CN111414257A - Pulsar signal multi-channel filtering method and device and storage medium - Google Patents
Pulsar signal multi-channel filtering method and device and storage medium Download PDFInfo
- Publication number
- CN111414257A CN111414257A CN202010234446.6A CN202010234446A CN111414257A CN 111414257 A CN111414257 A CN 111414257A CN 202010234446 A CN202010234446 A CN 202010234446A CN 111414257 A CN111414257 A CN 111414257A
- Authority
- CN
- China
- Prior art keywords
- impulse response
- finite
- gpu
- data
- length unit
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000001914 filtration Methods 0.000 title claims abstract description 57
- 238000000034 method Methods 0.000 title claims abstract description 32
- 230000004044 response Effects 0.000 claims abstract description 62
- 238000012545 processing Methods 0.000 claims abstract description 26
- 238000000354 decomposition reaction Methods 0.000 claims abstract description 13
- 230000006870 function Effects 0.000 claims description 31
- HPTJABJPZMULFH-UHFFFAOYSA-N 12-[(Cyclohexylcarbamoyl)amino]dodecanoic acid Chemical compound OC(=O)CCCCCCCCCCCNC(=O)NC1CCCCC1 HPTJABJPZMULFH-UHFFFAOYSA-N 0.000 claims description 5
- 238000010276 construction Methods 0.000 claims description 2
- 230000010354 integration Effects 0.000 claims description 2
- 238000001228 spectrum Methods 0.000 abstract description 5
- 238000010586 diagram Methods 0.000 description 20
- RTAQQCXQSZGOHL-UHFFFAOYSA-N Titanium Chemical compound [Ti] RTAQQCXQSZGOHL-UHFFFAOYSA-N 0.000 description 7
- 230000001133 acceleration Effects 0.000 description 7
- 238000005516 engineering process Methods 0.000 description 7
- 238000004364 calculation method Methods 0.000 description 5
- 230000003595 spectral effect Effects 0.000 description 5
- 238000011161 development Methods 0.000 description 3
- 230000005855 radiation Effects 0.000 description 2
- 238000012546 transfer Methods 0.000 description 2
- 238000007405 data analysis Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 230000010363 phase shift Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000000630 rising effect Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000010183 spectrum analysis Methods 0.000 description 1
- 238000009987 spinning Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
- G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/14—Fourier, Walsh or analogous domain transformations, e.g. Laplace, Hilbert, Karhunen-Loeve, transforms
- G06F17/141—Discrete Fourier transforms
- G06F17/142—Fast Fourier transforms, e.g. using a Cooley-Tukey type algorithm
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2209/00—Indexing scheme relating to G06F9/00
- G06F2209/50—Indexing scheme relating to G06F9/50
- G06F2209/5018—Thread allocation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Software Systems (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Analysis (AREA)
- Data Mining & Analysis (AREA)
- Computational Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Algebra (AREA)
- Databases & Information Systems (AREA)
- Discrete Mathematics (AREA)
- Image Processing (AREA)
- Complex Calculations (AREA)
Abstract
The disclosure provides a pulsar signal multi-channel filtering method, a pulsar signal multi-channel filtering device and a storage medium. The method comprises the following steps: constructing a window function; constructing a finite-length unit impulse response filter based on a window function; carrying out multi-phase decomposition on the finite length unit impulse response filter to obtain a plurality of finite length unit impulse response sub-filters with different phases; inputting coefficients of a plurality of finite-length unit impulse response sub-filters with different phases and baseband data of pulsar into a GPU (graphics processing unit) so as to perform finite-length unit impulse response multiphase filtering on the baseband data in the GPU; performing Fourier transform on the multi-phase filtering result to obtain multi-channel data, wherein the number of channels is equal to the number of threads in the GPU; and inputting the multi-channel data into a CPU (central processing unit) to obtain a data file in a filterbank format. The method provided by the disclosure can avoid spectrum leakage and can meet the real-time high-speed channel division requirement of radio signals.
Description
Technical Field
The present disclosure relates to the field of radio astronomical signal processing, and in particular, to a pulsar signal multi-channel filtering method, device and storage medium.
