JP2009060239A - Signal analyzer - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a signal analyzer which estimates the number of sequences correctly even if there are a plurality of periodic signal sequences of identical incoming time interval. <P>SOLUTION: In order to estimate the number of sequences and the incoming time interval of the period signal of radiation signals based on the observation data including the time of a predetermined timing indicative of the reception timing of a plurality of radiation signals, the analysis range of radiation signal incoming time interval and each width or the numbers when it is divided into a plurality of evaluation value subdivision sections and count value subdivision sections are set (6), the incoming time difference of radiation signal is calculated from the time of a predetermined timing (1), corresponding evaluation value subdivision section and count value subdivision section are selected from the incoming time difference (2), an accumulation evaluation value is calculated by calculating an evaluation value from the incoming time difference and the time of a predetermined timing and then accumulating the evaluation value for every evaluation value subdivision section (3), a count value is calculated by counting the incoming time differences selected in a selected subdivision section for every count value subdivision section (4), and the number of sequences and the incoming time interval of the periodic radiation signals are estimated from the accumulation evaluation value and the count value (5). <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention estimates the number of series of signals (periodic signals) whose period of arrival and disappearance of signals are periodic from received signals in which radiation signals radiated from a plurality of radio wave sources and the like are mixed. The present invention relates to a signal analysis apparatus that estimates the period of arrival time when there is a periodic signal.

  Conventionally, in such a method for estimating the number of sequences of periodic signals, a method for obtaining the number of sequences of periodic signals and their period based on the frequency of the arrival time intervals, with the difference between the arrival times of the two signals as the arrival time intervals. It has been proposed (see, for example, Patent Document 1).

JP-A-56-80925

  In the conventional technology as described above, when the arrival time intervals are different, it is possible to correctly estimate the number of sequences of the periodic signal of the radiation signal. However, in the case where a plurality of periodic signal sequences having the same arrival time interval coexist, there is a problem that the number of sequences cannot be estimated correctly.

  Therefore, an object of the present invention is to provide a signal analysis device that can correctly estimate the number of sequences even in the case where there are a plurality of sequences of periodic signals having the same arrival time interval.

  According to the present invention, based on observation data including a predetermined timing time of a radiation signal indicating a reception timing of each radiation signal of a reception signal in which a plurality of radiation signals are mixed, the arrival time of the radiation signals is a periodic signal having a periodicity. In the signal analysis apparatus for estimating the number of series and the arrival time interval, the analysis range of the arrival time interval of the radiation signal, the width and number of the analysis range of the arrival time interval divided into a plurality of evaluation value subdivision sections, and the arrival time Based on the width and number of divisions of the interval analysis range into a plurality of count value subdivision sections, a setting unit for setting parameters used for estimating the number of sequences in the estimation unit described below, and two based on a predetermined timing time of the radiation signal An arrival time difference calculation unit that calculates a difference in arrival times of radiation signals and outputs an end flag when the difference in arrival times of all combinations of radiation signals has been calculated; and While calculating the evaluation value based on the difference between the arrival time and the predetermined timing time of the radiation signal, the subdivision section selection unit that selects the corresponding evaluation value subdivision section and the count value subdivision section based on the difference of time The evaluation value accumulating unit for accumulating the evaluation value for each evaluation value subdivision section, and calculating the accumulated evaluation value, and the number of arrival time differences selected in the selected subdivision section are counted for each count value subdivision section. A signal comprising: a count unit that calculates a count value; and an estimation unit that estimates the number of sequences of radiation signals whose arrival time is periodic and the arrival time interval based on the cumulative evaluation value and the count value In the analyzer.

  In the present invention, the estimation unit first estimates the number of provisional series of periodic signals and their arrival time intervals based on the accumulated evaluation value. Further, using the estimated arrival time interval and the estimated arrival time interval, the number of series of periodic signals of radiation signals having the same arrival time interval is estimated again. As a result, even if there are a plurality of sequences having the same arrival time interval, the number of sequences is estimated again using the count value. Therefore, even if a plurality of periodic signal sequences having the same arrival time interval are mixed, the estimation is correctly performed. be able to.

