Pseudo noise generator

Abstract

A pseudo noise generator is disclosed, in which a first arbitrary random number generator generates two groups of first random number signals respectively corresponding to divided Amplitude Probability Distribitions, which are obtained by dividing a specified Amplitude Probability Distribition into two parts at a specified level. A second arbitrary random number generator generates two groups of second random number signals respectively defined by a specified Pulse Duration Distribution and a specified Pulse Spacing Destribution at said specified level. Ones of said two groups of first random number signals are selected in accordance with said specified Pulse Duration Distribution and said specified Pulse Spacing-Destribution defined at said specified level. The selected signals are converted, by a D/A converter, to pseudo noise of analog value in accordance with said Amplitude Probability Distribition, said specified Pulse Duration Distribution and said specified Pulse Spacing Destribution at said specified level.

Description

FIELD OF THE INVENTION

[0001] The present invention relates to a pseudo noise generator employed to generate pseudo noise, etc. for evaluate immunity of electric devices against electromagnetic interference waves.

DESCRIPTION OF PRIOR ART

[0002] In case of combining many kinds of noise, or of thermal noise or city noise to one another, amplitude of noise has the Gaussian distribution. In order to simulate generated noises to the Gaussian distribution, there have been proposed a noise generator for generating white-Gaussian noise by the use of a noise diode.

[0003] In narrow-band digital communication systems, there is a correlation between a bit error rate of the narrow-band digital communication system and an Amplitude Probability Distribution (refered to A P D) of electro-magnetic interference waves. Moreover, a report was published that the bit error rate of the communication system can be evaluated from the A P D of electro-magnetic interference waves. In this report, noise determined by a specified A P D is generated by the use of an arbitrary distribution random number generator (See: Paper Journal of The Institute of Electrical, Communication Engineers of Japan (A), vol. J 70-A, No. 11, pp1681-1690, Nov. 1987).

[0004] Important parameters for defining characteristic nature of noise are the A P D, a Crossing Rate Distribution (refered to C R D), a Pulse Duration Distribution (refered to P D D), and a Pulse Spacing Distribution (refered to P S D) etc. These parameters will now be described with reference to FIG. 13.

[0005] The A P D is defined as a time rate where the instantaneous value of a signal, such as electro-magnetic interference waves, exceeds a predetermined value, to show a total time length of the instantaneous value exceeding a level EK in a test time period of T0. The C R D is defined as a number of crossings per a unit time where the instantaneous value of the signal crosses the specified level EK to a positive direction (or a negative direction).

[0006] The P D D is defined as a probability distribution of a time Wi (k) where the instantaneous value of the signal exceeds a level EK in a test time period of T0. On the contrary, the P S D is defined as a probability distribution of a time Zi (k) where the instantaneous value of the signal lowers a level EK in a test time period of T0. In other words, the P D D and the P S D are probability distributions of time lengths from a time where the instantaneous value of the signal crosses the threshold level to a just succeeding time where the instantaneous value of the signal crosses the threshold level.

[0007] Moreover, a Probability Density Function is defined as a distribution of the level EK in the test time period of T0.

[0008] In a pseudo noise generator, the dispersion and the average of noise can be specified, but the A P D of noise cannot be specified since distribution of noise is limited to the white-Gaussian noise.

[0009] In an arbitrary distribution random generator, the A P D of noise be specified to generate noise with an arbitrary A P D. However, noise from the arbitrary distribution random generator assumes independent events having no time-correlation. On the contrary, noise from elctronic Ranges and ordinary electronic devices assumes non-independent events dependent to the period of a source voltage and the period of timing clock pulses. Accordingly, the C R D, the P D D and the P S D of noise from the arbitrary distribution random generator are different from the C R D, the P D D and the P S D of noise of non-independent events having time-correlation.

SUMMARY OF THE INVENTION

[0010] An object of the present invention is to provide a pseudo noise generator capable of specifying a Pulse Duration Distribution and a Pulse Spacing Distribution of noise at a specified amplitude level K in addition to an Amplitude Probability Distribution of noise.

