CN116482449B - Real-time detection method for fast electric storm - Google Patents
Real-time detection method for fast electric storm Download PDFInfo
- Publication number
- CN116482449B CN116482449B CN202310728309.1A CN202310728309A CN116482449B CN 116482449 B CN116482449 B CN 116482449B CN 202310728309 A CN202310728309 A CN 202310728309A CN 116482449 B CN116482449 B CN 116482449B
- Authority
- CN
- China
- Prior art keywords
- data
- candidate
- fast
- time
- storm
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 58
- 238000011897 real-time detection Methods 0.000 title claims abstract description 15
- 238000001514 detection method Methods 0.000 claims abstract description 36
- 238000012216 screening Methods 0.000 claims abstract description 27
- 230000010354 integration Effects 0.000 claims abstract description 6
- 230000009466 transformation Effects 0.000 claims abstract description 5
- 230000008569 process Effects 0.000 claims description 23
- 238000012545 processing Methods 0.000 claims description 15
- 239000006185 dispersion Substances 0.000 claims description 13
- 230000010287 polarization Effects 0.000 claims description 10
- 230000007246 mechanism Effects 0.000 claims description 9
- 230000001133 acceleration Effects 0.000 claims description 8
- 238000004364 calculation method Methods 0.000 claims description 8
- 239000013307 optical fiber Substances 0.000 claims description 4
- 230000003595 spectral effect Effects 0.000 claims description 4
- 230000009977 dual effect Effects 0.000 claims description 3
- 238000002360 preparation method Methods 0.000 claims description 3
- 230000010076 replication Effects 0.000 claims description 2
- 238000001228 spectrum Methods 0.000 claims description 2
- 230000035945 sensitivity Effects 0.000 abstract description 7
- 238000010891 electric arc Methods 0.000 abstract 2
- 238000010586 diagram Methods 0.000 description 6
- 230000008901 benefit Effects 0.000 description 5
- 238000011160 research Methods 0.000 description 5
- JJWKPURADFRFRB-UHFFFAOYSA-N carbonyl sulfide Chemical compound O=C=S JJWKPURADFRFRB-UHFFFAOYSA-N 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- UFHFLCQGNIYNRP-UHFFFAOYSA-N Hydrogen Chemical compound [H][H] UFHFLCQGNIYNRP-UHFFFAOYSA-N 0.000 description 2
- 230000008030 elimination Effects 0.000 description 2
- 238000003379 elimination reaction Methods 0.000 description 2
- 230000003203 everyday effect Effects 0.000 description 2
- 238000004880 explosion Methods 0.000 description 2
- 229910052739 hydrogen Inorganic materials 0.000 description 2
- 239000001257 hydrogen Substances 0.000 description 2
- 238000012827 research and development Methods 0.000 description 2
- 241000231739 Rutilus rutilus Species 0.000 description 1
- 238000010521 absorption reaction Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000013213 extrapolation Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 238000000752 ionisation method Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000003252 repetitive effect Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 239000011232 storage material Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R29/00—Arrangements for measuring or indicating electric quantities not covered by groups G01R19/00 - G01R27/00
- G01R29/08—Measuring electromagnetic field characteristics
- G01R29/0807—Measuring electromagnetic field characteristics characterised by the application
- G01R29/0814—Field measurements related to measuring influence on or from apparatus, components or humans, e.g. in ESD, EMI, EMC, EMP testing, measuring radiation leakage; detecting presence of micro- or radiowave emitters; dosimetry; testing shielding; measurements related to lightning
- G01R29/0842—Measurements related to lightning, e.g. measuring electric disturbances, warning systems
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R29/00—Arrangements for measuring or indicating electric quantities not covered by groups G01R19/00 - G01R27/00
- G01R29/08—Measuring electromagnetic field characteristics
- G01R29/0864—Measuring electromagnetic field characteristics characterised by constructional or functional features
- G01R29/0871—Complete apparatus or systems; circuits, e.g. receivers or amplifiers
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks
Abstract
The application discloses a real-time detection method of a rapid radio storm, which comprises the following steps: s1: the method comprises the steps of accessing rapid electric-arc detection data into a 500-meter-caliber spherical radio telescope (Five-handred-meter Aperture Spherical radio Telescope, FAST) multi-beam receiver digital back-end system, adopting a data parallel method, realizing the access and operation of the rapid electric-arc detection data under the condition of not affecting normal observation of FAST, and reading an observation plan; s2, storing data which corresponds to the candidate body and is subjected to fast Fourier transform and integration; s3: and (3) reading the sample data processed in the step (S2) in real time, identifying and classifying the sample data by adopting a multi-beam feed source screening strategy, and determining the candidate body. The real-time detection method of the rapid electric storm can utilize the sensitivity of FAST to rapid electric storm detection to realize real-time and efficient detection of the rapid electric storm.
Description
Technical Field
The application relates to the field of radioastronomy, in particular to a real-time detection method of a rapid radiostorm.
