RU2287900C1 - Method for measuring multiple-signal selectivity of broadband signal receiver - Google Patents

Abstract

FIELD: measurement technology.

SUBSTANCE: novelty is that proposed method for measuring multiple-signal selectivity of broadband signal receiver involves use of two types of frequency allocation in N (N > 2) interferences; first type includes choice of interference frequencies in vicinity of useful-signal intermediate frequency so as to ensure that third-order intermodulation components from N interferences occurred in each channel of protective unit; second type involves allocation of interference frequencies so that intermodulation components occurred only in 50% of protective-unit channels (number of interferences in both types) being equal); interference power level is increased in first type of interference frequency allocation before reducing output signal-to-noise ratio in proportion to cube of interference power and in second type, before initiating reduction of signal-to-noise ratio in proportion to interference power.

EFFECT: enhanced accuracy in measuring multiple-signal selectivity of broadband signal receiver in narrow-band interference environment.

2 cl, 3 dwg

2 cl, 3 dwg

Description

A method for measuring the multi-signal selectivity of a broadband signal receiver (PWM SHPS) relates to measuring technique and can be used in radio engineering to assess the real selectivity of a PWM ShPS.

For narrow-band receivers, work to refine the selectivity parameters was carried out for a long time and ended with standardization of both the selectivity characteristics and measurement methods (GOST 12252-86, GOST 22580-84).

There is neither a complete, universally recognized set of parameters for PWM NPS, nor established measurement methods. As a result, the prediction of receiver behavior, especially in an extremely severe electromagnetic environment, is unreliable.

Closest to the technical nature of the proposed is a method of measuring the selectivity of single-band radios of the land mobile service according to GOST 22579-86, adopted as a prototype.

According to the prototype method, the multi-signal selectivity of the receiver is determined in the following sequence:

a useful signal from the signal generator is supplied to the input of the receiver at the tuning frequency and the level of the useful signal U _c is set at which the signal-to-noise ratio is 12 dB;

increase the input signal level by 3 dB;

include noise generators at frequencies:

when measured above the frequency of the main receiving channel

f _r1 = f _nom 11,850 Hz, f _r2 = f _nom 21,850 Hz,

when measured below the frequency of the main reception channel

_r1 f = f _nom -8150 Hz, f _r2 = f _nom -18,730 Hz,

(where f _nom is the nominal receiver tuning frequency);

increase levels of interfering signals, keeping them the same, to the value at which intermodulation occurs;

the frequency of the second interference generator is adjusted so that intermodulation is manifested to the greatest extent;

increase the levels of interfering signals to a value of U _p , at which the signal-to-noise ratio at the receiver output becomes equal to 12 dB;

multi-signal selectivity of the receiver is determined by the formula

The disadvantage of the prototype method is that the measurement of multi-signal selectivity is carried out for receivers of narrowband signals, and for broadband receivers with a protection unit (BZ) from narrowband interference, this method is unsuitable.

To eliminate this drawback in the method of measuring the multi-signal selectivity of a broadband signal receiver with a narrow-band interference protection unit and having a linear output, including supplying a mixture of a useful signal and narrow-band noise to the input of the receiver and measuring the signal-to-noise ratio at its output, according to the invention, measurements are carried out for two types of frequency distribution N (N> 2) interference, in the first type of interference frequency is chosen near the average frequency of the useful signal so that the intermodulation components are one third its order of N interference fell into each channel of the protection unit, in the second form the interference frequency is distributed so that the intermodulation components fall into only half the channels of the protection unit (the amount of interference is the same in both types), and the interference power level is increased in the first type of distribution of interference frequencies before the start of the decrease in the signal-to-noise ratio at the output, it is proportional to the cube of the interference power, and in the second type of distribution, before the start of the decrease in the signal-to-noise ratio is proportional to the interference power.