Prior to the description of the embodiment of the present invention, first, the principle of the present invention will be described.
Consider a case where a radiation signal is received from only one radio wave source (the number of series is 1). The radiation signal from the radio wave source is assumed to be a periodic signal whose signal arrival time is a periodic signal, and the interval between arrival times is r. The total number of received signals is N (N is an integer equal to or greater than 1), and serial numbers from 1 to N are assigned to the received signals in the order received (see FIG. 2). If the arrival time of the radiation signal n (1 ≦ n ≦ N) is t _n , it can be expressed as equation (1).

t _n = t ₁ + (n−1) · r (1)

Next, a difference between arrival times of the two radiation signals (difference between arrival times) is calculated, and an evaluation value is calculated using the calculated difference. Assuming that the difference in arrival time between the radiation signal n and the radiation signal m (n <m) is Δt _{n, m} , the difference in arrival time can be calculated by equation (2). Also, the evaluation value is calculated as in equation (3).

Δt _{n, m} = t _m −t _n (2)
p _{n, m} = exp (2πjt _n / Δt _{n, m} ) (3)

  Here, assuming that the radiation signal n and the radiation signal m are adjacent signals, that is, m = n + 1, Expression (2) and Expression (3) can be transformed into Expression (4) and Expression (5), respectively.

Δt _{n, (n + 1)} = t ₁ + n · r− (t ₁ + (n−1) · r) = r (4)
p _{n, (n + 1)} = exp (2πjt _n / Δt _{n, (n + 1)} )
= Exp (2πj (t ₁ + (n−1) · r) / r)
= Exp (2πjt ₁ / r) (5)

From the above, when two signals whose difference in arrival time of the two signals is equal to r are adjacent to each other, Expression (5) is obtained, and the evaluation value is equal regardless of the value of n. Become.
Therefore, two signal difference the calculated evaluation value p _n of arrival _times, to calculate the _m, further difference Delta] t n of arrival _time, evaluation value for each size of _m p _n, accumulated evaluation value by adding _m If Δt _{n, m} is equal to the period r, the absolute value of the cumulative evaluation value becomes a large value. On the other hand, since the evaluation value calculated | required by Formula (5) does not become mutually equal for the radiation signal whose arrival time is not periodic, the absolute value of a cumulative evaluation value does not become a big value. Therefore, when the absolute value of the cumulative evaluation value is large, it can be recognized that a periodic signal exists in the received signal, and the number of series of periodic signals can be estimated from the number of the absolute value of the cumulative evaluation value increasing. Furthermore, the interval of the periodic signals can be known from the difference in arrival time when the cumulative evaluation value becomes large.

Next, a method for estimating the interval of periodic signals and the number of sequences from the magnitude of the cumulative evaluation value will be described. Assume that the difference in arrival time and the evaluation value are calculated for the 1 to N radiated signals according to equations (2) and (3), and the cumulative evaluation value is calculated for each difference in arrival time. In this case, the time during which the radiation signal was received is (t _N −t ₁ ). If a periodic signal whose arrival time interval is r is always received within this time, the number of combinations of two radiation signals whose arrival time difference is r is ((t _N −t ₁ ) / r) = N. -1. On the other hand, since the evaluation value of the periodic signal is equal regardless of n as in Expression (5), the absolute value of the cumulative evaluation value is equal to the number of combinations (N−1) at which the difference in arrival time is r.

Therefore, a threshold is determined based on the number of combinations ((t _N −t ₁ ) / r) calculated using the time (t _N −t ₁ ) during which the radiation signal was received, and this is used to determine the periodic signal. Estimate the number of series. However, the cumulative evaluation value of the periodic signal described above is a case where the periodic signal can always be received. If the periodic signal cannot be received, the cumulative evaluation value becomes small. Therefore, a ratio α (α <1) at which a periodic signal can be received is set, and a value obtained by multiplying this by the number of combinations is used as a threshold, and the cumulative evaluation value when the difference in arrival time is r is α ((t _N If it is greater than -t ₁ ) / r), it is determined to be a periodic signal.

  By the threshold processing as described above, the number of series of periodic signals can be estimated from the number of accumulated evaluation values exceeding the threshold, and the interval of periodic signals can be known from the difference in arrival time at that time. The above explanation is for the case where the number of series of periodic signals is 1, but even if there are a plurality of series, the number of series of periodic signals and the interval can be estimated in the same manner as described above if the intervals are different. it can.