Background
A pulsar is a fast spinning neutron star with a very high density and stable period that emits radiation in the direction of the magnetic poles, and the radio telescope on the earth receives periodic pulsed signals as its radiation sweeps across the earth. The radio telescope processes the received pulsar signal through the terminal equipment. A terminal system based on a Field Programmable Gate Array (FPGA) has been dominant for decades, but as the performance of a Graphics Processing Unit (GPU) is improved and a development tool chain is continuously perfected, a processing part for signals in a digital terminal has been silently shifted to a computing system with the GPU as a core.
In recent years, GPU and Unified Device Architecture (CUDA) technologies have been well applied in the field of pulsar signal processing. The high-speed development of the digital technology continuously improves the performance of the pulsar digital terminal, but at the same time, the ultra-wideband receiving technology continuously puts higher requirements on the processing speed and the processing capacity of the digital terminal. At present, a heterogeneous parallel computing system composed of a CPU and a GPU becomes the mainstream of future high-performance computing, which provides a powerful computing platform for real-time processing of pulsar data.
The multi-channel filtering is a core technology of pulsar signal processing, which divides channels of baseband pulsar data, increases data processing speed, generates multi-channel observation data, and generally uses Fast Fourier Transform (FFT) to divide the channels, but the existing pulsar signal multi-channel filtering technology has the following defects:
firstly, the existing pulsar signal channel division method generally adopts fast fourier transform to divide channels of pulsar signals, so that a filtering window is wide, and after the fourier transform, a final frequency domain contains actual frequency and other frequency components, namely, frequency spectrum leakage occurs, so that the research on weak radio pulsar signals is influenced.
In addition, the existing pulsar signal channel division method is difficult to meet the requirement of real-time channel division of astronomical mass data, the number of channels is very limited, and the resolution ratio is low. Some observation studies require high-resolution data analysis of millions of channels, and the currently used sub-channel technology cannot meet the requirement.
Furthermore, the existing hardware platform for real-time signal processing and high-speed channel division is expensive, such as an FPGA or an asic, and has a relatively complex implementation structure, poor flexibility and extensibility, and a high development period and cost of the system.
Therefore, a multi-channel filtering method for pulsar signals is needed to solve the problem of spectrum leakage and meet the real-time high-speed channel division requirement of radio signals.
Disclosure of Invention
The invention aims to provide a pulsar signal multi-channel filtering method, a pulsar signal multi-channel filtering device and a storage medium, so that spectrum leakage is avoided, and the real-time high-speed channel division requirement of radio signals is met.
In order to achieve the above object, an embodiment of the present disclosure provides a pulsar signal multi-channel filtering method, including:
constructing a window function;
constructing a finite-length unit impulse response filter based on the window function;
carrying out multi-phase decomposition on the finite length unit impulse response filter to obtain a plurality of finite length unit impulse response sub-filters with different phases;
inputting coefficients of the multiple finite-length unit impulse response sub-filters with different phases and baseband data of pulsar into a GPU (graphics processing unit) so as to perform finite-length unit impulse response polyphase filtering on the baseband data in the GPU;
performing Fourier transform on the multi-phase filtering result to obtain multi-channel data, wherein the number of channels is equal to the number of threads in the GPU;
and inputting the multi-channel data into a CPU to obtain a data file in a filterbank format.
In addition, the embodiment of the present disclosure further provides a pulsar signal multi-channel filtering device, including:
the window function constructing module is used for constructing a window function;
the filter construction module is used for constructing a finite-length unit impulse response filter based on the window function;
the multi-phase decomposition module is used for carrying out multi-phase decomposition on the finite length unit impulse response filter to obtain a plurality of finite length unit impulse response sub-filters with different phases;
the polyphase filtering module is used for inputting the coefficients of the multiple finite-length unit impulse response sub-filters with different phases and the baseband data of the pulsar into a GPU (graphics processing unit) so as to perform finite-length unit impulse response polyphase filtering on the baseband data in the GPU;
the Fourier transform module is used for carrying out Fourier transform on the multi-phase filtering result to obtain multi-channel data, wherein the number of channels is equal to the number of threads in the GPU;
and the data file acquisition module is used for inputting the multi-channel data into the CPU to obtain a data file in a filterbank format.