Background
Rapid radio storms are currently known as the brightest burst of radio bands in the universe. The estimated burst of pulses reaching the earth is as many as thousands to tens of thousands per day, and we have only guesses about their origin and physical mechanism, far from conclusive. According to the FRBCAT (Fast Radio Burst Catalog, FRBCAT) statistics, hundreds of fast-shot storms have been found so far, of which tens are identified as fast-shot storms of repeated bursts.
The FAST (Five-handred-meter Aperture Spherical radio Telescope, FAST) field of view is relatively small, and is less dominant than large field of view telescopes such as chem (Canadian Hydrogen Intensity Mapping Experiment, chem) in terms of the number expectations of discovering new FAST-storms. But has absolute advantages in terms of quality found. Along with the gradual expansion of various kinds of telescope in the world aiming at the observation of the rapid electric storm and the great increase of the discovery quantity, the discovery demand on high quality becomes more urgent, and the application is the key improvement of the discovery quality of the rapid electric storm. FAST has the unique advantage and potential of significant discovery in this area:
first, the findings that have been reported so far have hundreds of times, but with the deep research and the continuous increase in the number of findings, the high positioning accuracy and the burst-induced high-accuracy data brought about by high-sensitivity telescopes become very important and critical.
The rapid storm is a target on the cosmic distance, so the research on the absorption line in the signal can be used for researching the real distance of the rapid storm, the intrastar medium of the rapid storm host star system, the intersystem medium on the cosmic distance and the like. The method brings a new scientific breakthrough with wider significance to the final play of the real physical meaning of the rapid electric storm as the cosmic phenomenon.
Meanwhile, the estimated burst pulse reaching the earth every day is as many as thousands to tens of thousands, and FAST can find and capture the rapid electric storm quantity, which is not small.
FAST has extremely high sensitivity, and combines the comprehensive consideration of the quantity and quality of FAST discovery, so that FAST has unique advantages and opportunities of great discovery in the field.
Secondly, FAST is an ideal repeated storm observation device, and the best explosion signal-to-noise ratio and explosion rate statistical data in the world can be obtained. Thereby solving a problem of origin with respect to a rapid radio storm. The first repeat source FRB121102 has nearly 100% linear polarization and an extremely high faraday Rotation Measurement (RM). Whether all repeated sources have such a high RM will solve if the repeated sources originate from young stars or other mechanisms such as "combing" of the magnetic field of the foreground stars by the outflow of large mass black holes. In the research of repeated storms, FAST has made important progress, and in 2019 we have realized that FRB121102 repeated storms are found in real time first, and the most burst rate statistics and the highest burst signal-to-noise ratio are obtained.
Finally, FAST will likely detect the most distant FAST radio storm in the universe. The maximum DM that has been detected is 2596.1, the corresponding red shift value of 3. Based on extrapolation of known FAST shot data, FAST can detect FAST shots with a red shift above 10 if they exist. The discovery of these high red-shifted FRBs, especially fast-shot storms with red-shifted values above 6, will make an important contribution to the investigation of the evolution of the universe, especially the early universe re-ionization process.
Current observations are extremely lacking. The application is developed and utilized for key-counterpart authentication in the field, and needs to greatly improve the observation capability of FAST electric storm.
Disclosure of Invention
Aiming at the problems existing in the prior art, the application aims to provide a real-time detection method for a rapid electric storm, which can realize real-time and high-efficiency detection for the rapid electric storm by utilizing the sensitivity of FAST to the rapid electric storm detection.
In order to achieve the above object, the present application provides a real-time detection method of a rapid electric storm, the method comprising the steps of:
s1: the FAST electric storm detection data are accessed into a digital back-end system of the FAST multi-beam receiver, and a data parallel method is adopted to realize the normal observation of spectral lines and pulsar data of the FAST and access into the FAST electric storm detection data, and search for FAST electric storm signals in real time;
s2, storing sample data which corresponds to the candidate body and is subjected to fast Fourier transform and integration;
s3: reading the sample data processed in the step S2 in real time, identifying and classifying the sample data by adopting a multi-beam feed source screening strategy, and determining a candidate;
the method for screening candidates through the multi-beam feed source screening strategy and the screening process are specifically as follows:
after 19 computing nodes search candidate bodies in real time through GPU acceleration software Heimdall, all candidate bodies are sent to a recloser process which runs on a head node all the time; in order to ensure that the searched candidates come from the same radio source, before the data enter the recloser process for further processing, the time for transmitting the calculated node candidates to the head node is judged, and if the time for the candidates to reach the head node between different nodes is less than 10 seconds, the head node collects all the candidate data for further processing; if the arrival time exceeds 10 seconds, reporting errors in the data record; the recloser process compiles all candidates from each compute node into one file at a given time and adds one RFI flag to determine the probability of candidate being RFI by examining candidate recloses across multiple beams; the candidate screening process at the head node will then further screen the candidates based on the beam number of the detected candidate, the signal to noise ratio S/N (Signal to Noise ratio, S/N), the signal pulse width.
