the combined signal is defined. Sum And twice the second signal Therefore, if one of the following three conditions is met:

[0143] It is a local peak in the non-squared data FFT; the second larger peak. or ; or It refers to local peaks in squared data FFT that are above the adaptive threshold.

[0144] Based on this, a second near-frequency signal can be obtained, represented as follows:

[0145] .

[0146] if equal , , and At one of these times, continue with the second, larger peak. Let's analyze it. Similarly, let's compare it with... , , and In comparison, If equal to one of them, If it equals one of them, then we can directly obtain .if If one of the conditions is not met, proceed according to steps 1)-2) above. The determination of whether it is equal to 5MHz completes the separation of the second near-frequency signal (weaker signal).

[0147] This embodiment also proposes a near-frequency separation broadband digital receiver, specifically as follows: Figure 7 As shown, the digital receiver includes: an adaptive gain controller, a squarer, two analog-to-digital converters (ADC1 and ADC2), and a signal detection unit.

[0148] The adaptive gain controller automatically performs adaptive gain control on the signal based on the strength of the signal received by the RF antenna. The adaptive gain controller mainly includes a peak detection unit, which is used to select the appropriate gain level based on the detection of the peak value of the received signal.

[0149] The squarer is used to square the signal after adaptive gain control.

[0150] An analog-to-digital converter is used to directly receive and sample radio frequency antenna signals to obtain a set of non-square data.

[0151] Another analog-to-digital converter is used to receive the squared signal and sample it to obtain a set of squared data;

[0152] The signal detection unit evaluates the noise effect distribution of Gaussian white noise contained in non-square data and square data respectively, obtains Rayleigh distribution of two sets of Gaussian white noise FFT, and obtains the energy detection adaptive threshold of non-square data and square data respectively by combining the false alarm rate.

[0153] The signal detection unit analyzes the frequency points in the FFT spectrum of non-square data whose peak values ​​are greater than the adaptive threshold of energy detection based on the energy detection threshold of the non-square data and the window function, thereby completing the separation of the first near-frequency signal (stronger signal).

[0154] The signal detection unit analyzes the frequency points in the FFT spectrum of the squared data whose peak values ​​are greater than the adaptive threshold of its energy detection based on the energy detection threshold of the squared data, and compares them with the FFT spectrum of the non-squared data to complete the separation of the second near-frequency signal (weaker signal).

[0155] The signal detection unit in this embodiment can be based on a programmable logic device, such as an FPGA, to implement windowing operations, FFT transformations, and computational analysis.

[0156] The near-frequency signal digital separation technology proposed in this embodiment solves the problem that existing broadband data receivers have difficulty accurately detecting near-frequency signals under high instantaneous dynamic range.

[0157] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0158] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0159] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0160] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0161] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for digital separation of near-frequency signals, characterized in that, The method includes: The radio frequency antenna signal is directly received and sampled by an analog-to-digital converter to obtain a set of non-square data; the radio frequency antenna signal is then received and sampled by another analog-to-digital converter after adaptive gain control and squaring to obtain a set of square data. The noise effect distribution of Gaussian white noise contained in the non-square data and the square data is evaluated respectively to obtain the Rayleigh distribution of the two sets of Gaussian white noise FFT, and the energy detection adaptive threshold of the non-square data and the square data is obtained by combining the false alarm rate. Based on the energy detection adaptive threshold of the non-square data, and combined with the window function, analyze the frequency points in the FFT spectrum of the non-square data whose peak value is greater than its energy detection adaptive threshold, and complete the separation of the first near-frequency signal; Based on the energy detection adaptive threshold of the squared data, analyze the frequency points in the FFT spectrum of the squared data where the peak value is greater than its energy detection adaptive threshold, and compare them with the FFT spectrum of the non-squared data to complete the separation of the second near-frequency signal; The noise effect distribution evaluation process specifically includes: The Gaussian white noise signal in the data sampled by the analog-to-digital converter is windowed; Perform an FFT transform on the windowed signal to obtain the first noise distribution result; The first noise distribution result is stored, and the process is returned to the windowing step for a second loop until the prediction number is reached, at which point the Rayleigh distribution of the Gaussian white noise FFT is obtained. The energy detection adaptive threshold acquisition process includes: The desired false alarm rate is obtained by comparing the area from the threshold to infinity on the horizontal axis to the area from 0 to infinity on the horizontal axis in the Rayleigh distribution of the Gaussian white noise FFT. Based on the Rayleigh distribution results of the Gaussian white noise FFT, the false alarm rate is determined. Thus, the adaptive threshold for energy detection is calculated.

2. The near-frequency signal digital separation method according to claim 1, characterized in that, The process of obtaining the squared data includes: The amplitude intensity of the radio frequency antenna signal is detected, and adaptive gain control is performed on the radio frequency antenna signal based on the detection result; the radio frequency antenna signal is defined as a combined signal composed of two signals with added Gaussian white noise, the frequencies of which are respectively... and ; The frequency components of the signal after adaptive gain control are squared as follows: , , , , and ; The analog signal after squaring is received and sampled by an analog-to-digital converter to obtain the squared data.

3. The near-frequency signal digital separation method according to claim 2, characterized in that, Based on the detection results, adaptive gain control is performed on the radio frequency antenna signal, specifically including: If the sum of the amplitudes of the two signals is less than or equal to 0.01V, then the gain is set to 50. If the sum of the amplitudes of the two signals is greater than 0.01V and less than or equal to 0.05V, then the gain is set to 10. If the sum of the amplitudes of the two signals is greater than 0.05V and less than or equal to 0.1V, then the gain is set to 5; If the sum of the amplitudes of the two signals is greater than 0.1V, no gain amplification is performed, and the gain is set to 1.

