JPH08242259A - Testing device of digital equipment and method therefor - Google Patents

Abstract

PURPOSE: To rightly evaluate a digital equipment by estimating the bit error rate characteristic curve of the digital equipment from a calculated bit error rate, based on the ideal bit error rate characteristic curve of the digital equipment. CONSTITUTION: The demodulation digital data from a data demodulator is stored in a miry 23. For required various kinds of testing signals of C/N, the output from a demodulator is fetched into the memory 23 in the same way. Subsequently, the data of the memory 23 is successively read by a digital signal processor, the data is compared with an expected value by a comparison means 41 and a bit error rate BER is calculated for the comparison result by a BER arithmetic means 42. The BER calculated for each C/N is averaged and the BER is used as the subsequent BER. Namely, a BER characteristic curve is estimated by a BE:R characteristic curve estimation means 43 based on the ideal BER curve of a digital demodulator by using this BER and the corresponded C/N. As a result, the BER characteristic can be obtained in short measuring time.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a digital device for processing a carrier wave modulated by a digital signal such as a digital demodulator in mobile communication, or a digital modulator for generating and outputting a carrier wave by a digital signal. The present invention relates to a test apparatus and a test method for testing the bit error rate characteristic of digital equipment.

[0002]

2. Description of the Related Art There is a demodulator for digital mobile communication, which is composed of a semiconductor integrated circuit, and the bit error rate (hereinafter referred to as BER) characteristic in evaluating the above digital equipment including such a demodulator It is an important parameter. The BER characteristic also differs depending on the modulation method and demodulation method. This BER characteristic can be theoretically obtained,
The calculation method is known, and the ideal BER characteristic can be calculated.

For example, in a demodulator that receives π / 4DQPSK by differential detection, an input modulated carrier signal r (t) is input as shown in FIG. 3A. This r (t) is the transmission modulated carrier signal s (t) plus noise n (t). _{s (t) = AI k cos} (2πf c t) -AQ k sin (2
πf _c t) and can be expressed. Here, AI _k and AQ _k are modulated carrier signals s
The in-phase component and the quadrature component of the amplitude A of (t). This s
The reception input n (t) obtained by adding n (t) to (t) is expressed by the following equation.

R (t) = {AI _k + x (t)} cos (2
_{πf c t) {AQ k +} y (t)} sin (2πf c t) where x (t), y (t ) are in phase and quadrature components of the noise n (t) respectively, each of the sigma ² Gaussian distribution of variance. The input modulated carrier signal r (t) has its unnecessary component removed by a bandpass filter 11 and is supplied to a 1-symbol delay means 12 and multipliers 13 and 14, respectively. The output of the delay means 12 is supplied as it is to the multiplier 13 as a reference signal, and is also shifted by π / 2 in the phase shifter 15 and supplied to the multiplier 14 as a reference signal. The outputs of the multipliers 13 and 14 are low-pass filters, and are cos (Δθ) and sin (Δθ) as baseband signals.
Are extracted, these are identified as binary signals by the identifying means 18, and are further output as demodulated digital data by the parallel / serial converter 19. The probability density function pi (z) of the in-phase component of the differential detection can be expressed by the following equation.

Pi (z) = {1 / ((2π) ^1/2 · 2
^1/2 σ)} exp [-[z-Acos (Δθ) ² / (2-2
σ ² )] where Δθ is the phase difference between (I _k , Q _k ) and (I _k-1 , Q _k-1 ). Similarly, the probability density function Pq (z) of the orthogonal component of the differential detection can be expressed by the following equation. Pq (z) = {1 / (2π) ^1/2・ 2 ^1/2 σ)} exp
[-{Z + A sin (Δθ)} ² / (2-2σ ² )] In π / 4DQPSK, either Δθ = ± π / 4 or ± 3π / 4 is taken, and the error rate occurs equally in any Δθ. Here, Δθ = −π / 4 is calculated. At this time, the signal point is in the first quadrant. Therefore, the error rates Pi and Pq of the in-phase component and the quadrature component of the differential detection output can be expressed by the following equations, respectively.