[0011] To this end, a pseudo noise generator of the present invention comprises:

[0012] a first arbitrary random number generator for generating two groups of first random number signals respectively corresponding to divided Amplitude Probability Distribitions, which are obtained by dividing a specified Amplitude Probability Distribition into two parts at a specified level;

[0013] a second arbitrary random number generator for generating two groups of second random number signals respectively defined by a specified Pulse Duration Distribution and a specified Pulse Spacing Destribution at the specified level;

[0014] control means for selecting ones of said two groups of first random number signals in accordance with said specified Pulse Duration Distribution and said specified Pulse Spacing Destribution defined at the specified level; and

[0015] a D/A converter for converting the selected signals to pseudo noise of analog value; said pseudo noise being generated in accordance with said Amplitude Probability Distribition, and said specified Pulse Duration Distribution and said specified Pulse Spacing Destribution at the specified level.

BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The present invention will be described in detail below with reference to the accompanying drawings, in which;

[0017] FIG. 1 is a block diagram explanatory of the principle of the present invention;

[0018] FIG. 2 is a block diagram illustrating an embodiment of the present invention;

[0019] FIG. 3 is a block diagram illustrating an example of an arbitrary distribution random number generator 1;

[0020] FIG. 4 is a block diagram illustrating an example of each bit generator employed in arbitrary distribution random number generators 1 and 2;

[0021] FIG. 5 is a graph showing arrangement of data in a memory in the bit generator;

[0022] FIG. 6 is a block diagram illustrating an example of the arbitrary distribution random number generator 1 and time charts explanatory of operations of the same;

[0023] FIG. 7 is a block diagram illustrating an example of the arbitrary distribution random number generator 2 and time charts explanatory of operations of the same;

[0024] FIG. 8 is a block diagram illustrating an example of the controller 3 and time charts explanatory of operations of the same;

[0025] FIG. 9 shows characteristic curves illustrating conventional test results of the Amplitude Probability Distribition (a), the Crossing Rate Distribution (b) and the Probability Density Function (c) of electro-magnetic interference waves from electronic Ranges and pseudo noise in case of specifying the electromagnetic interference waves from electronic Ranges and the Amplitude Probability Distribition;

[0026] FIG. 10 shows characteristic curves illustrating conventional test results of the Pulse Duration Distribition of electromagnetic interference waves from electronic Ranges and pseudo noise in case of specifying the electromagnetic interference waves from electronic Ranges and the Amplitude Probability Distribition;

[0027] FIG. 11 shows characteristic curves illustrating the present invention's test results of the Amplitude Probability Distribition (a) the Crossing Rate Distribution (b) and the Probability Density Function (c) of electromagnetic interference waves from electronic Ranges and pseudo noise in case of specifying the electromagnetic interference waves from electronic Ranges and the Amplitude Probability Distribition;

[0028] FIG. 12 shows characteristic curves illustrating the present invention's test results of the Pulse Duration Distribition (a) and the Pulse Spacing Distribution (b) of electro-magnetic interference waves from electronic Ranges and pseudo noise in case of specifying the electromagnetic interference waves from electronic Ranges and the Amplitude Probability Distribition; and

[0029] FIG. 13 shows time charts explanarory of technical terms employed in this specification.

DETAILED DESCRIPTION

[0030] With reference to FIG. 1, the principle of the pseudo noise generator of the present invention provided in accordance with a specified A P D, a specified P D D and a specified P S D will first be descrined. This pseudo noise generator-of the present invention comprises four arbitrary distribution random number generators APD1, APD2 PDD0 and PSD0 and a selector SL.

