Pulse interval analysis device and pulse interval analysis method

Abstract

To provide a pulse interval analyzer capable of accurately identifying a target signal even under an interference environment.SOLUTION: A pulse interval analysis device includes a congestion separation unit, a pulse interval analysis unit, a database, an automatic identification unit, and a display control unit. The congestion separation unit separates a plurality of pulse signals in a wireless band into individual pulse signals. The pulse interval analyzer analyzes the time-series pulse intervals of the separated individual pulse signals and generates, for each of the pulse signals, a pulse interval change pattern that reflects the specifications of the pulse signal. The database holds a table in which a correspondence between a known signal and a pulse interval change pattern reflecting a radio wave specification value is registered. The automatic identification part collates the pulse interval change pattern with the contents of the table to identify individual pulse signals. The display control unit visually displays each pulse interval change pattern and the identification result for each pulse signal.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] FIELD Embodiments of the present invention relate to a pulse interval analysis device and a pulse interval analysis method. [Background technology]

[0002] Radar repeatedly transmits pulse signals (rectangular waves). The specifications of the pulse signals (radio wave specifications: frequency, pulse interval, etc.) vary depending on the radar and its purpose, and have unique specification values. On the other hand, by measuring the radio wave specifications of the pulse signal, it is possible to identify or estimate the radar equipment that is the source of the radio waves.

[0003] The first method for measuring the radio wave characteristics of a pulse signal is to use an oscilloscope. An oscilloscope visually displays the temporal change in the potential of the input signal in a two-dimensional graph, allowing parameters such as amplitude and pulse interval to be measured from the screen. Alternatively, a PI (Pulse Interval) raster can be used. The PI raster displays the input signal in a time direction according to a preset time width, visually indicating the timing of pulse arrival.

[0004] For example, these devices can be used to determine the parameter values ​​of a pulse signal, and the source of the signal can be identified based on the results. For example, if the parameter values ​​of a received pulse signal are the same as or similar to signal A registered in a known database, it can be assumed that the received pulse signal was transmitted from the same source as signal A. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Japanese Patent Application Laid-Open No. 2015-028447 Summary of the Invention [Problem to be solved by the invention]

[0006] In recent years, the radio wave characteristics of pulse signals have become more complex, making it difficult to quantify the characteristic values. In particular, the change patterns of pulse intervals have become significantly more complex. If the change patterns of pulse intervals are random and lack regularity, it is not possible to express their characteristics numerically. With existing technology, signal identification is performed manually by the user, but in some cases, signals that have become too complex cannot be identified.

[0007] In addition, it is rare to receive only one pulse signal; in most cases, multiple pulse signals are received simultaneously from multiple sources within the reception bandwidth. Existing technologies are vulnerable to radio wave environments that include interference, and measurement of radio wave parameters has been left to the user's painstaking manual separation and measurement.

[0008] As described above, it is difficult to measure and identify a target signal with high accuracy in today's circumstances where radio wave specifications are complicated, not to mention in poor radio wave environments such as interference. Therefore, an object of the present invention is to provide a pulse interval analyzer and a pulse interval analysis method that can accurately identify a target signal even in an interference environment. [Means for solving the problem]

[0009] According to an embodiment, a pulse interval analysis device includes a congestion separation unit, a pulse interval analysis unit, a database, an automatic identification unit, and a display control unit. The congestion separation unit separates multiple pulse signals in a radio band into individual pulse signals. The pulse interval analysis unit analyzes the time-series pulse intervals of the separated individual pulse signals and generates, for each individual pulse signal, a pulse interval change pattern that reflects the specifications of the pulse signal. The database holds a table that registers correspondences between known signals and pulse interval change patterns that reflect radio wave specification values. The automatic identification unit compares the pulse interval change pattern with the contents of the table to identify each individual pulse signal. The display control unit visually displays the pulse interval change pattern and the identification results for each individual pulse signal. [Brief explanation of the drawings]