[0006]

[Equation 1]

[0007]

[Equation 2] Where ∫ is −∞ to 0, γ ² is C / N, and er
f is erf (x) = (2 / √π) ∫exp (-z ² ) dz∫ from x to ∞ From this, the error rate Pe of the entire demodulator is as follows.

Pe = Pi + Pq−Pi × Pq = erf (γ / 2) − (1/4) {erf (γ / 2)} ² ≈erf (γ / 2) (1) Thus, the BER is calculated. However, in reality, the BER is deteriorated due to the fluctuation of the discrimination level, the angle fluctuation, the clock phase error, and the like. Therefore, it is desired to measure the BER characteristics of each digital device and evaluate the device.

In order to obtain the BER characteristic by measurement, a carrier signal (hereinafter referred to as a modulated carrier signal) modulated by a digital signal of various carrier level / noise level (hereinafter referred to as C / N) is generated, and this is generated. It is supplied as a test signal to a device under test, for example, a digital demodulator, the demodulated digital signal is compared with an expected value to calculate an erroneous ratio (BER), and the calculated BER is calculated for each C / N to obtain the BER. It has the characteristics.

[0010]

To accurately measure the BER, the reciprocal of the BER with respect to the C / N of the input signal is set to 10.
It is necessary to take 00 times more data. On the other hand, PHS
In the (Personal Handy System) receiver, the BER is required to be 10 ^-8 or more. Therefore, the PHS bit rate is 380 kb / s to measure the BER characteristics of this digital demodulator.
Need more days. For this reason, it is practically difficult to evaluate the BER characteristics of each of the mass-produced digital demodulators, and it has been omitted in the past.

[0011]

According to the invention of claim 1, a test signal of a carrier signal having various C / Ns and modulated into a digital signal is generated from a test signal generator to a digital device under test. Supplied. The BER of the output from the digital device under test is calculated, and the BER characteristic curve of the digital device under test is estimated from the measured BER for various C / Ns based on the ideal BER characteristic curve of the digital device under test.

This estimation is based on the ideal BER characteristic curve
Operation BER and its C for a curve with unknown residue
Substitute / N and determine the unknown residue so that the measurement error is minimized. As a digital device such as a modulator that modulates a carrier signal with a digital signal, a test signal generator generates a digital test signal and controls the parameters of the digital signal to generate an ideal digital device. The C / N of the output signal in the case can be set. Also, when the modulated carrier signal is input and the modulated carrier signal is output, the digital device under test demodulates the output signal with a demodulator having a known BER characteristic to calculate the BER.

[0013]

Embodiment FIG. 1A shows an embodiment of the present invention. The test signal from the test signal generator 21 is supplied to the digital device under test 22, and the output of the digital device under test 22 is stored in the memory 2.
Then, the digital signal processor (DSP) calculates the BER and estimates the BER characteristic curve, and displays the result on the display 25.

An example of a test on a digital demodulator configured as a semiconductor integrated circuit in digital mobile communication as the digital device under test 22 will be described below. In this case, the test signal generator 21 has various predetermined C / N.
The modulated carrier signal can be generated.
For example, as shown in FIG. 1B, a 9-stage PN pattern generator 26 generates a pseudo random pattern at a speed of 384 kb / s, and this digital signal is supplied to a π / 4DQPSK converter 27 and converted into a π / 4DQPSK signal. To be done. The in-phase signal is added with the in-phase component x (t) of the Gaussian noise n (t) by the adder 28, and n is added to the quadrature component of the modulation signal.
The quadrature component y (t) of (t) is added by the adder 29, the outputs of these adders are supplied to the multipliers 31 and 32, and the carrier signal of 1.2 MHz from the carrier generator 33, which is π / They are each multiplied by the two phase shifted ones. These multiplication outputs are added by the adder 34 to obtain a test signal of the modulated carrier signal. This carrier signal is usually an intermediate frequency signal that is an input to the digital demodulator.