The disclosed embodiments also provide a computer readable storage medium, on which computer instructions are stored, which when executed, implement the steps of the pulsar signal multi-channel filtering method described in any of the above embodiments.
According to the pulsar signal multi-channel filtering method, the decomposed multi-phase filter is designed, so that the problem of frequency spectrum leakage generated in the fast Fourier transform channel division process is effectively controlled, and the pulsar observation resolution can be effectively improved. Furthermore, the pulsar signal multi-channel filtering method provided by the disclosure is based on a high-speed channel division technology of a GPU and CUDA parallel architecture, and solves the problems of large calculation amount and poor real-time signal processing capability of a multi-channel filter algorithm; the method has the advantages that the number of the pulsar signal frequency channels is remarkably increased, the real-time channel division requirement of massive astronomical data is met, meanwhile, a low-cost hardware platform is used, a higher acceleration ratio is obtained, the cost of a real-time signal processing system is reduced, and the flexibility and the expandability of pulsar signal processing are improved.
Drawings
Fig. 1 is a flowchart of a pulsar signal multi-channel filtering method according to an embodiment of the present disclosure;
fig. 2 is a diagram of a pulsar signal multi-channel filtering framework provided by an embodiment of the present disclosure;
fig. 3 is a pulsar signal multi-channel filtering data flow diagram provided by an embodiment of the present disclosure;
FIG. 4 shows (a) a Hamming window time domain diagram and (b) a spectral response diagram;
FIG. 5 shows (a) a finite-length unit impulse response filter impulse response diagram and (b) a spectral response diagram;
fig. 6 is a schematic diagram of forming a plurality of sub-filters by performing polyphase decomposition on a finite-length single-bit impulse response in a multi-channel filtering method for a pulsar signal according to an embodiment of the present disclosure;
FIG. 7 shows (a) a diagram of a synthesized wave and (b) a diagram of the result of finite-length single-bit impulse response polyphase filtering of the synthesized wave in (a);
fig. 8 is a schematic diagram of multi-channel data in a multi-channel filtering method for pulsar signals according to an embodiment of the present disclosure;
FIG. 9 is a GPU multi-channel filter speed-up ratio for a tap number of 32;
FIG. 10 is a GPU multi-channel filter speed-up ratio for a channel number of 32768;
fig. 11 is a block diagram of a pulsar signal multi-channel filtering apparatus according to an embodiment of the present disclosure;
fig. 12 is a schematic diagram of a computer-readable storage medium provided by an embodiment of the disclosure.
Detailed Description
The embodiment of the disclosure provides a pulsar signal multi-channel filtering method, a pulsar signal multi-channel filtering device and a storage medium.
In order to make those skilled in the art better understand the technical solutions in the present disclosure, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present disclosure, and it is obvious that the described embodiments are only a part of the embodiments of the present disclosure, not all of the embodiments. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments in the present disclosure without any inventive step should fall within the scope of protection of the present disclosure.
With reference to fig. 1, a method for multi-channel filtering of a pulsar signal provided in an embodiment of the present disclosure includes:
s1: a window function is constructed.
In signal processing, in order to reduce spectral energy leakage, a clipping function is adopted to cut off a signal, wherein a window function is an important component of signal spectrum analysis and can be used for correcting the non-periodicity of the signal so as to reduce the signal measurement inaccuracy caused by the spectral energy leakage. The present disclosure adopts a hamming window function, and its expression is:
wherein w is a window function, a0＝0.54；a10.46; n is the window length; n is 0,1,2. The time domain plot and spectral response of the Hamming window function are shown in fig. 4:
s2: and constructing a finite-length unit impulse response filter based on the window function.
A Finite Impulse Response (FIR) filter is the most basic element in a signal processing system, and is a non-recursive filter. The finite-length single-bit impulse response filter can achieve strict linear phase shift, so that the system performance is kept stable, the unit impulse response is finite-length, and the filter can be designed to meet the requirements of special frequency characteristic shapes. Specifically, the finite-length single-bit impulse response filter may be:
wherein y is a finite-length single-bit impulse response filter, and h (k) is a fixed coefficient of the filter; x (n-k) is input data passing through k delay units; n is the window length. The transfer function of the system, H (z), can be expressed as:
a finite-length unit impulse response filter impulse response diagram designed by using a Hamming window function can be referred to as the diagram shown in FIG. 5 (a); the frequency response diagram can be referred to as shown in fig. 5 (b).