[0010] [Figure 1] FIG. 1 is a functional block diagram showing an example of a pulse interval analyzer according to an embodiment. [Figure 2] FIG. 2 is a diagram for explaining a state in which a plurality of pulse signals interfere with each other. [Figure 3] FIG. 3 is a diagram for explaining the processing of the congestion separating unit 3. In FIG. [Figure 4] FIG. 4 is a diagram showing an example of a pulse interval change pattern. [Figure 5] FIG. 5 is a diagram showing a pulse interval change pattern of a pulse signal in which pulses having four different pulse intervals appear periodically. [Figure 6] FIG. 6 is a diagram showing pulse intervals between a plurality of forward pulses and a plurality of backward pulses of a pulse pi. [Figure 7] FIG. 7 is a diagram for explaining signal identification by pattern matching. [Figure 8] FIG. 8 is a diagram showing an example of a display of a pulse interval change pattern of a plurality of pulse signals. [Figure 9] FIG. 9 is a flowchart showing an example of a processing procedure in the congestion separating unit 3. [Figure 10] FIG. 10 is a flowchart showing an example of a processing procedure in the pulse interval analyzer 4. [Figure 11] FIG. 11 is a flowchart showing an example of a processing procedure in the automatic identification unit 5. [Figure 12] FIG. 12 is a flowchart showing an example of a processing procedure in the display control unit 6. DETAILED DESCRIPTION OF THE INVENTION

[0011] Next, an embodiment will be described with reference to the drawings. Radars are mounted on aircraft, ships, etc., and emit radio waves (pulse signals) sequentially at certain time intervals depending on their purpose. In this embodiment, it is assumed that there are multiple such radars, and that pulse signals with different specifications transmitted from each radar arrive at a pulse interval analyzer in an interference state.

[0012] 1 is a functional block diagram showing an example of a pulse interval analyzer according to an embodiment. The pulse interval analyzer includes a receiving antenna 1, a receiving unit 2, a congestion separation unit 3, a pulse interval analyzer 4, an automatic identification unit 5, a display control unit 6, a monitor 7, and a database 8. In Fig. 1, a radio band pulse signal captured by a receiving antenna 1 undergoes demodulation processing by a receiving unit 2 to become a received pulse signal. The received pulse signals are sequentially sent to a congestion separation unit 3. The congestion separation unit 3 performs congestion separation processing on the received pulse signal output from the receiving unit 2, separating it into individual pulse signals.

[0013] Figure 2 is a diagram illustrating a state in which multiple pulse signals are interfering with each other. As shown in Figure 2(a), pulse signals 1, 2, and 3 each have different specifications, and can be said to be emitted from a different emission source. The pulse intervals of pulse signal 1 are generally irregular, but the pulse amplitude (PA) is constant. Pulse signal 2 is emitted at a constant pulse interval, but the pulse amplitude (PA) fluctuates. The pulse intervals of pulse signal 3 are constant, and so is the pulse amplitude (PA).

[0014] When these pulse signals reach the receiving antenna 1, they are received as a series of signals superimposed on the time axis, as shown in Figure 2(b).

[0015] 3 is a diagram for explaining the processing of the congestion separation unit 3. The congestion separation unit 3 first searches for adjacent candidate pulses for all pulses (FIG. 3(a)). That is, the congestion separation unit 3 selects a certain pulse in the received pulse signal as a reference pulse. Then, it calculates the pulse interval times between the pulse group located ahead on the time axis and the pulse group located behind.

[0016] In the congestion separation process, a constraint can be imposed that pulses be transmitted at time intervals that follow a specific rule. Based on this constraint, it is possible to determine whether a certain pulse is a candidate pulse adjacent to a reference pulse based on a preset upper and lower limit of the pulse interval time.

[0017] A pulse whose pulse interval time exceeds the upper limit may be connected to the reference pulse via another pulse, but it will not be directly adjacent to the reference pulse. Similarly, a pulse whose pulse interval time is below the lower limit will not be adjacent. Note that such pulses may be split pulses, and may originally be the same pulse as the reference pulse, so they may be merged depending on the situation. The congestion separation unit 3 performs this search process for adjacent candidate pulses for all pulses in the received pulse signal, changing the reference pulse.

[0018] Like the pulse interval time, the pulse amplitude (PA) does not vary significantly from one emission source to another. Therefore, pulse signals can be separated based on the pulse amplitude (PA) value (Figure 3(b)). However, because power attenuation occurs depending on the receiving environment, the amplitude will not be the same even for the same signal. Here too, by setting a reference pulse and comparing the ratio of the amplitude value with other pulses with a threshold, it is possible to determine whether the signal is the same or a different signal.