The C / N of the modulated carrier signal obtained depends on the magnitude of the Gaussian noise n (t) to be added. Each time the test signal is generated, noise of a magnitude corresponding to the target C / N may be added, but a test signal to which noise corresponding to various required C / Ns is added is created in advance, These are stored as digital data in the waveform memories 36 _{1 to} 36 _n in the test signal generator 21, respectively, and the data read from these waveform memories are converted into analog signals and output as test signals. You can

The test signal from the test signal generator 21 is generated by the number of repetitions corresponding to the C / N and supplied to the digital demodulator 22. The demodulated digital data from the digital demodulator 22 is stored in the memory 23. Similarly, the outputs from the digital demodulators are taken into the memory 23 for various test signals of C / N required. After that, the digital signal processor 24 sequentially reads the data in the memory 23,
As shown in FIG. 2A, the expected value is compared with the comparison means 41, and the BER calculation means 42 calculates the BER for the comparison result. The BER calculated for each C / N is averaged and used as the subsequent BER. That is, the BER characteristic curve estimating means 43 estimates the BER characteristic curve based on the ideal BER characteristic curve of the digital demodulator 22 using the BER and the corresponding C / N.

For example, in the above-described π / 4DQPSK differential detection demodulation, the ideal BER characteristic curve is shown by the equation (1), which is shown by the curve 45 in FIG. 2B. As described above, although there are various causes of BER deterioration of the demodulator, the deterioration factor can be converted into C / N, and when it is fixed, the BER characteristic curve of the demodulator moves in parallel with the ideal characteristic curve 45. It will be what was made. Considering the occurrence probabilities of various deterioration factors, the BER (Pn) of the demodulator is expressed by the following equation.

Pn = a · erf (b · γ / 2) (2) Therefore, the BER for the C / N obtained previously is substituted into the equation (2) to minimize the error from the measured value. a and b
Is obtained by regression processing, for example. At this time, the B
The ER is measured at a portion where the bend in the ideal BER characteristic curve 45 is relatively large, that is, C where the BER is about 10 ^−4.
The test signal having C / N smaller than / N is performed.

The BER of the digital demodulator consisting of the IC of Company A was measured by the above-mentioned method, and the result was in the state shown by the black dots in FIG. 2B. As a result of obtaining a and b in the equation (2) from this data, 0.5 and 0.8
It became 79. Substituting these values into the equation (2), the BER is calculated for each C / N, and the estimated BER characteristic curve 46
I got Similarly, the result of the measurement of the product of the company B becomes a mark X in FIG. 2B, and a and b are 0.5 and 0.73, respectively.
3 and the estimated BER characteristic curve is the curve 47 in FIG. 2B.

From these curves 46, C / N of 18 dB B
ER is 7.76 × 10 ^-7 , while C /
The BER was measured using 10 ^10- bit data with N of 18 dB, and the result was 7.96 × 10 ⁻⁷ . Further, when the BER with the C / N of 20 dB is obtained from the curve 47, it is 2.1
7 × 10 ⁻⁷ , while the result measured by the conventional method is 1.
It became 75 × 10 ^-7 . From these results, 2 in this example
It is understood that the BER can be predicted with an accuracy of ± 1 power or less. When each error in this result is converted into C / N, they are only 0.17 dB and 0.25 dB, respectively.

As described above, in the above embodiment, the BER is
The measurement may be performed in a state of worse than about 10 ^−4, so that the measurement time is significantly shortened as compared with the conventional method, and the measurement may be performed for about 1 second, for example. The obtained estimated BER characteristic curve is displayed on the display 25 as shown in FIG. 1A, for example. Also, by specifying the position on the display curve, B
ER and CN can be displayed numerically.

In the above description, the present invention is applied to the differential detection demodulator of the DQPSK signal, but the present invention can be applied to demodulators of other detection methods and other digital modulation methods. Further, the present invention can be applied to a digital device such as a filter or an equalizer that processes a digitally modulated carrier signal and outputs the digitally modulated carrier signal. In this case, the output of the digital device under test 22 is demodulated by the digital demodulator 48 having a known BER characteristic and supplied to the memory 23, as shown in FIG. .

The present invention also relates to the BER of the digital modulator.
It can also be applied to the measurement of characteristics. That is, FIG.
As shown in C, the PN pattern generator 26 is provided in the test signal generator 21, and the cycle or amplitude of the generated PN pattern is changed to change C / N equivalently. C / N setting means 49 is to change this C / N.
Done in. The PN pattern from the test signal generator 21 is supplied to the digital modulator under test 51 as a digital test signal. At this time, the C / N of the output modulated carrier signal and the setting of the C / N setting means 49 in the test signal generator 21 when the digital modulation supply 51 is ideal are obtained in advance. In this way, a test signal having a desired C / N is supplied to the digital modulator 51, the modulated output is demodulated by the digital demodulator 48 having a known BER characteristic, and the memory 23
Supply to

In any of the above cases, the test data may be directly supplied to the digital signal processor 24 without being taken in the memory 23 all at once, and the BER may be measured once.