S3: and carrying out multi-phase decomposition on the finite length unit impulse response filter to obtain a plurality of finite length unit impulse response sub-filters with different phases.
The finite length unit impulse response is subjected to polyphase decomposition to form a plurality of sub-filters, as shown in fig. 6, the original finite length unit impulse response filter is decomposed into a plurality of sub-filters, and the bandwidth of the filter is designed according to the actual application requirements.
S4: and inputting the coefficients of the multiple finite-length unit impulse response sub-filters with different phases and the baseband data of the pulsar into a GPU (graphics processing unit) so as to perform finite-length unit impulse response polyphase filtering on the baseband data in the GPU.
The GPU and the CPU are provided with independent memory spaces, and in a CUDA program, the GPU and the CPU cannot directly access parameters and variables of the other side. The CPU program code cannot access the storage space of the GPU side. Therefore, the finite-length single-bit impulse response sub-filter coefficients and the baseband data of the pulsar to be processed are firstly copied to the GPU, and then the calculation task is carried out in the GPU, so that the multiphase finite-length single-bit impulse response filter is realized in the GPU.
Specifically, the GPU starts multithreading by using a kernel function and initializes register variables defined by the GPU; traversing the baseband data by utilizing the multithreading index; the complex number is multiplied and the loop integration is performed in tap. For example, fig. 7(a) is a schematic diagram of a synthesized wave, and fig. 7(b) is a schematic diagram of a finite-length single-bit impulse response polyphase filtering result of the synthesized wave in (a).
S5: and performing Fourier transform on the multi-phase filtering result to obtain multi-channel data, wherein the number of channels is equal to the number of threads in the GPU.
And carrying out Fourier transform on the multi-phase filtering result to obtain multi-channel data. The calculation of the fourier transform uses CUDA parallel computing architecture cuFFT library to realize discrete fourier transform quickly, and the fourier transform operation on complex signals can use ExecC2C () function. The GPU threads and the thread blocks are arranged in one dimension in the x-axis direction, the number of channels is equal to the length of the x-axis, namely the number of channels is equal to the number of threads.
Referring to fig. 2, a frame diagram of a multi-channel filtering of pulsar signals is shown. It can be seen that, in the method provided by the present disclosure, a finite-length unit impulse response filter based on a window function is first designed, then a filter coefficient is subjected to polyphase decomposition to form a plurality of finite-length unit impulse response sub-filters, each path of sampling signal and the polyphase decomposed sub-filters are subjected to convolution operation, and finally fast fourier transform is performed. H in FIG. 21(z),...,HM-1(z) the system transfer function h (z) is decomposed into a number of branches of different phases, and then digital filtering is performed on each branch, denoted as a decomposed finite-length unit impulse response sub-filter.
Referring to fig. 3, a data flow diagram of multi-channel filtering of pulsar signals is shown. After the input data passes through the FIR filter, the real part and imaginary part of the complex data are separately ordered in sequence and then subjected to fast fourier transform.
For example, fig. 8 is a schematic diagram of 32-channel data obtained by performing multi-channel filtering on pulsar B0823+26 data, wherein the x-axis represents time and the y-axis represents channels in fig. 8.
S6: and inputting the multi-channel data into a CPU to obtain a data file in a filterbank format.
After all the calculation tasks are completed, the data are copied from the GPU video memory to the CPU memory, and meanwhile, the GPU memory space is released. The finally stored data is in a filterbank format and comprises file header information and a data part.
Specifically, the CPU and the GPU can be an Intel Xeon E5-1620 CPU and an NVIDIA GPU, namely TITAN V, TeslaK20 and Quadro P500. software environment is designed by adopting CUDA and L inux systems.