The protection block of the PWM ШПС carries out the function of a “whitening” filter, excluding from the spectrum of the receiver fed to the compression device of the ШПС sections affected by narrow-band interference [1]. As a rule, the number of frequency channels in the KB is several tens. The signal-to-noise ratio at the output of the compression device of the receiver with KB is determined by the formula [2]

where h ² is the signal-to-noise ratio at the output of the ShPS compression device,

- коэффициент сжатия сигнала, соответствующий полосе канала БЗ,

- the compression ratio of the signal corresponding to the bandwidth of the channel BZ,

Δf _to - equivalent bandwidth of the signal compression device (correlator),

ΔF is the bandwidth of the channel BZ,

i is the number of the frequency channel BZ,

M - the number of channels BZ,

P _si - part of the useful signal power falling into the passband of the i-th channel of the knowledge base,

P _wi - interference power in the passband of the i-th channel of the knowledge base.

The mechanisms of influence of narrow-band interference on the characteristics of the reception of a useful signal are the same for narrow-band PFP and PWM BPS. Deterioration of sensitivity due to the presence of interference is due to the following factors [3]:

1. The passage of interference due to linear selectivity (insufficient coefficient of squareness and the final value of the attenuation of the filters during detuning).

2. Non-linear lesion due to blocking effect.

3. Non-linear parametric effect, leading to the inverse transformation of the noise of the local oscillator when the input is affected by interference.

4. Nonlinear damage due to the appearance of intermodulation distortions of the form 2f ₁ -f ₂ , f ₁ + f ₂ -f ₃ .

As a rule, for correctly designed receivers, the mechanisms of action according to claims 1-3 begin to affect at higher interference levels than the mechanism according to claim 4. However, in a different electromagnetic environment, it is possible both a deterioration in the reception characteristics due to the influence of a small amount (in the limit of one) of very powerful interference with exposure mechanisms according to claims 1-3, and due to the influence of a sufficiently large number of relatively low-power interference with the mechanism impact according to claim 4. Therefore, in order to fully evaluate the selectivity characteristics of PfP, it is necessary to measure the characteristics corresponding to the impact mechanisms according to claims 1-3, separately from the impact mechanism according to claim 4.

The total interference power in the i-th channel of the knowledge base is determined by the following formula:

where D _ij is the attenuation (in power) by the filter of the i-th channel of the KB of the interference with a frequency corresponding to the average frequency of the j-th channel of the KB,

N _ij is the spectral density of the noise of the local oscillator,

K ₃ is the 3rd order clear factor [4].

In expression (2), the first term characterizes the internal noise of the PFP, the second — narrow-band interference directly in the BZ channel band, the third — the linear passage of interference due to non-ideal filters, the fourth — the conversion of local oscillator noise, and the fifth — intermodulation interference of the form 2f ₁ -f ₂ , sixth - intermodulation interference of the form f ₁ + f ₂ -f ₃ .

If we consider the power of all narrow-band interference affecting the PFP to be the same and equal to R _p , expression (2) can be rewritten in the following form:

- характеризует влияние конечности избирательности фильтров и обратного преобразования шумов гетеродина,

- characterizes the influence of the finiteness of the selectivity of the filters and the inverse noise conversion of the local oscillator,

Note that In _i is equal to the number of 3rd order intermodulation interference falling into the band of the i-th channel of the knowledge base.

Using (1) and (3), we write the expression for the deterioration of the signal-to-noise ratio Δ under the influence of N narrow-band interference on the PFP. In this case, we will take into account that the power of the useful signal in each of the KB channels is

where P _so is the power of the useful signal in each of the BR channels in the absence of narrow-band interference at the input, K _N (P _p ) ≤1 - characterizes the decrease (blocking) of the transmission coefficient of the path to the BR under the influence of N interference with a power of R _p each.

As can be seen from formulas (1) and (4), even with large values of K _{3, the} change in the signal-to-noise ratio under the influence of two interference (as in the prototype method) is very small.

For example, at M = 30, the signal-to-noise ratio at the output of an ideal (without non-linearity) PFM and PFM with high non-linearity, due to the presence of the KB, differs by 8% (0.32 dB).

It is very difficult to measure such a small change in real production conditions. If the receiving device is designed to receive discrete information and includes a decision circuit and a decoder, only the probability of error in receiving the symbol can be used for measurement. In this case, taking into account the random nature of the error probability estimation, such differences in the reception characteristics require enormous amounts of statistics, especially if the given reliability of the transmission is high. All this practically excludes the possibility of applying the methodology in accordance with GOST 22579-86 for measuring the multi-signal selectivity of PWM PWS with BZ.