[0019] Next, the congestion separation unit 3 concatenates adjacent candidate pulses to create one pulse train (FIG. 3(c)). In other words, the congestion separation unit 3 repeats the process of concatenating adjacent candidate pulses from the leading pulse, and concatenating the next adjacent candidate pulse from the concatenated pulse.

[0020] If there are multiple adjacent candidate pulses, the congestion separation unit 3 calculates the median of the pulse intervals from the histogram of the pulse intervals of the received pulse signal (Fig. 3(d)), and selects the pulse with the pulse interval closest to the median as the next pulse to be concatenated. The concatenated pulse train is considered to be a single pulse signal, and the above procedure is repeated for the remaining pulse signals that were not concatenated. By performing this series of steps in the congestion separation process, individual independent pulse signals can be obtained from a signal containing a mixture of multiple pulse signals.

[0021] Returning to FIG. 1, the explanation will be continued. The pulse interval analyzer 4 generates an image pattern (pulse interval change pattern) that represents the change in pulse interval over time from each pulse signal after convergence separation. In formulating this, all pulses in an arbitrary pulse signal are expressed as P={p i |i=1,2,...,N}, where N is the total number of pulses in the pulse signal P. Also, the arrival time of each pulse in the pulse signal P is expressed as T={t i |i=1,2,...,N}. The pulse arrival time can be either absolute time or relative time.

[0022] 4 is a diagram showing an example of a pulse interval change pattern. The processing of the pulse interval analyzer 4 will be described with reference to FIG. 4. In the oscilloscope display of FIG. 4(a), an arbitrary i-th pulse p i When we focus on p i is the pulse p located one step ahead i-1 and the pulse p located one behind i+1 have.

[0023] Considering a coordinate space in which the first axis is the pulse interval time with the forward pulse (Pre-PI axis) and the second axis is the pulse interval time with the backward pulse (Post-PI axis), pulse p i The position of is (x,y)=(t i -t i-1 ,t i+1 -t i Based on this result, the points in Figure 4(b) can be plotted as the pulse interval change pattern.

[0024] By changing the range of i = 1, 2, ..., N, a set of coordinate points corresponding to the entire received pulse signal is obtained in the coordinate space. This set of coordinate points is the result of converting the pulse interval changes inherent in the pulse signal into an image pattern (pulse interval change pattern).

[0025] Figure 5 shows an example of the pulse interval change pattern for a pulse signal in which pulses with four different pulse intervals appear periodically. Four plots like those in Figure 5(b) are obtained for the oscilloscope display in Figure 5(a).

[0026] In Figures 4 and 5, a pulse interval change pattern was created for one pulse from the time interval between one pulse before and one pulse after it. Next, we consider extending this to multiple pulses. For one pulse, the pulse intervals between each of the multiple forward pulses and each of the multiple backward pulses are calculated. For the i-th pulse p of the pulse signal P, i For all forward pulses, the pulse interval is (t i -t1,t i -t2,…,t i -t i-1 ) and the pulse interval between all the subsequent pulses is (t i+1 -t i ,t i+2 -t i ,…,t N -t i ) are Ni pulses. i There are (i-1) x (Ni) combinations of forward and backward pulses, and each combination corresponds to a coordinate point as shown in FIG. 6 and Table 1. FIG. 6 is a diagram showing the pulse intervals between multiple forward and backward pulses of pulse pi with pulse number i. Table 1 is a list showing the coordinate points of pulse pi with pulse number i.

[0027] [Table 1] Such coordinate points exist in the range of i=1, 2, . . . , N, so that N×(i−1)×(Ni) coordinate points are plotted in the coordinate space for the pulse signal P.

[0028] Elements of the coordinate point set of the pulse signal P may indicate the same coordinate point, and this is counted as the number of coordinate points (weight). Since a specific pulse interval change pattern that appears in the pulse signal appears repeatedly, coordinate points that correspond to the pattern have a large weight, and coordinate points that do not correspond to the pattern have a small weight. When plotting coordinate points in coordinate space, if they are displayed in shades according to the magnitude of the weight, coordinate points that correspond to the pattern will have a high density, and points that do not correspond to the pattern will have a low density, which makes it possible to eliminate noise.