Table 1 shows the data processing time consumption in milliseconds for CPU, P500, TeslaK20 and TITAN V when the tap number is 32 and the channel number is changed between 512 and 65536. As can be seen from table 1, as the number of channels increases, the more the algorithm is calculated, the more time is consumed. The influence of the number of channels on the multichannel filtering calculation of the CPU and the GPU is large, and the running time of the CPU is rapidly increased.
| Channels | CPU | Quadro P500 | Tesla | TITAN V | |
| 512 | 37.731 | 2.61149 | 3.79725 | 1.60874 | |
| 1024 | 75.260 | 5.99709 | 6.63386 | 2.96624 | |
| 2048 | 150.890 | 11.7379 | 12.6636 | 5.63158 | |
| 4096 | 305.913 | 23.3966 | 24.525 | 9.49686 | |
| 8192 | 627.096 | 48.0703 | 48.0394 | 21.8592 | |
| 16384 | 1277.341 | 96.3671 | 95.1814 | 36.6775 | |
| 32768 | 2608.485 | 200.754 | 189.296 | 86.5716 | |
| 65536 | 5766.886 | 383.541 | 377.767 | 142.146 |
TABLE 1
Table 2 shows the data processing time comparison of GPU and CPU platform in milliseconds when the number of channels is 32768 and the number of taps is 4-128. As can be seen from Table 2, the number of taps greatly affects the operating speed of the CPU and GPU algorithms, with larger taps consuming more time.
| taps | CPU | Quadro P500 | Tesla K20 | TITAN V |
| 4 | 646.299 | 145.863, | 173.119 | 80.851 |
| 8 | 920.825 | 148.401 | 175.083 | 83.309 |
| 16 | 1487.213 | 159.555 | 179.944 | 84.1333 |
| 32 | 2610.298 | 198.614 | 189.698 | 86.5778 |
| 64 | 5196.159 | 249.424 | 211.219 | 93.6895 |
| 128 | 10105.86 | 352.63 | 248.823 | 104.692 |
TABLE 2
The high-performance GPU parallel computing system is mainly used for obtaining a better acceleration ratio of a parallel algorithm, and improving the execution efficiency and the real-time data processing capacity of the algorithm. The speed-up ratio of the GPU platform is shown in figures 9 and 10.
FIG. 9 shows that as the number of lanes increases, the acceleration ratio of TITAN V increases, but there are some fluctuations, such as the acceleration ratio of 8192, 32768 lanes decreases. The acceleration ratio of P500, k20 is approximately between 10-15 times. For the number of channels 65535, the calculated speed and CPU ratio of TITAN V were increased by more than 40 times, and the acceleration ratio of P500 and k20 was about 20 times.
As can be seen from FIG. 10, as the number of taps increases, the speed-up ratio of the GPU parallel algorithm increases rapidly and keeps on rising trend, and it can be seen that the number of taps of the multi-channel filtering is a key factor influencing the speed-up ratio of the parallel algorithm. At tap 128, the acceleration ratio of TITAN V is already close to 100 times.
Referring to fig. 11, an embodiment of the present disclosure further provides a pulsar signal multi-channel filtering apparatus, including:
a window function constructing module 100, configured to construct a window function;
a filter constructing module 200, configured to construct a finite-length unit impulse response filter based on the window function;
a polyphase decomposition module 300, configured to perform polyphase decomposition on the finite length unit impulse response filter to obtain multiple finite length unit impulse response sub-filters with different phases;
a polyphase filtering module 400, configured to input coefficients of the multiple finite-length unit impulse response sub-filters with different phases and baseband data of a pulsar into a GPU, so as to perform finite-length unit impulse response polyphase filtering on the baseband data in the GPU;
a fourier transform module 500, configured to perform fourier transform on the polyphase filtering result to obtain multi-channel data, where the number of channels is equal to the number of threads in the GPU;
and a data file obtaining module 600, configured to input the multi-channel data into the CPU, so as to obtain a data file in a filterbank format.
Referring to fig. 12, an embodiment of the present disclosure further provides a computer-readable storage medium, on which computer instructions are stored, and when the instructions are executed, the steps of the pulsar signal multi-channel filtering method in any of the above-mentioned embodiments are implemented.