An analysis of formula (4) suggests a possible way to separately measure the PFP selectivity parameters associated with the mechanisms of action according to items 1 and 3, items 2 and 4.

We feed the PfP input with a useful signal mixed with N narrow-band interference. We choose the interference frequencies and their number so that for all i the inequality

with a minimum value of N. This corresponds to falling into the band of each of the M channels of the KB or narrow-band interference itself or a combination component of the form 2f ₁ -f ₂ or f ₁ + f ₂ -f ₃ .

For example, when five interference is applied to a knowledge base, consisting of 31 channels, in order to "clog" all the knowledge channels, the interference must be set to 11, 13, 17, 20, 21 of the knowledge base line.

In the absence of interference Δ = 1 (0 dB), R _p ≈ 10 R _{W int} , the channels of the protection unit, for which α _i = 1, are turned off, which leads to a slight decrease in Δ to the value Δ = (MN) / M, while the values of C _i P _p and K ₃ P _p ^{3 are} substantially less than P _{w int} . With a further increase in R _p in a certain interval, the required signal level does not change, because interference is completely suppressed. And then, after reaching the equality K ₃ P _p ³ = P _{W ext} , Δ decreases in proportion to P _p ³ . The dependence of Δ (on a logarithmic scale) on R _p for the considered distribution of interference frequencies is shown in figure 1.

Thus, having determined the value of R _p at which a further increase in the level of interference leads to a cubic drop in signal / noise at the output of the PFP compression device, the value of K ₃ is determined or, which is the same, the parameter of the multi-signal PFP selectivity characterizing the effect of nonlinear damage due to 3rd order intermodulation effect (mechanism according to claim 4).

To assess the selectivity parameters of PfP, characterizing the impact on the mechanisms of items 1, 2, 3, another distribution of interference frequencies is necessary. Its selection is made from the following considerations. Firstly, the amount of interference should be the same as before. Secondly, their frequency arrangement should minimize the influence of intermodulation components on the signal-to-noise value at the PFP output. Thirdly, the difference between the interference frequencies and the frequencies used in the first measurement should be minimal. The fulfillment of the first and third conditions is necessary to minimize changes in the effects according to items 1, 2, 3 in both dimensions. This makes it possible to provide measurements even in the case of close values of the threshold interference powers, starting from which signal / noise deterioration occurs due to the effects (mechanisms) of Claims 4 and 1, 1, 2, 3. A compromise solution for the distribution of interference frequencies consists in changing part of the frequencies from the set for the first measurement to the frequencies of the nearest BZ channels so that all interference frequencies correspond to the average frequencies of the even (or odd) BZ channels. For example, when five interference is applied to a KB consisting of 31 channels, so that the interference and their combinations of the third and higher orders fall into the band of only even (or odd) KB channels, the interference should be set to 11, 13, 17, 19, 21 of the BZ line. Due to this, due to the intermodulation effect, only half of the KB channels are disabled, even in the case of a large nonlinearity of the receiver path. This corresponds to a deterioration in the signal-to-noise ratio by 2 times (by -3dB) at a level of P _p greater than the value of P _p in figure 1. A further increase in the interference power will lead to a decrease in the signal-to-noise ratio only after the equality С _i Р _п ≈Р _{ш is} fulfilled. A subsequent increase in P _n will lead to a linear decrease in the signal-to-noise ratio. The dependence of Δ (on a logarithmic scale) from R _p for the second distribution of interference frequencies is shown in figure 2. The linear nature of the change in Δ from P _{p is} maintained to the point P _p when the blocking effect begins to appear. However, as noted earlier, for PWM PLCs, the measurement of blocking characteristics is not very significant, because an unacceptable loss in signal reception due to local oscillator noise occurs at significantly lower interference levels.

Thus, to assess the multi-signal selectivity of the receiver, measurements are made for two types of interference frequency distribution:

1) the interference is distributed near the average frequency of the useful signal so that third-order intermodulation products from N interference fall into each channel of the KB (this condition corresponds to the worst case for noise immunity of a broadband receiver);

2) the interference is distributed so that the intermodulation products fall into only half of the BZ channels (this condition corresponds to the minimum possible number of BZ channels suppressed by intermodulation).