[0031] In this case, setting of the P D D and the P S D is one point of the same level k; and an arbitrary distribution random number generator APD1 for generating random numbers of values exceeding the level k and an arbitrary distribution random number generator APD2 for generating random numbers of values under the level k are provided. These arbitrary distribution random number generators APD1 and APD2 are switched to meet with the specified P D D and P S D to generate random number codes-meeting with required A P D, P D D and P S D. Namely, generation of pseudo noise according to the present invention is performed as described below:

[0032] {circle over (1)} Binary codes i1 of N bits included in a pulse duration distribution pdd (i1) is generated from the arbitrary distribution random number generator PDD0 to determine a pulse duration length Ti1. During this pulse duration length Ti1, binary codes x1 of M bits included in an amplitude probability distribution apd1 (x1) is generated from the arbitrary distribution random number generator APD1 to obtain pseudo noise x.

[0033] {circle over (2)} A After the end of the pulse duration Ti1, binary codes i2 of N bits included in a pulse duration distribution psd (i2) is generated from the arbitrary distribution random number generator PSD0 to determine a pulse duration length Ti2. During this pulse duration length Ti2, binary codes x2 of M bits included in an amplitude probability distribution apd2 (x2) is generated from the arbitrary distribution random number generator APD2 to obtain an outout of pseudo noise x.

[0034] {circle over (3)} The above steps {circle over (1)} and {circle over (3)} are alternately performed.

[0035] In accordance with the above operations, binary numbers x of M bits meeting with the required A P D in addition to the required P D D and P S D are generated, and digital-to-analog converted to obtain an output of the pseudo noise generator.

[0036] As mentioned above, in the pseudo noise generator of the present invention, an arbitrary distribution random number generator 100 for generating exclusively the binary codes x1 of M bits included in an amplitude probability distribution apd1 (x1) or the binary codes x2 of M bits included in an amplitude probability distribution apd2 (x2) is employed in place of an arbitrary distribution random number generator for genarating a signal x included in a specified amplitude probability distribution (x).

[0037] A controller for outputting a memory-selecting signal s is provided to selet a first state generating the binary codes x1 or a second state generating the binary codes x2.

[0038] An arbitrary distribution random number generator 200 is provided for generating exclusively the binary codes i1 of M bits included in a pulse duration distribution pdd1 (i1) or the binary codes i2 of M bits included in a pulse spacing distribution psd (i2) to generate the memory selection signal s employed in the controller. Setting of the pdd1 (i1) and the psd (i2) is one point of the same level k.

[0039] A digital-analog converter is provided to convert the binary codes x (x1 or x2) to an analog value. An analog signal converted in the digital-analog converter is applied to an communication system through a cable or is radiated from an antenna by shifting the frequency band of the thereof by the use of an up-converter.

[0040] In the pseudo noise generator according to the present invention, the binary codes i1 included in a pulse duration distribution pdd1 (i1) is generated from the arbitrary distribution random number generator 200. During the pulse duration length Ti1 corresponding to the binary codes i1, the binary codes x1 included in an amplitude probability distribution apd1 (x1) is generated from the arbitrary distribution random number generator 100. Thereafter, the binary codes i2 included in a pulse spacing distribution psd (i2) is generated from an arbitrary distribution random number generator 200. During the pulse duration length Ti2, the binary codes x2 included in an amplitude probability distribution apd2 (x2) is generated from the arbitrary distribution random number generator 100. Switching between the amplitude probability distribution apd1 (x1) and the amplitude probability distribution apd2 (x2) and switching between the pulse duration distribution pdd1 (i1) and the pulse spacing distribution psd (i2) are performed in accordance with the memory selection signal s from the selecter SL.

[0041] Since the amplitude level of the binary codes x1 included in the amplitude probability distribution apd1 (x1) exceeds the value k while the amplitude level of the binary codes x2 included in the amplitude probability distribution apd2 (x2) lowers the value k, the setting of the P D D and the P S D at the amplitude level k is carried out in the pulse duration length Ti1 and the pulse duration time length Ti2 respectively.