  In the above-described method, since the cumulative evaluation value is calculated for each difference in arrival time, the estimation of the number of sequences is erroneous when a plurality of periodic signals having the same period are mixed. In view of this, a method is proposed in which the number of sequences can be accurately estimated even in the case where a plurality of sequences having the same periodic signal interval coexist using the above-described method and the count number of the difference in arrival time.

  This will be described using a specific example. Now, for example, as shown in FIG. 3, consider a case where radiation signals are received from two transmission sources (the number of sequences is 2). The two radiation signals are periodic signals whose arrival time intervals are both r, and the radiation signal from the transmission source A is received first, and then the radiation signal from the radiation source B (both not shown) is received. And The total number of received signals is N, and serial numbers from 1 to N are given in the order in which the received signals are received. Then, the arrival time of the radiated signal from the transmission source A (referred to as sequence A) can be expressed as equation (6), and the arrival time of the radiated signal from the transmission source B (referred to as sequence B) can be expressed as equation (7). it can.

t _n = t ₁ + {(n−1) / 2} · r (when n is an odd number) (6)
t _n = t ₂ + (n / 2-1) · r (when n is an even number) (7)

  Next, as in equations (2) and (3), the difference between the arrival times of the two radiation signals (radiation signals n and m) is calculated, and further an evaluation value is calculated. In the case where they are not equal, the cumulative evaluation value does not increase. Therefore, here, a case where the difference in arrival time is equal to the interval will be considered. In the case where the difference in arrival time is equal to the interval, when n is an odd number (radiation signal of series A), n can be expressed as in equation (8) using integer h, where m is the equation (9)

n = 2h-1 (8)
m = n + 2 = 2h + 1 (9)

  By substituting these into equations (2) and (3) and calculating the difference in arrival time and the evaluation value, equations (10) and (11) are obtained.

Δt _{(2h−1), (2h + 1)} = (t ₁ + h · r) − (t ₁ + (h−1) · r) = r (10)
p _{(2h−1), (2h + 1)} = exp (2πj (t ₁ + (h−1) · r) / r)
= Exp (2πjt ₁ / r) (11)

  On the other hand, when n is an even number (series B radiation signal), n can be expressed as in Expression (12), and m at this time is expressed as Expression (13).

n = 2h (12)
m = n + 2 = 2h + 2 (13)

  By substituting these into equations (2) and (3) and calculating the difference in arrival time and the evaluation value, equations (14) and (15) are obtained.

Δt _{(2h), (2h + 2)} = (t ₂ + h · r) − (t ₂ + (h−1) · r) = r (14)
p _{(2h), (2h + 2)} = exp (2πj (t ₂ + (h−1) · r) / r)
= Exp (2πjt ₂ / r) (15)

The difference between the arrival times is the same whether n is an odd number (series A) or n is an even number (series B). Therefore, the evaluation values calculated by equations (11) and (15) are added together. It becomes a cumulative evaluation value. In the case of the following expression (16), the evaluation values p _{(2h−1), (2h + 1)} and p _{(2h), (2h + 2)} are added to zero, but the condition of expression (17) is satisfied. If it is satisfied, the absolute value obtained by adding p _{(2h-1), (2h + 1)} and p _{(2h), (2h + 2)} is 1 or more. Therefore, the interval between periodic signals can be known in many cases by a method similar to the threshold processing described above. In the following description, the description will be continued assuming that the accumulated evaluation value exceeds the threshold when the difference in arrival time is r.

t ₂ −t ₁ = r / 2 (16)
t ₂ −t ₁ ≦ r / 3, or t ₂ −t ₁ ≧ 2r / 3 (17)

Next, the count value will be described. The count value is obtained by counting the difference in arrival time calculated by the equation (2) for each size. Referring to FIG. 3 as an example, first, two radiation signals (# 1 and # 2, # 3 and # 4,..., # (N-1) and # where the difference in arrival time is (t ₂ −t ₁ ). Since the number of combinations of N) is N / 2, the count value at (t ₂ -t ₁ ) is N / 2. In addition, the number of combinations of the radiation signals (# 2 and # 3,... # (N-2) and # (N-1)) in which the difference in arrival time is (r−t ₂ + t ₁ ) is (N / Therefore, the count value is (N / 2-1). Similarly, (# 1 and # 4,... # (N−3) and # (N)) count values when the difference in arrival times is (r + (t ₂ −t ₁ )) is (N / 2-1). (# 2 and # 5,... # (N−4) and # (N)) in (2r−t ₂ + t ₁ ), the count value is (N / 2-2) or the like.