Further, in step S1, the FAST-shot storm detection data is 19-beam dual-polarized feed data in the feed bin, which is connected to a FAST multi-beam receiver digital back-end system through an optical fiber, wherein the FAST multi-beam receiver digital back-end system comprises 10 ROACH2 boards (Reconfigurable Open-Architecture Compute Hardware version 2, ROACH 2), each ROACH2 board is provided with 2 analog-to-digital converters ADC (Analog to Digital Converter, ADC), each analog-to-digital converter ADC can be connected to single-beam dual-polarized data, and the data bandwidth is 500MHz (F nyq ) The sampled data is then integrated 11 times (acc_len) after 4096-point (nbin) fast fourier transform, and finally the data is encapsulated into a user datagram protocol (User Datagram Protocol, UDP) UDP packet stream.
Further, the encapsulated UDP data packet flows realize the copying and the propagation of data through a multicast mechanism of a network card, and then the UDP data packet flows are sent to 19 computing node servers through a high-speed 40/100GbE Ethernet switch.
Further, in step S2, 19 computing node servers with fixed IP and mac addresses are set to receive data sent by the 40/100GbE Ethernet switch, and each path of data has a data rate of 1GB/S.
Further, in step S2, three data processing threads are developed on each computing node to receive and process the UDP packet stream, first, the network thread is responsible for receiving and extracting the UDP packets and storing them in the designated input ring buffer; secondly, converting the two-path polarization data into a Stokes polarization form by a calculation thread, and writing the Stokes polarization form into an output ring buffer area; third, the output thread reads the data from the output ring buffer, stores it in SIGPROC filterbank data format, and stores it in the compute node's cache in preparation for incoherent dispersion.
Further, in step S2, in each computing node server cache SIGPROC filterbank data streams are received from the node to be processed by real-time incoherent dispersion cancellation in GPU acceleration software Heimdall.
Further, pulse searching in a large dispersion range is performed in real time, single pulses with dispersion values between 20 and 10000 are searched, the efficiency step length is 25%, and the master control node performs candidate screening according to the distribution of the single pulses in 19 beams.
Further, in step S3, the recloser on the head node determines in real time whether or not 5 or more of the 19 beams have detected pulses simultaneously, and if such candidates are detected, determines RFI to exclude ground RFI interference.
Further, for the candidates from which the ground RFI interference has been eliminated, the following determination conditions are set in the subsequent screening process:
S/N >10；
detecting the candidate adjacent beam quantity < =4;
maximum pulse width <128 milliseconds;
the number of allowed events/sec 2.
Further, in step S3, if the candidate is not screened by the recloser process, the candidate attribute detected by different beams is continuously and comprehensively determined in real time by the candidate screening thread on the head node, and the interference candidate is screened out according to the beam number, S/N, signal pulse width of the candidate, and the known pulsar, RFI distribution and interference frequency band of the aircraft in the field of view of FAST, which are compared with the frequency of the signal of the searched candidate, so as to further screen the candidate.
According to the real-time detection method of the rapid electric storm, the sensitivity of FAST to rapid electric storm detection can be utilized, and real-time and efficient detection of the rapid electric storm can be realized.
Drawings
Fig. 1 shows a diagram of a terminal system architecture of a fast-shot storm detection method according to the application;
FIG. 2 shows a flowchart of the method of fast shot storm detection according to the application;
fig. 3 shows a beam mask table diagram of a fast radio storm detection method according to the application;
fig. 4 shows a repetitive storm FRB121102 detected in real time by a rapid radio storm detection method according to the present application.
Detailed Description
The following description of the embodiments of the present application will be made more apparent and fully hereinafter with reference to the accompanying drawings, in which some, but not all embodiments of the application are shown. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
In the description of the present application, it should be noted that the directions or positional relationships indicated by the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc. are based on the directions or positional relationships shown in the drawings, are merely for convenience of describing the present application and simplifying the description, and do not indicate or imply that the devices or elements referred to must have a specific orientation, be configured and operated in a specific orientation, and thus should not be construed as limiting the present application. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance.
In the description of the present application, it should be noted that, unless explicitly specified and limited otherwise, the terms "mounted," "connected," and "connected" are to be construed broadly, and may be either fixedly connected, detachably connected, or integrally connected, for example; can be mechanically or electrically connected; can be directly connected or indirectly connected through an intermediate medium, and can be communication between two elements. The specific meaning of the above terms in the present application can be understood by those of ordinary skill in the art according to the specific circumstances.
Specific embodiments of the present application are described in detail below with reference to fig. 1-4. It should be understood that the detailed description and specific examples, while indicating and illustrating the application, are not intended to limit the application.
FAST is a short name of 500m caliber spherical radio telescope, which is composed of active reflecting surface system, receiver, terminal and observation base, etc. by 2023 month 3, FAST has found more than 740 pulsars, far exceeding the number of other telescopes found internationally.