4. The near-frequency signal digital separation method according to claim 1, characterized in that, The separation process of the first near-frequency signal includes: The non-square data is subjected to windowing and FFT transformation sequentially to obtain the FFT spectrum of the non-square data; The fine portion of the first near-frequency signal frequency is obtained from the FFT spectrum of the non-square data; Analyze the points in the FFT spectrum of the non-squared data where the peak value is greater than its energy detection adaptive threshold, find the frequency of the point with the largest peak value, and sum it with the fine part to obtain the first near-frequency signal.

5. The near-frequency signal digital separation method according to claim 4, characterized in that, The process of obtaining the fine portion of the frequency of the first near-frequency signal includes: Take an integer frequency point from the FFT spectrum of the non-square data, and denot its amplitude as . Take another one with Adjacent and located On the other side, the amplitude is higher The frequency point is denoted as its amplitude. ; definition arrive The distance between them is , The location is ,therefore and The amplitude of the position can be expressed as: ； ； and The ratios are as follows: ； According to the above formula, we get The expression is as follows: ； Therefore, fine frequency It can be represented as ； in, for The indicators This represents the frequency resolution of the FFT.

6. The near-frequency signal digital separation method according to claim 4, characterized in that, The separation process of the second near-frequency signal includes: Perform an FFT transform on the squared data to obtain the FFT spectrum of the squared data; Analyze the points in the FFT spectrum of the squared data where the peak value is greater than its energy detection adaptive threshold, find the two points with the largest peak values, and compare them with the FFT spectrum of the non-squared data to separate the second near-frequency signal.

7. The near-frequency signal digital separation method according to claim 6, characterized in that, Analyze the points in the FFT spectrum of the squared data where the peak value exceeds its energy detection adaptive threshold, find the two points with the largest peak values, and compare them with the FFT spectrum of the non-squared data to separate the second near-frequency signal. Specifically, this includes: Find the points in the FFT spectrum of the squared data that are greater than its energy detection adaptive threshold, and denote the frequency of the point with the largest peak as . The frequency of the second peak is denoted as ; Then the point frequency with the largest peak and , , and Comparison, if the point frequency If it equals one of the four, then the analysis of the second largest peak is performed; if it does not equal one of the four, then the analysis of whether it equals the frequency resolution of the FFT is performed. The specific process includes: 1) If the frequency of this point is equal to the frequency resolution of the FFT, then based on the FFT spectrum of the non-square data, determine the relative position of the second near-frequency signal and the first near-frequency signal, and then determine the frequency of the second near-frequency signal based on this relative position: ； 2) If the frequency at this point is not equal to the frequency resolution of the FFT, then define the difference between the first and second near-frequency signals. Sum and twice the second signal If one of the following three conditions is met: It is a local peak in the FFT spectrum of the non-squared data; the second larger peak or ; or The second near-frequency signal can be obtained by identifying local peaks in the FFT spectrum of the squared data that are above the adaptive threshold. ； If the frequency at that point equal , , and At one of these times, continue with the second, larger peak. Perform the analysis; similarly, compare it with... , , and Comparison, if If it equals one of them, then we can directly obtain ; if If one of the conditions is not met, proceed according to 1)-2). Determining whether it equals the frequency resolution of the FFT.

8. A near-frequency separation broadband digital receiver, characterized in that, The digital receiver includes: an adaptive gain controller, a squarer, two analog-to-digital converters, and a signal detection unit; The adaptive gain controller automatically performs adaptive gain control on the radio frequency antenna signal based on the strength of the radio frequency antenna signal; The squarer performs square processing on the signal output by the adaptive gain controller; One of the analog-to-digital converters directly receives the radio frequency antenna signal and samples it to obtain a set of non-square data. Another analog-to-digital converter receives the signal output by the squarer and samples it to obtain a set of squared data; The signal detection unit evaluates the noise effect distribution of the Gaussian white noise contained in the non-square data and the square data respectively, obtains the Rayleigh distribution of the two sets of Gaussian white noise FFT, and obtains the energy detection adaptive threshold of the non-square data and the square data respectively by combining the false alarm rate. The signal detection unit analyzes the frequency points in the FFT spectrum of the non-square data whose peak values ​​are greater than its energy detection adaptive threshold, based on the energy detection adaptive threshold of the non-square data and a window function, thereby completing the separation of the first near-frequency signal. The signal detection unit analyzes the frequency points in the FFT spectrum of the squared data whose peak values ​​are greater than the energy detection adaptive threshold based on the energy detection adaptive threshold of the squared data, and compares them with the FFT spectrum of the non-squared data to complete the separation of the second near-frequency signal. The noise effect distribution evaluation process specifically includes: The Gaussian white noise signal in the data sampled by the analog-to-digital converter is windowed; Perform an FFT transform on the windowed signal to obtain the first noise distribution result; The first noise distribution result is stored, and the process is returned to the windowing step for a second loop until the prediction number is reached, at which point the Rayleigh distribution of the Gaussian white noise FFT is obtained. The energy detection adaptive threshold acquisition process includes: The desired false alarm rate is obtained by comparing the area from the threshold to infinity on the horizontal axis to the area from 0 to infinity on the horizontal axis in the Rayleigh distribution of the Gaussian white noise FFT. Based on the Rayleigh distribution results of the Gaussian white noise FFT, the false alarm rate is determined. Thus, the adaptive threshold for energy detection is calculated.

Images