Here, since the interval of the arrival times is estimated to be r from the cumulative evaluation value, the number of signals of the periodic signal whose interval is r at the time (t _N -t ₁ ) during which the signal is received is considered. Assuming that N is an even number, the time during which a signal is received is given by the following equation.

t _N −t ₁ = (t ₂ + (N / 2-1) · r) −t ₁
= (T ₂ -t ₁ ) + (N / 2-1) · r (18)

Since the number of signals is obtained by dividing the time (t _N −t ₁ ) during which signals are received by r, the number of signals is (t ₂ −t ₁ ) / r + (N / 2-1).

This number of signals should theoretically be equal to the count value at the time of arrival (t ₂ −t ₁ ), which is the number of combinations of radiation signals with the difference of time of arrival being (t ₂ −t ₁ ). This is because the number of signals becomes one when there is one series of periodic signals with a period of r within the reception time. Therefore, a value obtained by multiplying the number of signals calculated from the reception time ((t ₂ −t ₁ ) / r + (N / 2-1)) by the parameter β is used as a threshold value, and compared with a count value having a difference in arrival time less than r. . If the count value exceeds the threshold value, it can be determined that there are a plurality of periodic signal sequences having the same interval, and if there is not, it can be determined that there are no periodic signals having the same interval. As for the number of series, when the number of count values exceeding the threshold is 2 (in the case of FIG. 3), the number of series is 2 if the sum of the differences in the arrival times of each other is equal to r. If the number of count values exceeded is 6, if two of the 6 arrival time differences are selected and the combination whose total value is r is 3, there are 3 periodic signals with the same interval. become. This is the principle of the present invention.

Embodiment 1 FIG.
Hereinafter, a signal analyzing apparatus according to the present invention will be described according to embodiments. FIG. 1 is a configuration diagram of a signal analysis apparatus according to Embodiment 1 of the present invention. FIG. 4 is a data flow diagram of the signal analyzer according to this embodiment. 1 and 4, the setting unit 6 includes an analysis range of the arrival time interval, a width and number obtained by dividing the analysis range of the arrival time interval into a plurality of evaluation value subdivided sections, and a plurality of analysis ranges of the arrival time interval. A threshold parameter used for estimating the width and number of divided count value subdivision sections and the number of periodic signal sequences is set. The arrival time difference calculation unit 1 is the observation data including the arrival time and the extinction time of each radiation signal of the reception signal in which a plurality of radiation signals observed by receiving radio waves with a receiver (not shown) are mixed. For example, regarding the arrival time of the radiation signal, the difference between the arrival times of any two radiation signals is calculated. Moreover, when the difference of the arrival times has been calculated for all combinations of radiation signals, an end flag is output.

  The subdivision section selection unit 2 selects a corresponding evaluation value subdivision section and count value subdivision section based on the magnitude of the difference of the arrival times, and further evaluate value accumulation section 3 corresponding to the selected evaluation value subdivision section. The arrival times and the like are output to the counting units 4-1 to 4 -G corresponding to the selected count value subdivision sections.

  The evaluation value accumulating units 3-1 to 3-K calculate an evaluation value by a predetermined method based on the difference between the arrival times and the arrival time, and calculate the accumulated evaluation value by accumulating the calculated evaluation values. . Further, the counting units 4-1 to 4-G calculate the count value by counting the number selected in the selected subdivision section. Based on the cumulative evaluation value and the count value, the estimation unit 5 estimates the number of radiated signal sequences with periodic arrival times and the arrival time interval.

The embodiment of the present invention will be described by taking as an example a case where the number of radiated signals included in a received signal is N and the arrival time of each signal is observed as t _n (1 ≦ n ≦ N).