[0029] Returning to Figure 1, we will continue the explanation. Database 8 stores a table in which known signals are registered in correspondence with pulse interval change patterns that reflect radio wave specification values. Automatic identification unit 5 automatically compares the characteristics of the pulse signal converted into a pulse interval change pattern (coordinate point distribution pattern) with reference pulse interval change patterns registered in advance in database 8, and discriminates the match rate using a predetermined threshold value to identify the received pulse signal. In other words, if the measured value of the input signal is the same as or similar to the specification values ​​of a certain signal A in database 8, it can be assumed that the received pulse signal is a known signal and was transmitted from the same emission source as signal A.

[0030] FIG. 7 is a diagram for explaining signal discrimination by pattern matching. As a method for discriminating between pulse interval variation patterns, pattern matching using the outer peripheral shape of the pattern can be performed. In this process, the automatic discrimination unit 5 discriminates between any coordinate point (x i ,y i ), the angles between the coordinate points and all other coordinate points are calculated, and the coordinate point with the largest outer periphery shape is determined. This process is repeated for all coordinate points, and the combination of coordinate points determined to have the largest angle is saved as a feature representing the outer periphery shape.

[0031] Then, the automatic identification unit 5 compares the features of all pulse interval change patterns registered in the database 8, and calculates the match rate by dividing the number of matching / similar coordinate points by the number of coordinate points possessed by the features.

[0032] Such pattern matching is possible by registering pulse interval change patterns in advance in database 8 and deriving feature quantities in the same way as for input signals. As signals registered in the database of received pulse signals and pulse interval change patterns to be registered in database 8, in addition to ideal target signals, patterns of imperfect pulse signals such as patterns in which the coordinate points are shifted or in which pulses are missing / omitted may be registered.

[0033] When receiving a pulse signal, whether intentional or not, there may be fluctuations in the arrival time of the received pulse signal, or pulse cracking or pulse leakage may occur, disrupting the pulse interval pattern. By registering these imperfect patterns, it is possible to automatically identify imperfect signals when they are received.

[0034] Continuing the explanation, returning to Fig. 1, the display control unit 6 visually displays the pulse interval change pattern and the automatic identification result for each of the multiple pulse signals included in the received pulse signal on the monitor 7, and presents this to the user.

[0035] FIG. 8 is a diagram showing an example of a display of a pulse interval change pattern of a plurality of pulse signals. FIG. 8 is a diagram showing an example of a pulse interval change pattern. As shown in FIG. 8( a), the pulse interval change pattern is displayed in a coordinate space obtained by rotating the XY coordinate space by 45 degrees counterclockwise. In the rotated display, if a line extending vertically from the origin (a vertical line) is drawn, coordinate points on the vertical line will have equal X and Y coordinate values. In other words, coordinate points on the vertical line indicate pulses whose leading pulse interval time and trailing pulse interval time are equal, and the pulse interval time increases the farther from the origin. Furthermore, the tendency of the pulse interval change pattern can be seen at a glance: in the region to the right of the vertical line, the leading pulse interval time is longer than the trailing pulse interval time, and in the region to the left of the vertical line, the trailing pulse interval time is longer than the leading pulse interval time.

[0036] Figure 8(b) shows an example of multiple pulse interval change patterns displayed simultaneously using this drawing method. In an XY coordinate space rotated 45 degrees counterclockwise, the drawing area is extended vertically and each pattern is drawn. [PRI(max)], [PRI(center)], and [PRI(min)] on the left side of the drawing area are labels that indicate the maximum, median, and minimum values ​​of the pulse interval time within the drawing area, and the label values ​​and drawing range are updated according to the settings entered by the user.

[0037] Figure 8(b) shows an example of four different pulse signals displayed simultaneously, arranged vertically because the average pulse interval time in each pulse signal is different for each pattern. The circles surrounding each pattern are examples of the automatic identification results, and labels may also be attached to each circle.