[0042] The amplitude probability distribution apd1 (x1) and the amplitude probability distribution apd2 (x2) are calculated from the amplitude probability distribution apd (x). If the amplitude probability distribution apd (k), the pulse duration distribution pdd1 (i1) and the pulse spacing distribution psd (i2) meet with the condition defined by an equation (1), the binary codes x (x1 or x2) generated from the arbitrary distribution random number generator 100 is included in the amplitude probability distribution apd (x). In this case, the notation M is a number of bits of the binary codes x (x1 or x2) generated from-the arbitrary distribution random number generator 100, and the notation N is a number of bits of the binary codes i (i1 or i2); 1$\begin{array}{cc}\sum _{i_2=0}^{2^N-1}\mathrm{psd}\left(i_2\right)T_{i^2}=\frac{1-\mathrm{apd}\left(k\right)}{\mathrm{apd}\left(k\right)}\sum _{i_1=0}^{2^N-1}\mathrm{psd}\left(i_1\right)T_{i^1}& \left(1\right)\end{array}$

[0043] In the present invention, designation of the pulse duration distribution P D D and the pulse spacing distribution P S D is not limited to designation of distribution having a distribution duration, but can be included designation of a special example of distribution having a defined value, as far as meeting with the condition of the equation (1).

[0044] (Embodiments)

[0045] In FIG. 2, an embodument of the pseudo noise generator of the present invention is illustrated. In this embodiment, arbitrary distribution random number generators of eight bits are employed. The embodiment comprises an arbitrary distribution random number generator 1, an arbitrary distribution random number generator 2, a controller 3 and a digital-to-analog (D/A) converter 4. The arbitrary distribution random number generator 1 is controlled by clock (1) and the memory selection signal s from the controller 3. The arbitrary distribution random number generator 2 is controlled by clock (2) and the memory selection signal s which are applied from the controller 3. The D/A converter 4 converts the binary codes x to an analog signal.

[0046] With reference to FIG. 3, the arbitrary distribution random number generator 1 comprises eight bit generators 1-1 to 1-8 and eight latch circuits 1-11 to 1-18 to generate the binary codes x (x1 or x2) of eight bits, in which the eight bit generators 1-1 to 1-8 and eight latch circuits, 1-11 to 1-18 are alternately connected in cascade so as to actuate each of the bit generators 1-1 to 1-8 at the rise-up instant of each clock pulses.

[0047] As shown in FIG. 4, each of the eight bit generators 1-1 to 1-8 comprises a uniform random number generator 11, a memory 12, and a comparator 13. In the memory 12, data y employed for determining each bit (a, b, . . . h) of the binary codes x (x1 or x2) are stored as follows.

[0048] The amplitude probability distribution apd1 (x1) correspods to a limited part of the binary codes x (x1) in the amplitude probability distribution apd (x) to generate the binary code x1, which is included in a range k≦x. The amplitude probability distribution apd1 (x1) is defined in an equation (2). 2$\begin{array}{cc}{\mathrm{apd}}_1\left(x_1\right)=\{\begin{array}{cc}1& \left(0\leqq x_1<k\right)\\ \frac{\mathrm{apd}\left(x_1\right)}{\mathrm{apd}\left(k\right)}& \left(k\leqq x_1\leqq 2^8-1\right)\end{array}& \left(2\right)\end{array}$

[0049] The amplitude probability distribution apd2 (x2) correspods to a limited part of the binary codes x (x2) in the amplitude probability distribution apd (x) to generate the binary code x2, which is included in a range k>x. The amplitude probability distribution apd2 (x2) is defined in an equation (3). 3$\begin{array}{cc}{\mathrm{apd}}_2\left(x_2\right)=\{\begin{array}{cc}\frac{\mathrm{apd}\left(x_2\right)-\mathrm{apd}\left(k\right)}{\mathrm{apd}\left(0\right)-\mathrm{apd}\left(k\right)}& \left(0\leqq x_2<k\right)\\ 0& \left(k\leqq x_2\leqq 2^8-1\right)\end{array}& \left(3\right)\end{array}$