The digital back end has strong universality and flexibility, and the research and development of the digital back end of firmware and hardware not only has important practical significance and scientific significance, but also has important strategic safety significance. The receiver and the back-end technology not only can be matched with the huge receiving area of FAST and the calm site condition to achieve the highest observation sensitivity in a low-frequency band, but also can be applied to newly built or upgraded telescopes with low cost, such as a 65 m telescope of an astronomical station in the sea and a 110 m telescope of an astronomical station in Xinjiang, and can be popularized to a 40 m telescope of an astronomical station in Yunnan, a 25 m telescope of an astronomical station in Xinjiang and the like for use, so that the observation performance is further improved. By means of the world leading technical indexes, the method can also compete for Arecibo and GBT telescope receiver projects in the United states, and can be used for making technical reserves for participation in SKA in China.
The application relates to a real-time and efficient rapid electric storm detection method, which is developed by combining software and hardware developed by a FAST telescope, and in order to detect the rapid electric storm, the rapid electric storm detection data of the application is accessed into the FAST, and the detection is performed by utilizing the technical advantages and high sensitivity of FAST observation.
The real-time detection method of the rapid electric storm comprises the following steps:
s1: the FAST shot detection data are accessed into a FAST multi-beam receiver, and the FAST normal observation spectral line and pulsar data are accessed simultaneously by adopting a data parallel method, and FAST shot signals are searched in real time;
s2, storing sample data which corresponds to the candidate body and is subjected to fast Fourier transform and integration;
s3: and (3) reading the sample data processed in the step (S2) in real time, identifying and classifying the sample data by adopting a multi-beam feed source screening strategy, and determining the candidate body.
Firstly, in step S1, a 19-beam dual-polarized feed signal in a feed bin is connected to a FAST multi-beam receiver digital back-end system through an optical fiber, and the FAST multi-beam receiver digital back-end system mainly collects radio frequency signals by a second-generation reconfigurable open architecture board ROACH2 developed by 10 astronomical signal processing and electronic research organizations. Each ROACH2 board is provided with 2 ADCs, each analog-to-digital converter ADC can be connected with single-beam dual-polarized data, and the data bandwidth is 500MHz (F nyq ) Then, the sampled data is integrated 11 times (acc_len) after 4096 points (nbin) fast Fourier transform, and finally the data is packaged into UDP data packet flow.
FAST is used as a single-antenna radio telescope with highest sensitivity and largest caliber, scientific targets are numerous, and observation time is limited. The rapid electric storm detecting method fully considers the point, copies and spreads the data by the multicast mechanism of the network card through the data packet flow packaged by ROACH2, and then sends the user datagram protocol data packet flow to 19 computing node servers through a high-speed 40/100GbE Ethernet switch. The FAST normal observation spectral line and pulsar data are realized, the FAST shot storm detection data are accessed, and FAST shot storm signals are searched in real time.
Secondly, in step S1, the observation mode changes much every day, and the invented rapid radio storm detection method can read the observation plan in real time and automatically match with the data form of different modes to perform pulse capturing in the observation period. In observation, the parameters to be set include the integrated length of data after fast fourier transform (acc_len) on the ROACH2 board, the parameter key value in the redis database is continuously accessed to adjust the acc_len value in the system, the value of acc_len is now set to 11, and then the start_flag bit is set to 1, and data recording is started.
In step S2, 19 computing node servers with fixed IP and mac addresses are set to receive data sent by the 40/100GbE Ethernet switch, and each path of data has a data rate of 1GB/S. One fundamental problem faced by FAST real-time FRB search back-end signal processing is: with the increase of the data rate of the network port of the computing node 10Gbe, especially when the rate is close to 1GB/s, the packet loss rate of the data receiving system will be higher and higher, but the data still needs to be transmitted on time. This is a significant challenge for high performance computing and software processing performance. Three data processing threads are arranged in the application and are used for receiving and processing UDP packets.
First, the network thread is responsible for receiving and extracting the UDP packets and storing them in the designated input ring buffer. Second, the computation thread converts the two-way polarization data into Stokes polarization form and writes it to the output ring buffer. Third, the output thread reads the data from the output ring buffer, stores it in SIGPROC filterbank format, and stores it in the compute node's cache in preparation for incoherent dispersion. In addition, by a Semaphore protection mechanism, it is ensured that the same segment of the ring buffer is not read and written concurrently. In addition, a hash key interface is also provided for monitoring the running state of each thread, including network port information, the use condition of the real-time input/output buffer area and the like. And 3 threads process the rapid radio storm detection data flow in parallel, so that the real-time processing and storage of high-speed data are realized.
In step S3, in each computing node server cache SIGPROC filterbank data stream, real-time incoherent dispersion elimination processing is performed in GPU acceleration software Heimdall connected to the node. In real-time searching, the dispersion value (Dispersion Measure, DM) ranges from 20 to 10000, and the efficiency step size is 25%, which is the maximum expected loss due to mismatch between the actual DM of the searched candidate and the selected DM step size.