The amount of interference is the same in both tests, and changing their frequencies should be the minimum possible.

The structural diagram of the measuring installation for implementing the proposed method for evaluating the multi-signal selectivity of the PWM PWM is shown in Fig. 3 and contains 1.1-1.N high-frequency interference generators whose outputs are connected to the corresponding inputs of the matching device 3, in turn, the output of which is through a serial connection with the equivalent Antennas 4 and PWM ШПС 5 are connected to the input of a signal-to-noise ratio meter or error probability meter. The output of the source of broadband signals 2 is connected to the corresponding input of the matching device 3.

Figure 3 introduced the following notation:

1.1-1.N - high-frequency interference generators;

2 - a source of broadband signals;

3 - matching device;

4 - equivalent receiver antenna;

5 - PfP ShPS;

6 - signal-to-noise ratio meter or error probability meter.

The measurement of the multi-signal selectivity of a broadband signal receiver is as follows.

For receivers having a linear output:

1) to the input of the receiver at the tuning frequency serves a useful signal from the generator 2 with a capacity of P _with ;

2) include interference generators 1.1-1.N at frequencies corresponding first to the first type of distribution, then to the second;

3) fix the signal-to-noise ratio;

4) increase the level of interference power R _p to the beginning of lowering the signal-to-noise ratio at the output in proportion to the cube of the interference power in the first embodiment of the distribution of interference frequencies and in proportion to the interference power in the second embodiment;

5) the multi-signal selectivity of the receiver is determined by the formula

For receivers with discrete information:

1) a useful signal from the generator 2 is supplied to the input of the receiver at the tuning frequency and the power of the useful signal P _s is set at which the signal-to-noise ratio at the output of the receiver is equal to the value typical for this type of receiver;

2) include interference generators 1.1-1.N at frequencies corresponding first to the first type of distribution, then to the second;

3) increase the power of the interference generators 1.1-1.N to a value of R _p , maintaining the signal-to-noise ratio unchanged by changing the power of the useful signal before the cubic growth of the desired signal begins with the first variant of the distribution of interference frequencies and linear growth with the second;

4) the multi-signal selectivity of the receiver is determined by the formula

Literature

1. Aces G.I. Statistical theory of complex signals. M .: Sov. radio, 1977.

2. Andronov I.S., Fink L.M. Discrete messaging over parallel channels. M .: Sov. Radio, 1971.

3. Bogdanovich B.M. Radio receivers with a large dynamic range. M .: Radio and communications, 1984.

4. Mushrooms E.B. Nonlinear phenomena in the transceiver path of communication equipment with transistors. M .: Communication, 1971.

Claims (2)

1. A method for measuring the multi-signal selectivity of a broadband signal receiver with a narrow-band interference protection unit, comprising supplying a mixture of a useful signal and narrow-band noise to the input of the receiver and measuring the signal-to-noise ratio at its output, characterized in that the measurements are made with two types of frequency distribution N narrow-band interference, where N> 2, and the number of narrow-band interference is the same in both forms, in the first form, the frequencies of narrow-band interference are chosen near the average frequency of the useful signal so that the inter third-order dividing components from N narrow-band interference fell into each channel of the protection unit; in the second form, the narrow-band interference frequencies are distributed near the average frequency of the useful signal so that the intermodulation components fall into only half of the channels of the protection unit, and the power level of narrow-band interference is increased in the first distribution type frequency of narrow-band interference before the start of lowering the signal-to-noise ratio at the receiver output is proportional to the power cube of narrow-band interference, and in the second form it is distributed I - to start reducing the signal / noise ratio is proportional to the power of narrowband interferers. 2. The method according to claim 1, characterized in that the measurements are carried out at a fixed signal-to-noise ratio or a fixed probability of error at the output of the receiver, supported by a change in the level of the input useful signal, and the power levels of narrow-band interference are increased before the cubic increase in the level of the useful signal occurs at the first in the form of frequency distribution of narrow-band interference and linear - in the second.

Images