[0050] The amplitude probability distribution apd1 (x1) and the amplitude probability distribution apd2 (x2) calculated in accordance with the steps defined in the equations (2) and (3) are converted to Conditional Probalitities pc1 (j,r), pc2 (j,r) and then stored as data y for determining each bit in the memory 5 as shown in FIG. 5. In this case, the notation j=1,2, . . . 8 and the notation r=0, 1, . . . 2j−1−1. 4$\begin{array}{cc}{\mathrm{pc}}_1\left(j,r\right)=\frac{{\mathrm{apd}}_1\left(\left(2r+1\right)·2^{M-j}\right)-{\mathrm{apd}}_1\left(\left(2r+2\right)·2^{M-j}\right)}{{\mathrm{apd}}_1(\left(2r·2^{M-j}\right)-{\mathrm{apd}}_1\left(\left(2r+2\right)·2^{M-j}\right)}& \left(4\right)\\ {\mathrm{pc}}_2\left(j,r\right)=\frac{{\mathrm{apd}}_2\left(\left(2r+1\right)·2^{M-j}\right)-{\mathrm{apd}}_2\left(\left(2r+2\right)·2^{M-j}\right)}{{\mathrm{apd}}_2(\left(2r·2^{M-j}\right)-{\mathrm{apd}}_2\left(\left(2r+2\right)·2^{M-j}\right)}& \left(5\right)\end{array}$

[0051] With reference to FIGS. 6 and 4, operations of the arbitrary distribution random number-generator 1 will be described.

[0052] In the arbitrary distribution random number generator 1, bit data A(s1, a1) from a first bit generator 1-1 is applied through a latch 1-11 to a second bit generator 1-2 at the rise instant of each pulse of the clock (1). The signal s1 is the memory selection signal s, and the signal a1 is a first bit a of the binary codes x from the first bit generator 1-1.

[0053] At the same time as the rise instant of each pulse of the clock (1), the uniform random number generator 11 in the second bit generator 1-2 having the construction shown in FIG. 4 generates uniform random numbers z. Thereafter, the data y employed in the second bit generator 1-2 are read out from the memory 12 by the use of the bit data A(s1, a1) as address data. The data y and z are compared with each other, so that a second bit b1 of an arbitrary distrinution number x at the output of the comparator 12 assumes a state “1” in case of y<z, while the second bit b1 assumes the state “0” in case of y≧z.

[0054] The second bit generator 1-2 applies bit data B (s1, a1, b1) to a third bit generator 1-3 (not shown) at the just suceeding rise instant of each pulse of the clock (1). New bit data of (s2, a2) are applied from the first bit generator 1-1 to the second bit generator 1-2, which generates second bit data b2 at the just succeeding clock pulse in response to the bit data of (s2, a2).

[0055] Each of other bit gernerators generates corresponding bit data from the bit data applied from, a bit generator of the just preceding stage, so that the bit data applied from the bit generator of the just preceding stage are combined with bit data generated at the bit generator of instant stage, the combined bit data are applied to a bit generator of the just suceeding state.

[0056] In this case, since the first bit generator 1-1 is of a first stage, only the memory selection signal s is applied to the first bit generator 1-1 as shown in FIG. 6 in place of bit data to be applied from a just preceding stage. Since a bit generator 1-8 is of a last stage, the memory selection signal s is not necessarily applied to a next stage. Accordingly, data x (x=a,b, . . . , h) from which the memory selection signal s is removed are applied to the D/A converter as shown in FIG. 6.

[0057] In accordance with the operations mentioned above, the arbitrary distribution random number generator 1 generates the binary code x1 included in the amplitude probability distribution apd1 (x1) in synchronism with the pulse of the clock (1) in case of the state “1” of the memory selection signal as from the controller 3. In case, of the sate “0” of the memory selection, signal s from the controller 3, the binary code x2 included in the amplitude probability distribution apd2 (x2) in synchronism with the pulse of the clock (1).

[0058] With reference to FIG. 7, construction and operations of the arbitrary distribution random number generator 2 will now be desribed. This arbitrary distribution random number generator 2 generates, alternately the binary codes i1 employed for determining the pulse duration time length Ti1 of the binary codes x1 included in the amplitude probability distribution apd1 (x1), or the binary codes i2 employed for determining the pulse duration length Ti2 of the binary codes x2 included in the amplitude probability distribution apd2 (x2).