In step S3, the method for screening the candidate through the multi-beam feed source screening strategy specifically includes: after each computing node searches for the candidate in real time through the GPU acceleration software Heimdall, all the candidates are sent to the recloser process running on the head node. In order to ensure that the searched candidates come from the same radio source, before the data enter the recloser for further processing, the time for sending the calculated node candidates to the head node is judged, and if the time for the candidates to reach the head node between different nodes is less than 10 seconds, the head node collects all the candidate data for further processing; if the arrival time exceeds 10 seconds, an error will be reported in the data record. The recloser process compiles all candidates from each computing node into one file at a given time and adds one radiofrequency interference RFI. Since the FRB is from a point source and its pulse signal energy is concentrated, if five or more candidates from different beams are detected simultaneously, or two or more from non-adjacent beams, then the detected candidate signal is likely to be RFI, which is the first step of screening for candidates. The candidate screening process on the head node will then further screen the candidates based on the beam number, S/N, signal pulse width of the detected candidate.
First, a beam mask is generated based on the detected candidate beam numbers, the mechanism of which is as follows: if a candidate signal is detected in the beam, we set the flag bit to 1; otherwise, the flag bit is set to 0. This produces a 19-bit binary beam mask with 2-bit RFI identification bits added to the top bits of the mask. For example, if two similar signals are detected in beams 1 and 2, the binary beam mask will be set to 00 0000000000000000011 and the corresponding decimal effective mask will be set to 3, and if 4 similar signals are detected in beams 2, 3, 9, and 10, the binary beam mask will be set to 00 0000000001100000110 and the corresponding decimal effective mask 774. This creates a valid mask table that is unlikely to be an RFI signal if the candidate beam number is found on the table. Meanwhile, the following determination conditions are set in the candidate screening process:
S/N >10；
detecting the candidate adjacent beam quantity < =4;
maximum pulse width <128 milliseconds;
the number of allowed events/sec 2.
Since candidates of the 19 calculation nodes are further screened on the head node, the calculation amount is very large, and the post_acc parameter value needs to be further set to reduce the time resolution of the candidate data, and the frequency resolution and the time resolution of the final candidate data are calculated as follows:
in the specific implementation of the present application,representing frequency resolution, +.>The bandwidth of the rapid radio storm detection data is represented, and the value of the bandwidth is 500MHz; />The number of spectrum channels is represented, and the value is 4096; and finally, the frequency resolution of the signal obtained by calculation is 122KHz. />And (5) representing the initial time resolution, and finally calculating to obtain the signal initial time resolution of 8.192 mu s.
The final time resolution of the signal isWherein->The integral length of the data after the fast Fourier transform on the ROACH2 plate is 11; />The length of the data integration which is done before the candidate signal is further screened on the head node is 3; and calculating to obtain the final signal time resolution of 270 mu s.
By adjusting the S/N value to adjust the system minimum detection threshold, we set the S/N value to 10. Furthermore, since a fast shot storm is a point source in the sky, it is unlikely to be detected by more than 4 beams. Also, the large number of fast-shot storms detected in the L-band are narrower in time, so maintaining the pulse width threshold close to 128 milliseconds does not remove any real candidates. Meanwhile, RFI is a significant trouble, whether in a rapid shot storm or pulsar observation, or in a conventional telescope observation. The elimination of RFI is particularly critical for the observation of rapid storms, because it directly or indirectly generates a large number of invalid candidates, which not only results in wasted calculation and storage materials, but also interferes with the instantaneous signal search quality of rapid storms. Comparing the known RFI distribution of FAST and the interference frequency band of the aircraft in the field of view with the frequency of the searched candidate signal, and screening out the interference candidates. Once a new candidate file is found, the candidate screening process changes the file state value in the redis database, then pushes the state information of the file on the computing node to the head node, queries the state information of the file on the head node at intervals, and deletes the data file in the data cache if the file state information is changed from-1 to 0; if the file state is changed from-1 to 1, the data file is stored in the disk array from the cache; if the file state is kept unchanged within 10 minutes due to the problem of RFI signal discrimination or the problem of the head node, another process deletes the files, so that the buffer area always keeps an idle storage position to ensure that real-time data is always written. To further improve the efficiency of the fast storm search, the head node finally lists the carefully examined candidate files, then reads the Redis database to determine the corresponding SIGPROC filterbank file, and finally draws the candidate waterfall.
In step S3, a repeat storm FRB121102 is finally detected using the rapid radio storm detection method.
In addition, fig. 1 and 2 show the terminal architecture and workflow of the fast-shot detection method of the present application, and fig. 3 and 4 show a beam mask table and a real-time detected repeat storm FRB121102. The figures are illustrated as follows:
FIG. 1: a terminal system architecture diagram of the rapid electric storm detection method according to the application;
1. the photoelectric converter receives 19-beam dual-polarized feed source data transmitted by the optical fiber and converts the optical signal into an electric signal.
The hydrogen clock provides 1PPS and 10MHz reference frequency, wherein 1PPS signals are distributed to 10 ROACH2 boards through a pulse distributor, so that synchronization with a FAST multi-beam receiver digital back-end system is realized; the 10MHz reference frequency is provided to the ADC on the ROACH2 board by a frequency synthesizer generating a 1GHz sampling frequency.
2. The 10 ROACH2 boards digitize 19-beam dual-polarized feed source data, perform fast Fourier transform and integrate, make the processed data into UDP packets, and send the data through a 10GbE network port.