The above embodiments in the present specification are all described in a progressive manner, and the same and similar parts among the embodiments may be referred to each other, and each embodiment is described with emphasis on being different from other embodiments.
The above description is only a few embodiments of the present application, and although the embodiments disclosed in the present application are as described above, the above description is only for the convenience of understanding the technical solutions of the present application, and is not intended to limit the present application. It will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the disclosure as defined by the appended claims.
Claims (9)
1. A pulsar signal multi-channel filtering method is characterized by comprising the following steps:
constructing a window function;
constructing a finite-length unit impulse response filter based on the window function;
carrying out multi-phase decomposition on the finite length unit impulse response filter to obtain a plurality of finite length unit impulse response sub-filters with different phases;
inputting coefficients of the multiple finite-length unit impulse response sub-filters with different phases and baseband data of pulsar into a GPU (graphics processing unit) so as to perform finite-length unit impulse response polyphase filtering on the baseband data in the GPU;
performing Fourier transform on the multi-phase filtering result to obtain multi-channel data, wherein the number of channels is equal to the number of threads in the GPU;
and inputting the multi-channel data into a CPU to obtain a data file in a filterbank format.
2. The method of claim 1, wherein the window function is a hamming window function, and the expression is:
wherein w is a window function, a0＝0.54；a10.46; n is the window length; n is 0,1,2.
3. The method of claim 1, wherein the finite-length unit impulse response filter is:
wherein y is a finite length unit impulse response filter, and h (k) is a fixed coefficient of the finite length unit impulse response filter; x (n-k) is input data passing through k delay units; n is the window length.
4. The method of claim 1, wherein the performing, in the GPU, finite-length single-bit impulse response polyphase filtering on the baseband data comprises:
multithreading is initiated using the kernel function,
initializing register variables defined by the GPU;
traversing the baseband data using the multi-threaded index;
the complex number multiplication is calculated and the loop integration is performed within tap.
5. The method of claim 1, wherein a CUDA parallel computing architecture cuFFT library is utilized to perform Fourier transform on the polyphase filtering result to obtain multi-channel data.
6. The method of claim 1, wherein the multi-channel data is located in a video memory of the GPU, and wherein the multi-channel data is input into a memory of the CPU from the video memory of the GPU.
7. The method of claim 1, further comprising initializing the GPU and the CPU prior to inputting data into the GPU and the CPU, the initializing comprising at least: opening up the GPU video memory space; allocating the CPU memory space; designing a Fourier transform length; and setting kernel function parameters.
8. A pulsar signal multi-channel filtering device, comprising:
the window function constructing module is used for constructing a window function;
the filter construction module is used for constructing a finite-length unit impulse response filter based on the window function;
the multi-phase decomposition module is used for carrying out multi-phase decomposition on the finite length unit impulse response filter to obtain a plurality of finite length unit impulse response sub-filters with different phases;
the polyphase filtering module is used for inputting the coefficients of the multiple finite-length unit impulse response sub-filters with different phases and the baseband data of the pulsar into a GPU (graphics processing unit) so as to perform finite-length unit impulse response polyphase filtering on the baseband data in the GPU;
the Fourier transform module is used for carrying out Fourier transform on the multi-phase filtering result to obtain multi-channel data, wherein the number of channels is equal to the number of threads in the GPU;
and the data file acquisition module is used for inputting the multi-channel data into a CPU (central processing unit) to obtain a data file in a filterbank format.