[0059] This arbitrary distribution random number generator 2 has the construction and operations similar to those of the arbitrary distribution random number generator 1. In this arbitrary distribution random number generator 2, however, the clock (2) is employed in place of the clock (1). In each memory 12 of bit generators 2-1 to 2-8, data for generating the pulse duration distribution pdd1 (i1) and the pulse spacing distribution psd (i2) in place of the amplitude probability distribution apd1 (x1) and the amplitude probability distribution apd2 (x2). From a function n (Ti1) of a number n and the pulse duration length Ti1, the pulse duration distribution pdd1 (i1) can be calculated in accordance with an equation (6). 5$\begin{array}{cc}\mathrm{pdd}\left(i_1\right)=\frac{n\left(T_{i^1}\right)}{\sum n\left(T_{i^1}\right)}& \left(6\right)\end{array}$

[0060] The pulse spacing distribution psd (i2) is calculated in the similar manner to those of the pulse duration distribution pdd1 (i1) in accordance with an equation (7). 6$\begin{array}{cc}\mathrm{psd}\left(i_2\right)=\frac{m\left(T_{i^2}\right)}{\sum m\left(T_{i^2}\right)}& \left(7\right)\end{array}$

[0061] The pulse spacing distribution psd (i2) and the pulse duration distribution pdd1 (i1) are converted in accordance with equations (8) and (9) to trial conditional probability values pc1 (j,r) and pc2 (j,r), which are stored in the memory of the arbitrary distribution random number generator 2. In this case, the binary codes i1 and i2 are defined by bit signals j and r in the equation (8). However, the bit signals j and r are defined as j=1,2, . . . ,8 and r=0,1, . . . 2j−1−1. The bit signal r is determined by trial results until the (j−1)-th trial. Data arrangement in the memory 12 of each bit generation 2-1 to 2-8 is shown in FIG. 5 in the similar manner to the arbitrary distribution random number generator 1. 7$\begin{array}{cc}{\mathrm{pc}}_1\left(j,r\right)=\frac{\mathrm{pdd}\left(\left(2r+1\right)·2^{8-j}\right)-\mathrm{pdd}\left(\left(2r+2\right)·2^{8-j}\right)}{\mathrm{pdd}\left(2r·2^{8-j}\right)-\mathrm{pdd}\left(\left(2r+2\right)·2^{8-j}\right)}& \left(8\right)\\ {\mathrm{pc}}_2\left(j,k\right)=\frac{\mathrm{psd}\left(\left(2r+1\right)·2^{8-j}\right)-\mathrm{psd}\left(\left(2r+2\right)·2^{8-j}\right)}{\mathrm{psd}\left(2r·2^{8-j}\right)-\mathrm{psd}\left(\left(2r+2\right)·2^{8-j}\right)}& \left(9\right)\end{array}$

[0062] With reference to FIG. 8, the controller comprises an I-T converter 20 receiving the binary codes i, a down counter 21 of 32 bits receiving output data T* of the I-T converter 20 under control of the clock (1), a memory selection signal generator 22 receiving the carry output c of the down counter 21, and a clock generator 23 receiving the carry output c of the down counter 21 and the clock (1). The I-T converter 20 is a memory for data of thirty-two bits which is controlled with address codes i (=i1 or i2) of eight address bits to read out the stored data T* (*=i1 or i2).

[0063] The controller 3 sets the memory selection signal s to the state “1” during the time length Ti1 and to the state “0” during the time length. Ti2 to control the pulse duration length Ti1 of an arbitrary distribution random number included in the amplitude probability distribution apd1 (x1) and the pulse duration length Ti2 of an arbitrary distribution random number included in the amplitude probability distribution apd2 (x2) as shown in time charts of FIG. 8.

[0064] If the memory selection signal s assumes the state “1”, the counting value c of the down counter 21 is decreased by “1”. In this period of the state “1”, the arbitrary distribution random number generator 1 generates the amplitude probability distribution apd1 (x1).