[0025]

As described above, according to the present invention, B
Since only the part where the ER is large is measured and the BER characteristic curve is estimated and the BER characteristic is obtained, the BER characteristic can be obtained in a significantly shorter measurement time than before.
Moreover, because it can be estimated with a fairly high accuracy,
The BER characteristic can be used as the evaluation of the digital device, and the digital device can be correctly evaluated.

[Brief description of drawings]

FIG. 1A is a block diagram showing an embodiment of the present invention, and B is a block diagram showing an example of a test signal generator thereof.

2A is a block diagram functionally showing processing contents of a digital signal processor 24 in FIG. 1, and FIG. 2B is an ideal BER characteristic curve, BER measured in the embodiment of the present invention, and BER estimated in the present invention. It is a figure which shows a characteristic curve.

3A is a block diagram showing a delay detector for a DQPSK signal, FIG. 3B is a block diagram showing a main part of another embodiment of the present invention, and C is a block showing a main part of still another embodiment of the present invention. It is a figure.

Claims (7)

[Claims] 1. A device for testing a digital device that processes a carrier wave modulated by a digital signal (hereinafter referred to as a modulated carrier wave), the carrier signal having various carrier wave levels / noise levels and modulated by the digital signal. The test signal generator for generating the test signal of (1) and supplying it to the digital device, the error rate calculation means for calculating the bit error rate of the output of the digital device, and the test signals of the various carrier level / noise level described above. A test of a digital device, comprising: an estimation means for estimating a bit error rate characteristic curve of the digital device from the calculated bit error rate based on an ideal bit error rate characteristic curve of the digital device. apparatus. 2. The digital device is a demodulator, and the error rate calculation means is means for comparing the output of the demodulator with an expected value and calculating the error rate. Digital equipment testing equipment. 3. The digital device is a device for inputting a modulated carrier wave and outputting a modulated carrier wave, and including demodulation means for demodulating an output of the digital equipment as a digital signal and supplying the demodulated signal to the error rate calculating means. Claim 1 characterized by
Test equipment for the described digital equipment. 4. A device for testing a digital device that modulates a carrier wave with a digital signal and outputs the modulated signal, the parameter of the digital signal being generated as a test signal and supplied to the digital device in an ideal state. A test signal generator capable of outputting modulated carrier waves of various carrier level / noise level in the above, a digital demodulator with a known bit error rate for demodulating the output modulated carrier wave of the above digital equipment, and the above digital Based on the ideal bit error rate characteristic curve of the digital modulator, from the error rate calculation means for calculating the bit error rate of the output of the demodulator and the calculated bit error rate corresponding to the various carrier level / noise level. And a means for estimating the bit error rate characteristic curve of the digital device, and a digital device test comprising: Location. 5. The estimation means minimizes the arithmetic error rate by substituting the arithmetic error rate and its carrier level / noise level for a curve in which an unknown constant is added to the ideal bit error rate characteristic curve. 5. The apparatus for testing digital equipment according to claim 1, which is a means for determining the unknown constant. 6. The test signal generator according to claim 1, further comprising a waveform storage means for storing the test signals of various carrier wave levels / noise levels in advance and reading and outputting the test signals. The test equipment for the digital device according to Crab. 7. A method for testing a digital device that processes a carrier wave modulated by a digital signal or a digital device that modulates a carrier wave signal with a digital signal and outputs the modulated carrier wave signal. A test signal of various carrier level / noise level including a relatively large part and having a smaller carrier level / noise level or a test signal having such a carrier level / noise level is supplied to the digital device, Estimate the bit error rate characteristic curve of the digital device based on the ideal bit error rate characteristic curve by calculating the output bit error rate and using the calculated bit error rate of each carrier level / noise level. A method for testing digital equipment characterized by.