3. The replication and distribution of the data are realized through a network card multicast mechanism, and the 40/100GbE switch distributes UDP data packets to 19 computing nodes.
4. The 19 computing nodes receive and process the data, store the data in a cache, search pulses in real time through a GPU acceleration program Heimdall, and finally judge candidates from the 19 computing nodes by a head node server.
Fig. 2: a workflow diagram of a rapid shot storm detection method according to the application;
1. the high-speed data receiving and processing thread on each computing node receives UDP data packets in real time, converts each beam of dual-polarized data into Stokes form and stores the Stokes form into a server cache;
2. searching candidate bodies in real time through a GPU acceleration program Heimdall;
3. the recloser judges whether 5 or more than 5 wave beams detect signals simultaneously in real time, and if so, the recloser judges as RFI; if the candidate screening thread does not continue to pass, screening out the interference candidates according to the detected wave beam number, S/N, signal pulse width of the candidates, and the comparison between the known pulsar of FAST, RFI distribution and the interference frequency band of the aircraft in the field of view and the searched signal frequency of the candidate, and further screening the candidates. After drawing the candidate waterfall diagram, the data file is stored in the disk array from the cache.
Fig. 3: a beam mask table diagram of a fast shot storm detection method according to the application;
the left panel shows 19 feeds and a dual polarization layout, with the right table being a beam combination where it is possible to detect truly fast radio storm pulses. We do this binary beam mask table, beam 1, beam 2, left to right, through to beam 19, and once the beam detects a candidate signal, set the position to 1 and the rest to 0. This effective mask table is finally obtained, while converting the binary values into decimal values. The decimal mask value is used as one of the criteria for determining whether the system detects a true signal.
Fig. 4: the rapid shot storm detection method according to the present application detects repeated storm FRB121102 in real time.
The fast radio storm search terminal detects the repeated storm FRB121102 in real time. The lowest graph shows the signal detected in real time, showing the candidate waterfall graph without dispersion, the abscissa being the time axis, the ordinate being the frequency axis, the color representing the intensity; the middle graph shows a dispersion-eliminated candidate waterfall graph; the uppermost plot shows the profile of the pulse, showing the S/N of the signal detected in real time.
In the application, the FAST multi-beam receiver digital back-end system is recorded in the following steps: jiang P, tang N Y, hou L G, et al The fundamental performance of FAST with 19-beam receiver at L band [ J ]. Research in Astronomy and Astrophysics, 2020, 20 (5): 064.
The application has the technical advantages and beneficial effects that:
firstly, the application successfully realizes a real-time detection method of a FAST radio storm on a digital back-end system of the FAST multi-beam receiver, realizes multiple detection of repeated storm FRB121102 and discovers a plurality of new pulsars.
Meanwhile, in order to improve the rapid radio storm searching efficiency, the anti-interference capability is improved by accessing multiple information in real time. Telescope pointing data (known pulsar information in an observation path and satellite and aircraft data in a field of view are determined), electromagnetic interference data recorded on site, known fixed interference source data and the like are read in the system, and interference candidates are eliminated.
Finally, through the realization of a complete real-time detection method of the rapid electric storm, practical experience is provided for a plurality of telescope projects in China. The developed software and hardware technology and achievements not only provide support for the digital back-end system of the FAST multi-beam receiver, but also can be applied to more ground and space telescopes in China. The method has strong universality, flexibility and portability, is the research and development of software, firmware and hardware with independent intellectual property rights in China, and has important practical significance and scientific significance and important strategic safety significance. In the future, the development of related technologies will be led to improve the international competitiveness and influence.
Claims (10)
1. The real-time detection method of the rapid radio storm is characterized by comprising the following steps:
s1: the FAST electric storm detection data are accessed into a digital back-end system of the FAST multi-beam receiver, and a data parallel method is adopted to realize the normal observation of spectral lines and pulsar data of the FAST and access into the FAST electric storm detection data, and search for FAST electric storm signals in real time;
s2, storing sample data which corresponds to the candidate body and is subjected to fast Fourier transform and integration;
s3: reading the sample data processed in the step S2 in real time, identifying and classifying the sample data by adopting a multi-beam feed source screening strategy, and determining a candidate;
the method for screening the candidates through the multi-beam feed source screening strategy specifically comprises the following steps:
after 19 computing nodes search candidate bodies in real time through GPU acceleration software, all candidate bodies are sent to a recloser process which runs on a head node all the time; in order to ensure that the searched candidates come from the same radio source, before the data enter the recloser process for further processing, the time of sending the node candidates to the head node is judged, and if the time of the candidates reaching the head node among different nodes is less than 10 seconds, the head node collects all the candidate data for further processing; if the arrival time exceeds 10 seconds, reporting errors in the data record; the recloser process compiles all candidates from each compute node into one file at a given time and adds a radio frequency interference RFI flag, determining the probability of candidate being RFI by examining candidate recloses across multiple beams; then the candidate screening process on the head node further screens the candidate according to the beam number, the signal-to-noise ratio S/N and the signal pulse width of the detected candidate;
a beam mask is generated based on the detected candidate beam numbers, by the following mechanism: if a candidate signal is detected in the beam, a flag bit is set to 1; otherwise, the flag bit is set to 0; generating a 19-bit binary beam mask, adding 2-bit RFI discrimination bits to the highest bit of the mask, and generating a valid mask table, wherein if the searched candidate beam number is on the table, the candidate beam number is not an RFI signal;
the candidates of the 19 calculation nodes are further screened on the head node, and the post_acc parameter value is further set to reduce the time resolution of the candidate data, wherein the frequency resolution and the time resolution of the candidate data are calculated as follows:
representing frequency resolution, +.>The bandwidth of the rapid radio storm detection data is represented, and the value of the bandwidth is 500MHz; />The number of spectrum channels is represented, and the value is 4096; finally, the frequency resolution of the signal obtained by calculation is 122KHz; />Representing the initial time resolution, and finally calculating to obtain the signal initial time resolution of 8.192 mu s;
the final time resolution of the signal isWherein->The integral length of the data after the fast Fourier transform on the ROACH2 plate is 11; />The length of the data integration which is done before the candidate signal is further screened on the head node is 3; and calculating to obtain the final signal time resolution of 270 mu s.