[0065] When the counting value c of the down counter 21 reaches the zero state, the memory selection signal s assumes the state “0”. In this period of the state “0”, the arbitrary distribution random number generator 1 generates the amplitude probability distribution apd2 (x2).

[0066] In response to the zero state of the down counter 21, the pulse duration length Ti2 is newly set in the down counter 21 at the rise instant of a just succeeding pulse of the clock (1). At the same time as the zero state of the down counter 21, a pulse p2 of the clock (2) comes at the rise instant of a just succeeding pulse of the clock (1) to actuate the arbitrary distribution random number generator 2 so as to newly generate the binary code i1′, which is converted to the output data T1′ in the I-T converter 20.

[0067] The controller 3 descreases successively the conting value of the down copunter 21 as the similar manner to the case of the state “1” of the memory selection signal s. In this period, the arbitrary distribution random number generator 1 generates the binary code x2 included in the amplitude probability distribution apd2 (x2). When the counting value of the down counter 21 reaches the zero state, the state “0” of the memory selection signal s is reversed to the state “1”, so that the output data T1′ is set to the down counter 21 to be restored to the initial state.

[0068] As mentioned above, the pulse duration length Ti1 and the pulse duration length Ti2 are alternately set to the down counter 21 in the controller 3 to alternately reverse the state “1” and the state “0” of the memory selection signal s in response to the pulse duration length. Ti1 and the pulse duration length Ti2, respectively, so that the binary code xi included, in the amplitude probability distribution apd1 (x1) and the binary code x2 included in the amplitude probability distribution apd2 (x2) are alternately generated.

[0069] Simulation results of pseudo noise generation by a computer associated with the device of the present invention will be described. In this simulation test, the A P D, the P D D and the P S D of the electro-magnetic interference waves from an electronic range are measured and adopted. Since the measured values of the A P D, the P D D and the P S D are are obtained from actual electromagnetic interference waves, these measured values meets necessrily with the condition of the equation (1). FIGS. 9A, 9B, 9C and FIG. 10 are simulation results of pseudo noise distributions by conventional devices, where only the electromagnetic interference waves from an electronic range and the A P D are specified. A characteristic curve of heavy line in FIG. 9A indicates the A P D of the electromagnetic interference waves from an electronic range and is adopted as the A P D by conventional devices. A characteristic curve of heavy line in FIG. 9B indicates the C R D of the electromagnetic interference waves from an electronic range, and a characteristic curve of heavy line in FIG. 9C indicates the P D F of the electro-magnetic interference waves from an electronic range. By using marks ◯, Δ and X in FIGS. 9A, 9B and 9C, the A P D, the C R D and the P D F of the pseudo noise are illustrated. In FIG. 10, the P D D of the actual electro-magnetic interference waves and the P D D of the pseudo noise are illustrated with a heavy line and marks &Circlesolid;, respectively. As understood from FIGS. 9A, 9B, 9C and 10, the actual electromagnetic interference waves and the pseudou noise are substantially agreeable with each other with respect to the A P D and the P D F but different from each other with reference to the C R D and the P D D.

[0070] On the contrary, FIGS. 11A, 11B and 11C and FIG. 12 show test result of pseudou noise according to the present invention in case of specifying the actual electromagnetic interference waves and the A P D, the P D D and the P S D. A heavy line in FIG. 11A indicates the A P D of the actual electromagnetic interference waves and is employed as a specified value of the A P D. A heavy line in FIG. 11B indicates the C P D of the actual electromagnetic interference waves, while a heavy line in FIG. 11C indicates the P D F of the actual electromagnetic interference waves. In FIGS. 11A, 11B and 11C, marks ◯, Δ and X indicates the A P D, the iC R D and the P D F of the actual electro-magnetic interference waves, respectively. In FIG. 12A, a heavy line indicates the P D D of the actual electromagnetic interference waves, while a heavy line indicates the P S D of the actual electromagnetic interference waves. In FIGS. 12A and 12B, the P D D and the P S D of pseudou noise generated in case of the P D D and the P S D specified are indicated by marks ◯ and X, respectively. As understood from FIG. 11A to FIG. 12B, the actual electromagnetic interference waves and the pseudou noise are substantially agreeable with each other with respect to the P D D and the P S D in addition to the A P D and the P D F but still different from each other with reference to the C R D.