[0038] Received pulse signals are broadly divided into known signals that are registered in database 8 and unknown signals that are not registered in database 8. If the automatic identification result is a known signal, the signal name associated with the name registered in database 8, the match rate, and the measurement results of the radio wave specifications are displayed as a label. If the automatic identification result is an unknown signal, the signal name is unknown, so a label such as "unknown" can be added. Note that if signals that pose a threat are known in advance, known signals can be classified into threat signals and other signals, and an explicit label can be added to the threat signals.

[0039] By viewing a display such as that shown in Figure 8 on the monitor 7, the user can comprehensively check the pulse interval change patterns of multiple signals and the automatic identification results. After checking the results, the user can register the unknown signal along with the measured radio wave parameters in the database 8, or can eliminate known unnecessary signals and focus only on signals of interest. In addition, when signals are being observed in real time, the user can be notified by a warning display or alarm sound when a threat signal is automatically identified.

[0040] 9 is a flowchart showing an example of a processing procedure in the congestion separation unit 3. The congestion separation unit 3 reads received pulse signal data (number of pulses M) (step S11). Then, after initializing an index i=1 (step S12), the congestion separation unit 3 searches for adjacent candidate pulses ahead and behind the pulse i behind the received pulse signal from the upper and lower limits of the pulse interval time (step S13). The congestion separation unit 3 also searches for adjacent candidate pulses ahead and behind the pulse i behind the received pulse signal from the amplitude value of pulse number i (step S14). The processing of steps S13 and S14 is repeated while incrementing i (step S15) until i>M (step S16).

[0041] Next, the congestion separation unit 3 creates a PI histogram from the pulse interval time of the received pulse signal (step S17). After initializing an index i=1 (step S18), the congestion separation unit 3 concatenates pulse number i of the received pulse signal with an adjacent candidate pulse that approximates the median of the PI histogram (step S19), and excludes the concatenated pulse from the adjacent candidate pulses (step S20). The processes of steps S19 and S20 are repeated while incrementing i (step S15) until i>M (step S16).

[0042] 10 is a flowchart showing an example of a processing procedure in the pulse interval analyzer 4. The pulse interval analyzer 4 reads received pulse signal data (number of pulses N) (step S31). After initializing an index i to 1 (step S32), the pulse interval analyzer 4 measures all leading pulse interval times of pulse number i of the received pulse signal (step S33) and measures all trailing pulse interval times of pulse number i of the received pulse signal (step S34). The pulse interval analyzer 4 then obtains a coordinate point in the XY coordinate space from all combinations of pulse interval times of pulse number i of the received pulse signal (step S35). The processing of steps S33, S34, and S35 is repeated while incrementing i (step S36) until i>N (step S37).

[0043] Next, the pulse interval analyzer 4 counts the number of each coordinate point from the set of coordinate points of the received pulse signal (step S38), removes coordinate points with a small number of coordinate points as noise (step S39), and records the remaining set of coordinate points as a pulse interval change pattern (step S40).

[0044] 11 is a flowchart showing an example of a processing procedure in the automatic identification unit 5. The automatic identification unit 5 reads the pulse interval change pattern (number of coordinate points L) (step S51). Then, after initializing an index i to 1 (step S52), the automatic identification unit 5 calculates the angles between the i-th coordinate point and all other coordinate points (step S53) and selects the coordinate point with the largest outer circumferential shape (step S54). The processing of steps S53 and S54 is repeated while incrementing i (step S55) until i>L (step S56).

[0045] Then, the automatic identification unit 5 acquires a set of coordinate points that represent the outer periphery of the pulse interval variation pattern (step S57), and reads the pulse interval variation pattern (number of patterns D) from the database 8 (step S58).

[0046] After initializing the index i to 1 (step S59), the automatic classification unit 5 calculates the match rate with the i-th pulse interval variation pattern in the database (step S60). The process of step S60 is repeated while incrementing i (step S61) until i>D (step S62). The automatic classification unit 5 then adopts the pulse interval variation pattern with the highest match rate as the classification result (step S63).