  First, the setting unit 6 manually sets in advance the range of the arrival time interval of the radiation signal for calculating the cumulative evaluation value and the width and number when the range is divided into a plurality of ranges. The range of the arrival time interval is set assuming a periodic signal to be analyzed, and the width when the range is divided into a plurality of values is equal to the resolution of time information such as the arrival time of the radiation signal, A value obtained by multiplying the resolution by a certain value is used. In the following, the divided intervals of arrival time intervals are referred to as evaluation value subdivision intervals.

  Further, the setting unit 6 manually sets in advance the range of the difference in arrival time of the radiation signal for calculating the count value, and the width and number when the range is divided into a plurality of parts. The range of the difference in arrival time is set assuming a periodic signal to be analyzed, and the width when the range is divided into a plurality of values is equal to the resolution of time information such as the arrival time of the radiation signal, A value obtained by multiplying the resolution by a certain value is used. In the following, the divided interval of arrival times is referred to as a count value subdivision interval.

  Furthermore, the setting unit 6 manually sets parameters α, β, and γ related to thresholds required when estimating the number of sequences in the estimation unit 5 described later. These parameters are set assuming a periodic signal to be analyzed.

In the following description, it is assumed that the range of the arrival time interval for calculating the cumulative evaluation value is set to τ _min ⁽ⁱ⁾ or more and less than τ _max ⁽ⁱ⁾ and the width is divided into K by Δτ ⁽ⁱ⁾ . Assume that the range of the arrival time interval of the evaluation value subdivision section k (1 ≦ k ≦ K) is set to τ ⁽ⁱ⁾ _(k−1) or more and less than τ ⁽ⁱ⁾ _k . In the above description, the evaluation value subdivision sections do not overlap at all. However, the ranges of the evaluation value subdivision sections may be set to overlap each other.

In the following description, it is assumed that the range of arrival time differences for calculating the count value is set to τ _min ^(d) or more and less than τ _max ^(d) , and the width is divided into G by Δτ ^(d) . It is assumed that the range of the difference in arrival time of the count value subdivision interval g (1 ≦ g ≦ G) is set to τ ^(d) _(g−1) or more and less than τ ^(d) _g . In the above description, the count value subdivision sections do not overlap at all, but the ranges of the count value subdivision sections may be set to overlap each other.

The arrival time difference calculation unit 1 calculates the difference Δt _{n, m} of arrival times for any two signals (n, m) out of a plurality of received signals, using Equation (2) _, and calculates the calculated difference Δt _{n , M} and the arrival time t _n of the signal n are output to the subdivision section selection unit 2. However, the arrival time difference calculation unit 1 performs output to the subdivision section selection unit 2 only when the difference of the calculated arrival times is within the interval of arrival times set by the setting unit 6. Further, when the difference in arrival time has been calculated for all combinations of signals in the observation data, an end flag is output to the evaluation value accumulating units 3-1 to 3-K and the counting units 4-1 to 4-K. .

The subdivision section selection unit 2 selects a corresponding evaluation value subdivision section k (1 ≦ k ≦ K) based on the magnitude of the arrival time difference Δt _{n, m} output from the arrival time difference calculation unit 1. Since the range of the evaluation value subdivision section k is set to τ ⁽ⁱ⁾ _(k-1) or more and less than τ ⁽ⁱ⁾ _k by the setting unit 6, the evaluation value subdivision section k that satisfies the following equation (19) is selected. To do.

τ ⁽ⁱ⁾ _(k−1) ≦ Δt _{n, m} <τ ⁽ⁱ⁾ _k (19)

Then, the arrival time difference Δt _{n, m} and the arrival time t _n are output to the evaluation value accumulating unit 3-k corresponding to the evaluation value subdivision section k. In the case where the difference Delta] t _n, evaluation value accumulation unit 3-k corresponding to _m of the arriving time there are a plurality, in all evaluation value accumulation unit 3-k corresponding difference Delta] t n arrival _time, and _m The arrival time t _n is output.

Further, the subdivision section selection unit 2 selects the corresponding count value subdivision section g (1 ≦ g ≦ G) based on the magnitude of the arrival time difference Δt _{n, m} output from the arrival time difference calculation unit 1. . Since the range of the count value subdivision interval g is set by the setting unit 6 to be not less than τ ^(d) _{(g-1) and} less than τ ^(d) _g , the count value subdivision interval g satisfying the following equation (20) is selected. To do.

τ ^(d) _(g−1) ≦ Δt _{n, m} <τ ^(d) _g (20)

Then, the arrival time difference Δt _{n, m} is output to the counting unit 4-g corresponding to the count value subdivision section g. In the case where the difference Delta] t n of arrival _time, the count unit 4-g corresponding to _m there are a plurality, in all the count unit 4-g corresponding outputs a difference Delta] t _{n, m} of the arrival time.

In the evaluation value accumulating unit 3-k, the evaluation value _{pn, m} is calculated by Expression (3) using the arrival time difference Δt _{n, m} and the arrival time t _n output from the subdivision section selection unit 2. Furthermore, the calculated evaluation value is added to the cumulative value (cumulative evaluation value) of the evaluation value calculated so far by the evaluation value accumulating unit 3-k, and the cumulative evaluation value is updated. Moreover, when the end flag output when the calculation of the arrival time differences of all the radiation signals is completed is input from the arrival time difference calculation unit 1, the accumulation calculated by the evaluation value accumulation unit 3-k so far The evaluation value and the range (τ ⁽ⁱ⁾ _(k−1) , τ ⁽ⁱ⁾ _k ) of the evaluation value subdivision interval are output to the estimation unit 5.

In the counting unit 4-g, when the arrival time difference Δt _{n, m} output from the subdivision section selecting unit 2 is input, the count value that has been counted so far by the counting unit 4-g is counted up. Update the value. Further, when an end flag that is output when the calculation of the difference between the arrival times of all the radiation signals is completed is input, the count value counted by the count unit 4-g and the range of the count value subdivision section (τ ^(d) _(g−1) and τ ^(d) _g ) are output to the estimation unit 5.

In the estimation unit 5, the cumulative evaluation value and the range of evaluation value subdivision intervals (τ ⁽ⁱ⁾ _(k−1) , τ ⁽ⁱ⁾ _k ) are input from the evaluation value accumulating unit 3-k (1 ≦ k ≦ K). Then, the presence / absence of a periodic signal is first determined based on the absolute value of the cumulative evaluation value. As the method, the center value ^(c) τ ⁽ⁱ⁾ _k of the evaluation value subdivision section k is calculated by the equation (21), and the threshold th ⁽ⁱ⁾ _k in the evaluation value subdivision section k is calculated by the equation (21).

^(c) τ ⁽ⁱ⁾ _k = (τ ⁽ⁱ⁾ _(k−1) + τ ⁽ⁱ⁾ _k ) / 2 (1 ≦ k ≦ K) (21)
th ⁽ⁱ⁾ _k = α · (t _N −t ₁ ) / ^(c) τ ⁽ⁱ⁾ _k (1 ≦ k ≦ K) (22)

Here, α is a real number of 1 or less set by the setting unit 6. Next, the absolute value of the cumulative evaluation value output from each evaluation value accumulating unit 3-k is compared with the threshold value th ⁽ⁱ⁾ _k calculated above, and the absolute value of the cumulative evaluation value exceeding the threshold value is searched for. If not, the processing is completed assuming that the analyzed periodic signals 1 to N do not include a series of periodic signals. Here, the accumulated evaluation value output from the evaluation value accumulating unit 3-k ′ is an absolute value. The explanation will proceed as if the value was exceeded.

In such a case, the count value output from the count unit 4-g (1 ≦ g ≦ G) and the range of the count value subdivision section (τ ^(d) _(g−1) , τ ^(d) _g ) are used. The number of series of periodic signals having a period of τ ⁽ⁱ⁾ _(k′−1) or more and less than τ ⁽ⁱ⁾ _{k ′} is estimated. First, the threshold value th ^(d) in each count value subdivision section is calculated by the equation (23) using the center value ^(c) τ ⁽ⁱ⁾ _{k ′} of the evaluation value subdivision section k ′.

th ^(d) = β · (t _N −t ₁ ) / ^(c) τ ⁽ⁱ⁾ _{k ′} (23)

Note that the threshold th ^(d) to be compared with the count value is constant regardless of g. Β is a real number of 1 or less set by the setting unit 6. Then, the center value ^(c) τ ^(d) _g of the count value subdivision section g shown in the equation (24) is calculated, and ^(c) τ ^(d) _g is the center value ^(c) τ ^{( i) A} count value subdivision interval g smaller than _{k ′} is extracted. Further, the count value of the extracted count value subdivision section is compared with th ^(d) to search for a count value exceeding the threshold value. As a result, if there is no count value exceeding the threshold value, it is determined that the number of sequences of periodic signals whose arrival time interval is not less than τ ⁽ⁱ⁾ _{(k−1) and} less than τ ⁽ⁱ⁾ _k is 1. When the count values of the count value subdivision interval g ′ and the count value subdivision interval g ″ exceed the threshold value, the arrival time interval is equal to or greater than τ ^{(i )} _{(k−1) and} τ ^{(i ) The} number of series of periodic signals less than _k is determined to be 2.

^(c) τ ^(d) _g = (τ ^(d) _(g−1) + τ ^(d) _g ) / 2 (1 ≦ g ≦ G) (24)
(1-γ) · ^(c) τ ⁽ⁱ⁾ _{k ′} ≦ ^(c) τ ^(d) _{g ′} + ^(c) τ ^(d) _{g ″} ≦ (1 + γ) · ^(c) τ ⁽ⁱ⁾ _{k ′} (25)
Here, γ is a real number less than 1 set by the setting unit 6.

Further, there are six count values exceeding the threshold th ^(d) , and the two center values of the count value subdivision intervals are combined and added, and they are expressed by (1-γ) · ^{(( c) When} three combinations of τ ⁽ⁱ⁾ _{k ′} or more and (1 + γ) · ^(c) τ ⁽ⁱ⁾ _{k ′} or less are made, the interval of arrival times is τ ⁽ⁱ⁾ _(k−1) or more and τ ^{( i) The} number of series of periodic signals less than _k is determined to be 3.

Note that there are cases where two periodic signal sequences having the same arrival time interval (center value) are mixed and the arrival times are both synchronously jittered. In such a case, the count exceeding the threshold th ^(d) The number of values is 1. Determining Therefore, when the number of the count value exceeding the threshold value is 1, the interval of arrival time tau ⁽ⁱ⁾ the number of sequences of _(k-1) or tau ⁽ⁱ⁾ _k less of the periodic signal is 1 and Is done.

In addition, there are cases in which three periodic signal sequences having the same three arrival time intervals (center values) are mixed and the arrival times are jittered together. In such a case, the threshold value th ^(d) is set. The number of count values exceeding 2 is 2. However, in this case, for the two count values exceeding the threshold value, even if the center values of the respective count value subdivision intervals are added, the added value generally does not become a value close to ^(c) τ ⁽ⁱ⁾ _{k ′.} . Therefore, when the number of count values exceeding the threshold th ^(d) is 2 and the condition of the expression (25) is not satisfied, a period in which the interval of arrival times is not less than τ ⁽ⁱ⁾ _{(k−1) and} less than τ ⁽ⁱ⁾ _k The number of signal sequences is determined to be three.

In the above description, the method using the arrival time of radio waves has been described. However, the above method is the center in which all the radio waves disappear (disappearance time), and the arrival time and disappearance time of radio waves are added and divided by 2. Even when the time is used, the same processing as described above can be performed, and the same effect can be obtained. That is, when the disappearance time of the radiated signals n was t _'n, calculates the difference between arrival time in equation (26) or equation (27).

Δt _{n, m} = t ′ _m −t ′ _n (26)
Δt _{n, m} = (t _m + t ′ _m ) / 2− (t _n + t ′ _n ) / 2 (27)

  Furthermore, the evaluation value accumulating unit can obtain the same effect as the method described above even if the evaluation value is calculated by the equation (28) or the equation (29) and the accumulated value of the evaluation value is used.

p _{n, m} = exp (2πjt ′ _n / Δt _{n, m} ) (28)
p _{n, m} = exp {πj (t _n + t ′ _n ) / Δt _{n, m} } (29)

The same effect can be obtained by calculating the evaluation values _{pn, m} based on the equations (30) to (32).

p _{n, m} = exp (2πjt _m / Δt _{n, m} ) (30)
p _{n, m} = exp (2πjt ′ _m / Δt _{n, m} ) (31)
p _{n, m} = exp {πj (t _m + t ′ _m ) / Δt _{n, m} } (32)

  Furthermore, the present invention is applicable not only to radio waves but also to sound waves and light waves.

It is a block diagram of the signal analyzer in one embodiment of this invention. It is a time chart of 1 series of periodic radiation signals for explaining operation of a signal analysis device. It is a time chart of the periodic radiation signal of the 2 series of the same period for demonstrating operation | movement of a signal analyzer. It is a data flow figure in the signal analyzer in one embodiment of this invention.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1 Arrival time difference calculation part, 2 Subdivision section selection part, 3-1 to 3-K evaluation value accumulation part, 4-1 to 4-G count part, 5 estimation part, 6 setting part.

Claims (9)

Based on the observation data including the predetermined timing time of the radiated signal indicating the reception timing of each radiated signal of the received signal in which a plurality of radiated signals are mixed, the arrival time of the radiated signal is the number of periodic signal series and arrival In a signal analyzer for estimating a time interval,
The analysis range of the arrival time interval of the radiated signal, the width and number of the analysis range of the arrival time interval divided into a plurality of evaluation value subdivision intervals, and the analysis range of the arrival time interval divided into a plurality of count value subdivision intervals A setting unit for setting parameters used for estimating the width and the number of sequences, and the number of series,
An arrival time difference calculation unit that calculates a difference between arrival times of two radiation signals based on a predetermined timing time of the radiation signals, and outputs an end flag when the difference between arrival times is calculated for a combination of radiation signals;
A subdivision section selection unit that selects the corresponding evaluation value subdivision section and count value subdivision section based on the difference of the arrival times;
An evaluation value is calculated based on the difference between the arrival times and the predetermined timing time of the radiation signal, and the evaluation value is accumulated for each evaluation value subdivision section, and an evaluation value accumulating unit that calculates a cumulative evaluation value;
A count unit for calculating the count value by counting the number of arrival time differences selected in the selected subdivision section for each count value subdivision section;
Based on the cumulative evaluation value and the count value, an estimation unit that estimates the number of radiated signal sequences and the arrival time interval whose arrival time is periodic;
A signal analysis apparatus comprising:   The signal analyzer according to claim 1, wherein the setting unit sets the count value subdivision sections to overlap each other.   The signal analyzer according to claim 1, wherein the setting unit sets the evaluation value subdivision sections to overlap each other.   The predetermined timing time of the radiation signal is an arrival time of the radiation signal, and the arrival time difference calculation unit calculates a difference between arrival times of two radiation signals and outputs the difference as an arrival time difference. Item 4. The signal analysis device according to any one of Items 1 to 3.   The estimation unit estimates the number of temporary sequences of the periodic signal and the arrival time interval based on the time when the radiation signal is received and the cumulative evaluation value, and then counts the arrival time interval of the periodic signal and the time when the radiation signal is received. The signal analysis apparatus according to claim 4, wherein the final number of sequences of the periodic signal is estimated using the value.   The predetermined timing time of the radiation signal is composed of the disappearance time of the radiation signal, and the predetermined timing time difference calculation unit calculates a difference between the disappearance times of the two radiation signals and outputs the difference as an arrival time difference. The signal analysis apparatus according to claim 1.   The predetermined timing time of the radiation signal is composed of a central time obtained by dividing the sum of the arrival time and the extinction time of the radiation signal by 2, and the predetermined timing time difference calculation unit calculates a difference between the central times of the two radiation signals. The signal analyzer according to claim 1, wherein the signal analyzer outputs the difference as arrival time differences.   The evaluation value accumulating unit calculates an evaluation value based on a difference between arrival times and the disappearance time, and accumulates the evaluation value for each evaluation value subdivision section to calculate a cumulative evaluation value. The signal analyzer according to claim 6.   The evaluation value accumulating unit calculates an evaluation value based on the difference between arrival times and the central time, and accumulates the evaluation value for each evaluation value subdivision section to calculate a cumulative evaluation value. The signal analysis device according to claim 7.

Images