Rapid signal pre-sorting implementation method based on multistage assembly line technology

Abstract

The invention provides a rapid signal pre-sorting implementation method based on a multistage pipeline technology, which mainly comprises the following steps of: 1, performing parameter measurement and data dimension reduction on a signal, and extracting frequency information RF and amplitude information PA of the signal; 2, multi-stage assembly lines are constructed for the to-be-processed frequency information RF and amplitude information PA, wherein the number of the assembly lines is M; 3, determining an amplitude threshold according to the amplitude information PA; 4, judging the relevancy of the frequency values in each stage of assembly line, and determining a weight value; 5, calculating the total weight, and judging whether the signals in the time period are effective signals or not according to the total weight; 6, threshold detection: judging whether the signal is a false alarm according to an amplitude threshold; and 7, outputting effective signals greater than the amplitude threshold. And thus, rapid signal pre-sorting is completed. The method has the characteristics of quick system response, simple and efficient logic judgment, hardware resource saving and the like.

Description

Technical Field

[0001] The present invention relates to the technical field of digital signal processing, and further to the application of pipeline technology to realize rapid signal detection. Background Art

[0002] With the complexity of the current radar combat environment and the improvement of electronic countermeasure technology, the current new system radar generally adopts frequency agility technology, which can reduce the probability of being intercepted by reconnaissance aircraft on the one hand, and reduce the interference level of jammers on the other hand. It can be said that frequency agility is a relatively effective anti-reconnaissance and anti-interference means of radar. Therefore, the rapid signal pre-sorting of frequency agile radar is an important prerequisite for reconnaissance and interference of new system radar.

[0003] However, the currently used fast signal pre-sorting technology generally has the disadvantages of complex algorithm engineering implementation and long processing time for high pre-sorting accuracy, or the disadvantages of short processing time algorithm engineering implementation being greatly affected by interference. In addition, the electromagnetic environment of modern battlefields is changing rapidly, and in high-density pulse stream signal environments, multipath and simultaneous arrival increase the difficulty of fast signal pre-sorting technology. Summary of the invention

[0004] In view of the above technical problems, the present invention designs a fast signal pre-sorting implementation method based on multi-stage pipeline technology according to the characteristics of fast signal pre-sorting technology, which can ensure the accuracy of pre-sorting while having extremely high real-time performance and a simple implementation process. The fast signal pre-sorting technology can be applied to the frequency guidance of reconnaissance receivers, electronic warfare jammers and other equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0005] Figure 1 It is a schematic diagram of constructing a multi-stage pipeline according to the present invention.

[0006] Figure 2 It is a flow chart of the implementation method of the present invention. DETAILED DESCRIPTION

[0007] In order to more fully and clearly reflect the purpose and technical features of the present invention, the implementation mode of the present invention is further described in detail below in conjunction with the accompanying drawings of the embodiments of the present invention.

[0008] Before pre-sorting, the radar signal receiver needs to extract features from the received pulse signal to form a pulse description word (PDW). PDW is mainly composed of parameters such as frequency RF, arrival time TOA, arrival angle DOA, pulse width PW, pulse amplitude PA and intra-pulse modulation parameters (PMP). In the fast frequency measurement receiver (IFM), one of the main purposes of signal pre-sorting is to provide fast frequency guidance for subsequent high-precision parameter measurement and signal processing. Therefore, there is no need to form a complete PDW, only frequency information RF and amplitude information PA are required, and the effect of data dimensionality reduction is achieved at the same time.

[0009] The key to the fast signal pre-sorting technology based on multi-stage pipeline is to build a multi-stage pipeline, such as Figure 1 shown.

[0010] The frequency information in a time window is input into the pipeline in chronological order, and the frequency correlation is determined in each level of the pipeline to determine the weight. The weights of each level of the pipeline are added together, and the calculation formula is as follows:

[0011]

[0012] Where K i That is, the relevant weight of the frequency information in each level of the pipeline, and the judgment standard is set to p. When P≤p, the frequency information in this time period is irrelevant or weakly correlated, and it is judged as noise, and the output keeps the last pre-sorting result; when P>p, the frequency information in this time period is strongly correlated, and it is judged as a signal, and the pre-sorting result is updated, and the output is the current signal frequency value and amplitude value, thus completing the main process of signal pre-sorting.

[0013] In order to avoid excessive false alarm rate caused by interference in complex environments, threshold detection should also be performed.