[0047] 12 is a flowchart showing an example of a processing procedure in the display control unit 6. The display control unit 6 rotates the XY coordinate space by 45 degrees counterclockwise (step S71). Next, the display control unit 6 reads the pulse interval change pattern (number of patterns P) (step S72). Then, after initializing an index i to 1 (step S73), the display control unit 6 rotates the set of coordinate points of the pulse interval change pattern of the i-th received pulse signal by 45 degrees counterclockwise (step S74).

[0048] Next, the display control unit 6 specifies a density according to the number of coordinate points of the coordinate point set of the pulse interval change pattern (step S75). Then, the display control unit 6 draws the coordinate point set of the pulse interval change pattern of the i-th received pulse signal in the XY coordinate space (step S76). The processes of steps S74, S75, and S76 are repeated while incrementing i (step S77) until i>P (step S78). Then, the display control unit 6 superimposes and displays the automatic identification result on the pulse interval change pattern (step S79).

[0049] As described above, in the embodiment, in a radio interference environment in which a large number of pulse signals are transmitted from a plurality of radars, the plurality of pulse signals are first separated into individual pulse signals. Then, each pulse signal is converted into a coordinate point set pattern in a two-dimensional coordinate space based on the change characteristics of the pulse intervals. After the converted pattern is automatically identified, the pattern and the identification results are comprehensively presented to the user.

[0050] That is, the congestion separation unit 3 separates the multiple incoming pulse signals into individual pulse signals. For each separated pulse signal, the pulse interval analysis unit 4 converts the changes in pulse intervals into a pattern image. The received pulse signal is then compared with similar reference pattern images registered in advance in a database 8, and the automatic identification unit 5 performs automatic identification of the received pulse signal. Finally, the display control unit 6 displays the pattern images corresponding to all the received pulse signals together with the identification results on one screen.

[0051] With this configuration, even in an interference-ridden radio wave environment, the radio wave characteristics of a complex pulse signal can be automatically measured and identified, and the results can be presented to the user. The user can comprehensively check the results of the automatic identification of multiple pulse signals, enabling real-time situational awareness of the radio wave environment to be updated. As a result, according to the embodiments, it is possible to provide a pulse interval analysis device and a pulse interval analysis method that can accurately identify a target signal even in an interference-ridden environment.

[0052] Although an embodiment has been described, this embodiment is presented as an example and is not intended to limit the scope of the invention. This novel embodiment can be embodied in various other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. This embodiment and its modifications are included within the scope and spirit of the invention, and are also included in the inventions described in the claims and their equivalents. [Explanation of symbols]

[0053] 1...receiving antenna, 2...receiving unit, 3...congestion separation unit, 4...pulse interval analysis unit, 5...automatic identification unit, 6...display control unit, 7...monitor, 8...database.

Claims

1. a congestion separation unit that separates a plurality of pulse signals in a radio band into individual pulse signals; a pulse interval analysis unit that analyzes pulse intervals of the time series of the separated individual pulse signals and generates a pulse interval change pattern for each of the individual pulse signals, the pattern reflecting specifications of the pulse signals; a database that holds a table that registers correspondence between known signals and pulse interval change patterns that reflect radio wave specification values; an automatic identification unit that identifies each of the pulse signals by comparing the pulse interval change pattern with the contents of the table; a display control unit that visually displays the pulse interval change pattern and the result of the classification for each of the individual pulse signals.

2. the database holds a reference pattern image obtained by converting the pulse interval change pattern serving as a reference into an image; 2. The pulse interval analyzer according to claim 1, wherein the automatic identification section identifies the individual pulse signals by comparing a pattern image obtained by visualizing the pulse interval change pattern with the reference pattern image.

3. a pulse interval analyzer separating a plurality of pulse signals in a radio band into individual pulse signals; a step in which the pulse interval analyzer analyzes pulse intervals of the time series of the separated individual pulse signals and generates a pulse interval change pattern for each of the individual pulse signals, the pattern reflecting the parameters of the pulse signals; a step in which the pulse interval analyzer stores in a database a table in which correspondences between known signals and pulse interval change patterns that reflect radio wave specification values ​​are registered; a step in which the pulse interval analyzer compares the pulse interval variation pattern with the contents of the table to identify the individual pulse signals; a step in which the pulse interval analyzer visually displays each pulse interval variation pattern and the results of the classification for each of the individual pulse signals.

Images