9. A computer readable storage medium having stored thereon computer instructions which, when executed, implement the steps of the method of any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010234446.6A CN111414257A (en) | 2020-03-30 | 2020-03-30 | Pulsar signal multi-channel filtering method and device and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010234446.6A CN111414257A (en) | 2020-03-30 | 2020-03-30 | Pulsar signal multi-channel filtering method and device and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111414257A true CN111414257A (en) | 2020-07-14 |
Family
ID=71491546
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010234446.6A Pending CN111414257A (en) | 2020-03-30 | 2020-03-30 | Pulsar signal multi-channel filtering method and device and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111414257A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114415933A (en) * | 2021-12-29 | 2022-04-29 | 北京无线电计量测试研究所 | FPGA-based cross-correlation arithmetic unit, processing method and signal processing system |
| CN115587288A (en) * | 2022-12-07 | 2023-01-10 | 中国科学院国家天文台 | Non-2-power special point spectrum calculation method and system |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105915193A (en) * | 2016-06-21 | 2016-08-31 | 电子科技大学 | Improved generation method of multiphase filter |
| CN107749762A (en) * | 2017-09-26 | 2018-03-02 | 天津光电通信技术有限公司 | A kind of multiphase filtering digital channelizing implementation method |
| CN110455282A (en) * | 2019-08-15 | 2019-11-15 | 中国科学院新疆天文台 | A kind of digital termination system applied to observations of pulsar |
-
2020
- 2020-03-30 CN CN202010234446.6A patent/CN111414257A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105915193A (en) * | 2016-06-21 | 2016-08-31 | 电子科技大学 | Improved generation method of multiphase filter |
| CN107749762A (en) * | 2017-09-26 | 2018-03-02 | 天津光电通信技术有限公司 | A kind of multiphase filtering digital channelizing implementation method |
| CN110455282A (en) * | 2019-08-15 | 2019-11-15 | 中国科学院新疆天文台 | A kind of digital termination system applied to observations of pulsar |
Non-Patent Citations (1)
| Title |
|---|
| 托乎提努尔等: "基于CUDA的射电天文多相滤波器组设计", 《天文研究与技术(国家天文台台刊)》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114415933A (en) * | 2021-12-29 | 2022-04-29 | 北京无线电计量测试研究所 | FPGA-based cross-correlation arithmetic unit, processing method and signal processing system |
| CN114415933B (en) * | 2021-12-29 | 2024-03-29 | 北京无线电计量测试研究所 | Cross-correlation operator based on FPGA, processing method and signal processing system |
| CN115587288A (en) * | 2022-12-07 | 2023-01-10 | 中国科学院国家天文台 | Non-2-power special point spectrum calculation method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10367477B2 (en) | Sparse cascaded-integrator-comb filters | |
| Plaumann et al. | Determinantal representations of hyperbolic plane curves: an elementary approach | |
| Sreenivas et al. | FFT algorithm for both input and output pruning | |
| CN111414257A (en) | Pulsar signal multi-channel filtering method and device and storage medium | |
| US9015219B2 (en) | Apparatus for signal processing | |
| CN111786688B (en) | Broadband parallel channelization receiving method based on embedded GPU | |
| CN105991137A (en) | Systems and methods of variable fractional rate digital resampling | |
| CN104967428B (en) | Frequency domain implementation method for FPGA high-order and high-speed FIR filter | |
| US9705476B2 (en) | Optimal factoring of FIR filters | |
| US9189457B2 (en) | Method for distributing computing load in a data processing system | |
| Kim et al. | Implementation of a high-throughput low-latency polyphase channelizer on GPUs | |
| EP3928445B1 (en) | Resource conserving weighted overlap-add channelizer | |
| Kim et al. | An efficient GPU implementation of an arbitrary resampling polyphase channelizer | |
| Pelkowitz | Frequency domain analysis of wraparound error in fast convolution algorithms | |
| Jameil et al. | Efficient FIR Filter Architecture using FPGA | |
| Alam et al. | On the implementation of time-multiplexed frequency-response masking filters | |
| Kim et al. | Implementation of a low-complexity low-latency arbitrary resampler on GPUs | |
| Zierhofer | Data window with tunable side lobe ripple decay | |
| Ma et al. | Implementation of a digital down converter using graphics processing unit | |
| Raj et al. | A paradigm of distributed arithmetic (DA) approaches for digital FIR filter | |
| GB2530289A (en) | Low-complexity channelisation scheme catering for signals of arbitrary centre frequency and bandwidth | |
| Al-safi et al. | GPU based implementation of a 64-channel polyphase channalizer | |
| Kim et al. | Implementation of a Multirate Resampler for Multi-carrier Systems on GPUs | |
| Nakahara et al. | On a wideband fast Fourier transform for a radio telescope | |
| Faulkner et al. | GPU synthesis of RF channeliser outputs for a variable bandwidth microwave digital receiver |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20200714 |