2. The method for detecting the rapid electric storm in real time according to claim 1, wherein in the step S1, the rapid electric storm detection data are 19 wave beam dual polarized feed source data in a feed source bin, the rapid electric storm detection data are accessed into a digital back end system of a FAST multi-wave beam receiver through optical fibers, the digital back end system of the FAST multi-wave beam receiver comprises 10 ROACH2 boards, each ROACH2 board is provided with 2 analog-to-digital converters (ADCs), each analog-to-digital converter (ADC) is accessed into single wave beam dual polarized data, the data bandwidth is 500MHz, then the sampled data are integrated for 11 times after 4096 points of FAST Fourier transformation, and finally the data are packaged into User Datagram Protocol (UDP) data packet streams.
3. The method for real-time detection of a rapid radio storm according to claim 2, wherein the encapsulated UDP packet stream is passed through a multicast mechanism of a network card to realize data replication and propagation, and then the UDP packet stream is sent to 19 compute node servers through a high-speed 40/100GbE ethernet switch.
4. A method of real-time detection of a rapid electric storm according to claim 3 wherein in step S2 19 computing node servers having fixed IP and mac addresses set up receive data sent from a 40/100GbE ethernet switch at a data rate of 1GB/S per path.
5. The method according to claim 4, wherein in step S2, three data processing threads are provided on each computing node to receive and process the UDP packet stream, and first, the network thread is responsible for receiving and extracting the UDP packets and storing them in a designated input ring buffer; secondly, converting the two-path polarization data into a Stokes polarization form by a calculation thread, and writing the Stokes polarization form into an output ring buffer area; third, the output thread reads the data from the output ring buffer, stores it in SIGPROC filterbank data format, and stores it in the compute node's cache in preparation for incoherent dispersion.
6. The method of claim 5, wherein the SIGPROC filterbank data stream in each compute node server cache is received into the GPU acceleration software on the node for real-time incoherent dispersion cancellation.
7. The method according to claim 1, wherein in step S2, the real-time large dispersion range pulse search is performed to search for single pulses with dispersion values between 20-10000, the efficiency step size is 25%, and then the head node performs candidate screening according to the distribution of single pulses in 19 beams.
8. The method according to claim 1, wherein in step S3, the recloser on the head node determines in real time whether 5 or more than 5 of the 19 beams are simultaneously pulsed, and if such candidates are detected, determines radio frequency interference RFI to exclude ground RFI interference.
9. The method for real-time detection of a rapid radio storm according to claim 1 or 2, wherein the following determination conditions are set in the subsequent screening process for candidates from which ground RFI interference has been excluded:
S/N（Signal to Noise ratio, S/N）>10；
detecting the candidate adjacent beam quantity < =4;
maximum pulse width <128 milliseconds;
the number of allowed events/sec 2.