[0014] Reference Figure 1 and Figure 2 The method for implementing fast signal pre-sorting based on multi-stage pipeline technology includes the following steps:

[0015] Step 1: Measure the signal parameters and reduce the data dimension, extract the frequency value RF and amplitude value PA of the signal flow;

[0016] Step 2: Cache the frequency value RF and the amplitude value PA into the register at a delay of one clock cycle, cache M clock cycles in total, and build an M-level pipeline;

[0017] Step 3: Calculate the correlation of the frequency values ​​RF in each level of the pipeline within the same clock cycle: whenever there are two frequency values ​​RF whose difference does not exceed 1.5 times the frequency measurement accuracy of the electronic receiver, the correlation K is recorded. i is 1 / M, i is a natural number;

[0018] Step 4: Sum the correlations in each pipeline level, with a weight of P;

[0019] Step 5: Make a decision on the weight P, and set the judgment criteria p to be 1 / 2 and 3 / 4 respectively;

[0020] When the weight value P≤1 / 2, it is determined to be noise or interference, and the processing flow is terminated directly;

[0021] When 1 / 2<weight P≤3 / 4, go to step 6;

[0022] When the weight P>3 / 4, go to step 7;

[0023] Step 6: Perform threshold detection. If the amplitude value PA is greater than the threshold T, it is determined to be a signal, otherwise it is target noise;

[0024] Step 7: Perform threshold detection. If the amplitude value PA is greater than the threshold T, it is determined as a strong signal, otherwise it is determined as a weak signal;

[0025] Step 8: Output the pre-sorting value, which includes the frequency value RF and amplitude value PA of the target signal, strong signal or weak signal. So far, the fast frequency pre-sorting process has been completed.

[0026] Among them, the amplitude value PA is stored in FIFO in the order of clock cycles, and the storage depth is N (N<M). At the same time, the amplitude value PA is accumulated, and the accumulation result is recorded as SUM. After FIFO stores N PA values, the latest PA value is added to SUM, and the PA value output by FIFO is subtracted. The result divided by N is the average amplitude MEAN. This operation is calculated once for each clock cycle, and twice the MEAN value is the threshold T.

[0027] On the basis of reasonable selection of pipeline length and decision conditions, the signal pre-sorting strategy in different application scenarios can be further adjusted. For example, a long pipeline with high correlation can be used to filter interference and noise in a complex electromagnetic environment; a short pipeline with low correlation can be used for rapid discovery of time-sensitive signals; or a combination of long and short pipelines can be used to separate different types of signals in a complex electromagnetic environment under multiple decision mechanisms.

[0028] Although the content of the present invention has been described in detail through the above preferred embodiments, it should be appreciated that the above description should not be considered as a limitation of the present invention. After reading the above content, it will be apparent to those skilled in the art that various modifications and substitutions of the present invention will occur. Therefore, the protection scope of the present invention should be limited by the appended claims.

Claims

1. A fast signal pre-sorting implementation method based on multi-stage pipeline technology, characterized in that: Includes steps: Step 1: Measure the signal parameters and reduce the data dimension, extract the frequency value RF and amplitude value PA of the signal flow; Step 2: Cache the frequency value RF and the amplitude value PA into the register at a delay of one clock cycle, cache M clock cycles in total, and build an M-level pipeline; Step 3: Calculate the correlation of the frequency values ​​RF in each level of the pipeline within the same clock cycle: whenever there are two frequency values ​​RF whose difference does not exceed 1.5 times the frequency measurement accuracy of the electronic receiver, the correlation K is recorded. i is 1 / M, i is a natural number; Step 4: Sum the correlations in each pipeline level, with a weight of P; Step 5: Make a decision on the weight P, and set the judgment criteria p to be 1 / 2 and 3 / 4 respectively; When the weight P≤1 / 2, it is determined to be noise or interference, and the processing flow ends directly; When 1 / 2<weight P≤3 / 4, go to step 6; When the weight P>3 / 4, go to step 7; Step 6: Perform threshold detection. If the amplitude value PA is greater than the threshold T, it is determined to be a signal, otherwise it is target noise; Step 7: Perform threshold detection. If the amplitude value PA is greater than the threshold T, it is determined as a strong signal, otherwise it is determined as a weak signal; Step 8: Output pre-sorting values, which include the frequency value RF and amplitude value PA of the target signal, strong signal or weak signal.

2. The method for implementing fast signal pre-sorting based on multi-stage pipeline technology according to claim 1, characterized in that: The amplitude value PA is stored in FIFO in the order of clock cycles, with a storage depth of N (N<M). At the same time, the amplitude value PA is accumulated and the accumulated result is recorded as SUM. After FIFO stores N PA values, the latest PA value is added to SUM, and the PA value output by FIFO is subtracted. The result divided by N is the average amplitude MEAN. This operation is calculated once for each clock cycle, and twice the MEAN value is the threshold T.

Images