[0071] As mentioned above in detals, noise of non-independent events having time-correlation of desired characteristics can be generated in accordance with the present invention of relatively simplified construction and control operations by designating the amplitude probability distribution apd (x), the pulse duration distribution pdd1 (i1) and the pulse spacing distribution psd (i2). Therefore, merits of the present invention are very effective for evaluate immunity of electric devices against electromagnetic interference waves.

Claims

1. A pseudo noise generator comprising: a first arbitrary random number generator for generating two groups of first random number signals respectively corresponding to divided Amplitude Probability Distribitions, which are obtained by dividing a specified Amplitude Probability Distribition into two parts at a specified level; a second arbitrary random number generator for generating two groups of second random number signals respectively defined by a specified Pulse Duration Distribution and a specified Pulse Spacing Destribution at said specified level; control means for selecting ones of said two groups of first random number signals in accordance with said specified Pulse Duration Distribution and said specified Pulse Spacing Destribution defined at said specified level; and a D/A converter for converting the selected signals to pseudo noise of analog value; said pseudo noise being generated in accordance with said Amplitude Probability Distribition, and said specified Pulse Duration Distribution and said specified Pulse Spacing Destribution at said specified level.

2. A pseudo noise generator comprising: first terminal means receiving first clock pulses; a controller-for generating, in addition to a selection signal, second clock pulses counted down by one forth from said first clock pulses; a first arbitrary random number generator for generating, under control with said first clock pulses and said selection signal, two groups of first random number signals respectively corresponding to divided Amplitude Probability Distribitions, which are obtained by dividing a specified Amplitude Probability Distribition into two parts at a specified level; a second arbitrary random number generator for generating, under control of said second clock pulses and said selection signal, two groups of second random number signals respectively defined by a specified Pulse Duration Distribution and a specified Pulse Spacing Destribution at said specified level; and a D/A converter for converting the selected signals to pseudo noise of analog value; said pseudo noise being generated in accordance with said Amplitude Probability Distribition, and said specified Pulse Duration Distribution and said specified Pulse Spacing Destribution at said specified level; said controller controlling said first arbitrary random number generator by said selection signal to selecting ones of said two groups of first random number signals in accordance with said specified Pulse Duration Distribution and said specified Pulse Spacing Destribution defined at said specified level.

3. A pseudo noise generator according to claim 2, in which said first arbitrary random number generator comprises eight stages connected successively in cascade, each of stages comprising a cascade connection of a bit generator and a latch circuits, said eight stages being controlled by said first clock pulses, said selection signal being applied to a first stage of said eight stages, said first random number signals being obtained from the last stage of said eight stages.

4. A pseudo noise generator according to claim 2, in which: said second arbitrary random number generator comprises eight stages connected successively in cascade, each of stages comprising a cascade connection of a bit generator and a latch circuits, said eight stages being controlled, by said second clock pulses, said selection signal being applied to a first stage of said eight stages, said second random number signals being obtained from the last stage of said eight stages.

5. A pseudo noise generator according to claim 3 or 4, in which said bit generator comprises a uniform random generator for generating uniform random numbers z controlled under said first clock pulses, a memory for storing data neccessary to determine each bit y of said random number signals to be generated, and a comparator for generating the state “1” or the state “0” in accordance with comparison results I (where y<z) and II (where y≧z), respectively.

6. A pseudo noise generator according to claim 2, in which: said controller comprises an I-T converter for converting said second arbitrary random numbers i to time length data T*, a down counter counting said time length data T* under control of said first clock pulses, a clock generator for generate said second clock pulses from the counting state of said down counter under control of said first clock pulses, and a signal generator for generating said selection signal from the counting state of said down counter under control of said first clock pulses.