10. The method according to claim 9, wherein in step S3, if the candidate is not screened by the recloser process, the candidate attribute detected by different beams is continuously and comprehensively determined in real time by the candidate screening thread on the head node, and the candidate is further screened according to the beam number, S/N, signal pulse width, and the known pulsar, RFI distribution and interference frequency band of the aircraft in the field of view of the candidate, and compared with the signal frequency of the searched candidate.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310728309.1A CN116482449B (en) | 2023-06-20 | 2023-06-20 | Real-time detection method for fast electric storm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310728309.1A CN116482449B (en) | 2023-06-20 | 2023-06-20 | Real-time detection method for fast electric storm |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116482449A CN116482449A (en) | 2023-07-25 |
| CN116482449B true CN116482449B (en) | 2023-09-08 |
Family
ID=87223549
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310728309.1A Active CN116482449B (en) | 2023-06-20 | 2023-06-20 | Real-time detection method for fast electric storm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116482449B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117556266B (en) * | 2024-01-11 | 2024-03-22 | 之江实验室 | Signal identification model training method, signal identification method and device |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2016097832A1 (en) * | 2014-12-16 | 2016-06-23 | Kamil Idris | Communication between radio terminals on an extraterrestrial body using a space based component and an ancillary component located on the extraterrestrial body. |
| CN106771653A (en) * | 2016-11-25 | 2017-05-31 | 中国科学院新疆天文台 | A kind of sudden and violent real-time detection devices, systems and methods of Rapid Radio |
| CN109657594A (en) * | 2018-04-18 | 2019-04-19 | 贵州师范大学 | A kind of radio pulsar identification auxiliary method being folded into figure based on secondary data |
| CN109991479A (en) * | 2019-03-22 | 2019-07-09 | 中国科学院新疆天文台 | The sudden and violent real-time detection apparatus, system and method for the Rapid Radio of multibeam receiver |
| CN110427878A (en) * | 2019-07-31 | 2019-11-08 | 中国科学院新疆天文台 | A kind of sudden and violent signal recognition method of Rapid Radio and system |
| CN111079608A (en) * | 2019-12-09 | 2020-04-28 | 中国科学院新疆天文台 | Quick radio storm real-time searching method |
-
2023
- 2023-06-20 CN CN202310728309.1A patent/CN116482449B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2016097832A1 (en) * | 2014-12-16 | 2016-06-23 | Kamil Idris | Communication between radio terminals on an extraterrestrial body using a space based component and an ancillary component located on the extraterrestrial body. |
| CN106771653A (en) * | 2016-11-25 | 2017-05-31 | 中国科学院新疆天文台 | A kind of sudden and violent real-time detection devices, systems and methods of Rapid Radio |
| CN109657594A (en) * | 2018-04-18 | 2019-04-19 | 贵州师范大学 | A kind of radio pulsar identification auxiliary method being folded into figure based on secondary data |
| CN109991479A (en) * | 2019-03-22 | 2019-07-09 | 中国科学院新疆天文台 | The sudden and violent real-time detection apparatus, system and method for the Rapid Radio of multibeam receiver |
| CN110427878A (en) * | 2019-07-31 | 2019-11-08 | 中国科学院新疆天文台 | A kind of sudden and violent signal recognition method of Rapid Radio and system |
| CN111079608A (en) * | 2019-12-09 | 2020-04-28 | 中国科学院新疆天文台 | Quick radio storm real-time searching method |
Non-Patent Citations (1)
| Title |
|---|
| 基于FAST望远镜的地外文明共时观测;张志嵩;张海燕;朱岩;金乘进;张同杰;李菂;朱丽春;;深空探测学报(第02期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116482449A (en) | 2023-07-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116482449B (en) | Real-time detection method for fast electric storm | |
| MacMahon et al. | The breakthrough listen search for intelligent life: a wideband data recorder system for the Robert C. Byrd Green Bank Telescope | |
| Allison et al. | Design and performance of an interferometric trigger array for radio detection of high-energy neutrinos | |
| US6043771A (en) | Compact, sensitive, low power device for broadband radar detection | |
| CN109991479B (en) | Fast radio storm real-time detection device, system and method of multi-beam receiver | |
| CN106226761A (en) | A kind of high-performance is concerned with higher-frequency radar multifrequency detection method | |
| CN107329127B (en) | A kind of phase linearity analysis method and system for radar system DBF Function detection | |
| CN111257655B (en) | Intercepted distance testing device for radio frequency sensor | |
| Men et al. | Piggyback search for fast radio bursts using Nanshan 26 m and Kunming 40 m radio telescopes–I. Observing and data analysis systems, discovery of a mysterious peryton | |
| US7719457B1 (en) | Digitally tuned digital radio frequency memory | |
| US11131742B2 (en) | Reconfigurable analog-to-digital conversion sampling of antennas for phase interferometry | |
| CN113438042B (en) | Real-time electromagnetic environment monitoring system and method | |
| CN111487473B (en) | Automatic detection system and method for rapid radio storm | |
| Clarke et al. | A multi-beam radio transient detector with real-time de-dispersion over a wide dm range | |
| Sabri et al. | The significance of e-CALLISTO system and construction of log periodic dipole antenna (LPDA) and CALLISTO system for solar radio burst study | |
| Nakajima et al. | New Nobeyama Radio Heliograph | |
| CN115801144B (en) | Passive reconnaissance system | |
| US7062001B2 (en) | Adaptive multi-channel, multi-function | |
| Petrin et al. | Measurement and analysis of urban spectrum usage | |
| Bosquez et al. | System for receiving NOAA meteorological satellite images using software defined radio | |
| EP3384311A1 (en) | Method for calibrating an electronically scanned sector antenna and corresponding measuring device | |
| CN115632710A (en) | Coherent laser communication terminal internal field test method | |
| CN111504362B (en) | Multifunctional processing system and method for observation signals | |
| CN214591471U (en) | Low-cost miniaturized full-airspace full-frequency-band signal interception device | |
| Marsh | Phased array digital beamforming algorithms and applications |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |