GB/T 11299.5-1989 Satellite communication earth station radio equipment measurement methods Part 1: Measurements common to subsystems and subsystem combinations Section 5: Noise temperature measurement

This standard specifies the measurement methods for noise temperature and noise figure. This standard is applicable to the measurement of noise temperature and noise figure of linear subsystems and/or subsystems combined with appropriate interfaces. The measurements of special systems or subsystems are given in the second and third parts of this series of standards. GB/T 11299.5-1989 Satellite communication earth station radio equipment measurement methods Part 1: General measurements for subsystems and subsystem combinations Section 5: Noise temperature measurement GB/T11299.5-1989 standard download decompression password: www.bzxz.net

National Standard of the People's Republic of China
Methods of measurement for radio equipment used in satellite earth stationsPart 1: Measurements common to sub-systemsand combinations of sub-systemsSection Five-Noise temperature measurementsThis standard is one of the standards in the series of "Methods of measurement for radio equipment used in satellite earth stations"1 Subject content and scope of application
This standard specifies the measurement methods for noise temperature and noise figure. GB11299.5-89
This standard is applicable to the measurement of noise temperature and noise figure of linear subsystems and/or combinations of subsystems with appropriate interfaces: the measurements of special systems or subsystems are given in the second and third parts of this series of standards. 2 Introduction
Noise is generated externally or internally by the equipment. External noise mainly comes from noise sources outside the earth, as well as thermal radiation from the atmosphere and the earth's surface. Internal noise comes from thermal noise and circuit noise, such as shot noise in vacuum tubes, intermodulation noise, and fluctuation noise in semiconductors and fluctuation noise caused by the movement of magnetic domain interfaces in ferrite devices. It is appropriate to use noise temperature to measure the noise power generated by a system or subsystem. Noise temperature is always an "equivalent" temperature, not an actual temperature. It is a measure of the effect of all noise sources (thermal noise sources and non-thermal noise sources). 3 Definitions
The following definitions apply to this standard.
3.1 Noise Power Spectral Density
Noise power spectral density is defined as
N(f)dPn(f)
Formula: N(f) - noise power spectral density, which is a function of frequency (W/Hz)
dp, (f) is the total noise power contained in the frequency interval df. Note: In practical applications, the noise power spectral density can be considered as the noise power contained in a 1Hz bandwidth, represented by N. 3.2 Available Noise Power Density
Available noise power density refers to the noise power density transferred from the noise source to the matched load. 3.3 Noise Temperature
Noise temperature is the ratio of available noise power density to the Boltzmann constant: Approved by the Ministry of Electronics Industry of the People's Republic of China on March 1, 1989 (1)
Implementation on January 1, 1990
GB 11299.5-
Where:. —Noise temperature, and is a function of frequency; N
k——Boltzmann constant, k1.3805×10-2J/K. 89
Noise temperature is always an "equivalent" temperature rather than an actual temperature. Even when the noise power comes from a heat source, the noise temperature is not just the thermal radiation of a single object at a single temperature. T, retaining the subscript indicates equivalent temperature, omitting the subscript indicates actual temperature, 3.4 Average noise temperature
Average noise temperature T, defined as:
Where: B-
Noise bandwidth (see 3.5);
PN-Total noise power within the noise bandwidth (B). 3.5 Noise Bandwidth
If G(f) represents the available power gain of a noiseless linear network with respect to frequency, and G. is the available power gain of the network at the nominal center frequency or reference frequency, then its noise bandwidth (B) is defined as: the bandwidth of an ideal noiseless filter with rectangular frequency characteristics and available power gain G, and the noise power output is equal to the output of the actual filter (see Figure 1). Here it is assumed that the noise power density does not vary with frequency. Therefore, the noise power at the output of the filter can be expressed by the formula: N·G(f)df = NBG. (W)
Where: N
The available noise power density at the input of a filter is obtained by equation (4) and the noise bandwidth (B) is:
\G(f)df(Hz)
**(4))
It should be noted here that the value of the noise bandwidth (B) is not a constant parameter of the filter. It depends on the choice of the center frequency (), and the center frequency (f) determines the available power gain (G,). 3.6 Operating noise temperature or system noise temperature The operating noise temperature of a subsystem or a combination of subsystems refers to the noise temperature composed of all noise sources generated by external and internal (i.e., within the device under test) and is expressed as T. p. For a two-port device or multiple two-port II devices connected in series, the operating noise temperature reflected at the first-level input port is given by the following equation:
T =T +T.
Where: T—noise temperature of the input gate; T. The equivalent input noise temperature of a device under test consisting of a device or a chain of devices. "System noise temperature" is often used to refer to the noise temperature of a complete communication system, denoted by T. 6
"Operating noise temperature" is not only applicable to complete communication systems, but also to more general situations involving individual noise sources, individual systems or subsystems under test, and arbitrary configurations of loads. 3.7 Reference noise source
A reference noise source is a noise source whose noise temperature remains constant during measurement. It is usually not necessary to know the noise temperature of the reference noise source. If the noise temperature of a reference noise source is known, it is called a standard noise source. The noise temperature of the reference noise source may be the ambient temperature, a typical example being a hot load at ambient temperature. The reference noise source often consists of a hot load or a cold load and an attenuator at or near ambient temperature (with a base attenuation value of 1) (see Figure 2). In this case, the reference noise temperature (T) is given by: 44
Where: T. is the noise temperature of the reference noise source; T is the ambient temperature of the attenuator.
3.8 Equivalent input noise temperature of a two-port device GB 11299.5---
The equivalent input noise temperature of a one-port device is an imaginary noise temperature. When added to an ideal noise-free two-port device with the same input impedance and gain as the actual device, the noise temperature will produce the same output noise power density as the actual device.
When a two-port device with an equivalent input noise temperature of T is connected to a noise source with a temperature of Tu, the noise power density (N) output by the device is:
N -- (Tu + T)hG(W/Hz)
Where: - the gain of the one-port device.
If T is the average equivalent input noise temperature within a given bandwidth (B), the output noise power (P) of the two-port device is: Ph -- (Tsou -+ T,)kGB
where: B is the noise bandwidth;
T is assumed to be constant within the noise bandwidth (B). (9)
Unless otherwise specified, the equivalent input noise temperature refers to the average noise temperature within a given bandwidth, expressed as T. The "input\reference plane" is selected at the input port of the device under test. 3.9 Average noise figure
The average noise figure (below) of the port device is the ratio of the total noise power (P) delivered to the matched load by the device when the noise temperature at its input port is 290K to the noise power (P.) that can be obtained at the output of an ideal noise-free device under the same conditions. Pa
-KT,GB
where: T,--290K.
For devices with gain in multiple frequency bands (such as the image frequency of a heterodyne system), the denominator Ps only includes the noise power at the input port in the same frequency band as the modulation signal. This situation can be applied to satellite communication systems and is called "narrowband noise figure". The relationship between the average noise figure (F) and the average equivalent input noise temperature (T.) can be obtained by the following formula: P, a 290kGB + kT, GB
kGB(290 + T.)
According to formulas (10) and (11), we can get:
AGB(290±T)=1+ 20
290kGB
T, = 290(F -- 1)
Long-term average noise factor is usually referred to as noise factor, expressed as F. 4-Generally, the measurement methods of noise factor (F) and equivalent input noise temperature (T) are divided into broadband method and narrowband method: the broadband method often uses a noise generator as the measurement signal generator, while the narrowband method uses a continuous wave signal generator as the measurement signal generator. The commonly used broadband measurement methods are:wwW.bzxz.Net
a. Y factor method;
b.3dB attenuator method;
Automatic noise figure meter (ANFM) method.
GB11299.5—89
The continuous wave method using unmodulated signals is the most widely used band measurement method. This method is used from very low frequencies to tens of light hertz. The choice of method in a specific case will depend on many factors, which are: a. The required accuracy; the required instrument; the availability of the instrument; the frequency range; the type of device under test; the convenience; the measurement speed. Table 1 summarizes the main features of the various measurement methods. The device under test can be a single subsystem (such as a low noise amplifier) ​​or a combination of subsystems (such as a low noise amplifier and a downconverter). The second and third parts of this series of standards will give test plans suitable for special systems or subsystems. The main features of the various measurement methods are based on the method. ||(a) Power meter method
(h) Variable attenuator method
3d13 attenuator method
(a) Variable source method
(b) Fixed source method
Automatic noise figure meter (ANFM) method
Continuous wave method
5 Measurement method
5.1 Y factor method
There are two commonly used Y factor methods:
a. Power meter method;
b. Variable attenuator method
10%～25%
5%～20%
5%～~20%
5%～20%
The main difference between these two methods is the measurement method of the Y factor. 5.1.1 Power meter method
Frequency range
10kHz～ 30GHz
10kHz-- 30GHz
IMHz~30GHz
10k Hz --- 30Gz
10MHz ~- 30GHz
1kHz～40GH2
Measurement speed
As shown in Figure 3, this method uses a pair of random noise generators and a power meter. The noise temperature (T) of the hot noise generator is higher than the noise temperature (T\) of the cold noise generator. The hot and cold noise generators provide known available power to the device under test, and the output power is measured with a power meter. The ratio of the two output powers corresponding to the two input powers is the Y factor. The equivalent input noise temperature (T) and noise factor (F) can be calculated based on the measured Y factor and the noise temperature of the two known noise sources. This method has high accuracy. , especially when the measurement configuration is automated, under optimal conditions the measurement error can be as small as 1% (0.(4dB); the typical error is 2%~8% (0.1～0.36dB). Therefore, this method is usually selected when high accuracy and high precision measurement are required and a precise power meter is available.
The measurement steps are as follows:
Refer to Figure 3, the thermal noise generator is connected to the input port of the device under test, and the power meter reading is recorded for 1h. GB11299.5--89
Disconnect the thermal noise generator, connect the cold noise generator to the input port of the device under test, and record the power meter reading.. b.
Calculate the Y factor from formula (14):
The equivalent input noise temperature (T.) can be calculated as follows: (T+T)KGB
P=(T+ TOKGB
Where: G—gain of the device under test;
B-——noise bandwidth of the device under test;
Th-——noise temperature of the thermal noise generator; T.
Noise temperature of the cold noise generator.
From formula (15):
According to the value of T., the noise figure (F) can be calculated from the following formula: F=1+
Th—YT
= 1 + 290(Y=1)
The noise figure F(dB) expressed in decibels is:
F(dB) - 10 logioF
5.1.2 Variable attenuator method
. (16)
As shown in Figure 4, this measurement method is similar to that of 5.1.1 except that a precision variable attenuator and a signal level indicator are used to measure the Y factor. The accuracy achievable by this method is similar to that achievable by the power meter method, but this method does not require a precision power meter. The main measurement error comes from the attenuator error rather than the error of the noise source or the signal level indicator. Therefore, it is necessary to ensure that the variable attenuator used has sufficient High enough accuracy. The measurement steps are as follows:
Refer to Figure 4, connect the cold noise generator to the input port of the device under test, adjust the variable attenuator so that the indicator is close to full scale. Record the indicator reading T and the attenuator reading A expressed in decibels. (dI3). b. Disconnect the cold noise generator and connect the hot noise generator to the input port of the device under test. c.
Adjust the attenuator so that the indicator gets the same reading T. , record the attenuator reading A expressed in decibels (dB) Considering the following situation, the Y factor can be calculated:
P,(dB)— An(dB) = P(dB) - A(dB)19）
Where: P':(dB) and P(dB) are the noise powers at the output terminal I1 of the device under test when its input port is connected to the hot and cold noise generators respectively.
Mountain formula (19) is obtained:
Y(dB) P,(dB) - P(dB)
=Ar(dB) A(dB)
According to Y(dB), the Y factor is obtained:
Y= 1or
Knowing the Y factor, the equivalent input noise temperature (T.) and noise figure (F) can be calculated according to 5.1.1. This method is not suitable for measuring low-gain devices under test because the noise influence of the attenuator is too large to reduce the measurement accuracy. 5.2 3dB attenuator method
GB11299.5---89
This method is similar to the method in 5.1.2, except that a 3dB precision fixed attenuator is used instead of a variable attenuator, and a fixed Y factor value of 2 (i.e. 3dB) is used. Therefore, a noise source with a continuously adjustable output level must be used. There are two methods that can be used. The most common method is to use a temperature-limited thermionic vacuum diode as a shot noise source. The output level of this noise source is easily adjusted.
Another method is to use a fixed-level noise generator and a variable attenuator, the latter of which is used to adjust the noise level added to the device under test. However, this method is not commonly used.
Which method to choose depends on the available instruments and the frequency range of the device under test. 5.2.1 Variable Source Method
As shown in Figure 5, the diode noise generator provides an adjustable known input power to the device under test, and the noise bandwidth should be greater than the bandwidth of the device under test.
The signal level indicator provides a reference level of the output power, but its absolute value is not important. Due to the limitation of the available frequency range of the shot noise generator, this method is usually only used in the lock rate range of 1MHz3GHz, and the measurement accuracy can reach 5% (0.2dB). The measurement steps are as follows:
a. Referring to Figure 5, the output port of the device under test is connected to the signal level indicator, and the emission current of the diode noise generator is reduced (the generator is still connected to the device under test). In this state, the noise power at the output of the device under test caused by the residual input thermal noise is: P=(T+T)IGB
Where: G and B are the gain and bandwidth of the device under test; T. — is the equivalent input noise temperature of the device under test; T is the ambient temperature of the generator source impedance.
b. Adjust the sensitivity of the signal level indicator so that its indication is close to full scale, and record the indicator reading I. c. Disconnect the device under test from the signal level indicator, and connect the 3dB attenuator to the test system, (22)
d. Increase the emission current of the noise generator until the same reading I. is obtained on the indicator. In this state, the noise power at the output of the device under test is:
P, = (T + T)kGB = 2P
Where: T, — total (shot and thermal) noise temperature of the diode noise generator. e. Record the reading of the noise figure F in decibels indicated on the noise generator. If the transmit current is not calibrated in terms of noise figure (dB), the value of T. is calculated by equations (22) and (23): (T, +T)RGB
(T, +T)AGB
T = T. - 2T.
The noise figure (F) is obtained from the following equation:
5.2.2 Fixed source method
As shown in Figure 6, a fixed level noise generator, a calibrated variable attenuator, a 3dB fixed attenuator and a signal level indicator are used. The generator/attenuator combination plays the same role as the diode noise generator described in Section 5.2.1. When a highly precise noise generator and attenuator are used, this method has high accuracy. Under optimal conditions, the measurement error can be as small as 2% (0.1dB), and the typical error is 5% to 20% (0.2 to 1dB). This method is applicable to the frequency range of 10kHz to 30GHz. The measurement steps are as follows:
Refer to Figure 6, the output port of the device under test is connected to the signal level indicator, the noise generator is turned off, and the attenuator is still connected. a.
GB 11299. 5 -- 89
In this state, the noise power at the output end of the device under test is: P, =(T+T)KGB
Formula: Te
Equivalent input noise temperature of the device under test; T\,---the ambient temperature of the generator source impedance. Adjust the sensitivity of the signal level indicator so that its indication is close to full scale, and record the indicator reading 1. h.
Disconnect the device under test from the signal level indicator and connect the 3d3 attenuator to the test system. C
Turn on the noise generator and adjust the variable attenuator until the same reading I is obtained on the indicator. d
In this state, the noise power at the output of the device under test is: P2= {T,α+T,(1
where: Th
noise temperature of the noise generator;
transmission coefficient of the variable attenuator;
T-ambient temperature of the variable attenuator;
T—equivalent input noise temperature of the device under test. - α) + TJkGB - 2P
e. Record the attenuation value A of the variable attenuator calibrated in decibels, and use formula (29) to convert it into the transmission coefficient α, α = 1o=4at
The T value is obtained from formulas (27) and (28):
T + T(α) +TKGB
T + TJGB
Therefore:
T=(T, T)αT,
5.3 Automatic Noise Figure Meter (ANFM) Method
As shown in Figure 7, this method uses a switch-type random noise generator and an automatic noise figure meter (ANFM) to automatically measure the noise figure
The automatic noise figure meter periodically switches the noise generator between two known output power levels and automatically calculates the noise figure (F), the value of which is directly displayed on the panel. This method has moderate accuracy because the analog circuits in the switch-type noise generator and the automatic noise figure meter are difficult to calibrate.
Under optimal conditions, the measurement error can be as small as 5% (0.2dB), and the typical error is 5% to 20% (0.2~1dB). The nominal usable frequency band of commercial instruments is 10MHz to 30GHz.
The measurement steps are as follows:
Refer to Figure 7, the noise generator current and other control knobs are adjusted according to the manufacturer's instructions, and the front panel switch is set to display the noise factor on the panel. If necessary, the F value can be calculated using the following formula: F 10
Where: F (dB) is the noise factor expressed in decibels. And T. is calculated according to formula (33):
T. - 290(F - 1)
5.4 Continuous Wave Method
The continuous wave method is a narrowband measurement method, as shown in Figure 8. It mainly uses a variable frequency sine wave signal generator. If the noise bandwidth of the device under test is known, this method can be used to replace those methods using a noise generator. At the test frequency (f), the power input from the signal generator to the device under test is known as P, and the output power is measured in two states: the continuous wave signal is cut off, and the output power (P,) in this case is completely noise. a.
b. Connect the CW signal. In this case, the output power (P2) is a mixture of the noise and the applied CW signal. GB 11299. 5-89
The noise figure 1 can be calculated based on the output power (P) of the signal generator and the measured output power of the device under test. The measurement steps are as follows:
Refer to Figure 8, tune the CW generator to the frequency. , and set the output level to CEN (the generator is still connected to the device under test). a.
Record the output power (P) of the system. In this state, the noise power (P) at the output end of the device under test is: P=(T+T)KGB
武: T
Equivalent input noise temperature of the device under test; - Ambient temperature of the source impedance of the CW generator. . (31)
b. The frequency of the continuous wave signal generator is still , and the output level is at least 20dB greater than P, but the device under test must not be saturated. Record the available power level (P) of the signal generator and the output power P2 of the device under test. P2 = PG + (T. + T) KGB ||tt || T. The value is calculated by the following formula:
P = (T. + T) kGB + PG
(T+ TRGB
1+(T+T)KB
The noise figure (F) is obtained by the following formula:
This means that the noise bandwidth of the device under test must be known, otherwise, the method given in Appendix A should be used for measurement. 6 Result expression
The measurement results should be expressed as follows:
The equivalent input noise temperature (T.) is expressed in units of absolute temperature (K); the noise figure (F) is expressed in units of decibels (dB). 7
Details to be specified
When this measurement is required, the equipment technical conditions should include the following: Measurement method used:
Required measurement equipment configuration accuracy:
Measurement Frequency range of the base;
Maximum input level applied to the device under test; Reference plane selected when making measurements.
Thermal noise
Filter
GB11299.5-89
Noise within bandwidth (B)
Figure 1 Schematic diagram of flat noise bandwidth
Noise source
Variable attenuator
(ambient temperature is T.
Attenuator adjusts the noise temperature of the reference source
Thermal noise
Generator
Device under test
Generator
Power meter| |tt||Figure 3 Equipment configuration for measuring Y factor using power meter method Hot noise
Generator
Cold noise
Generator
Device under test
Attenuator
Signal level
Indicator
Figure 4 Equipment configuration for measuring Y factor using attenuator method Temperature-limited diode
Noise generator
Device under test
3dB attenuator
Signal level
Indicator
Figure 5 Equipment configuration for measuring noise figure using 3dB attenuator method with variable level noise source 51
Noise generator
Variable attenuation Figure 6 Equipment configuration for measuring noise figure using the 3dB attenuator method of a fixed-level noise source Switched noise generator Equipment under test Automatic noise coefficient meter Figure 7 Configuration for automatic measurement of noise figure Continuous wave signal Generator Splitter Counter Figure 8 Equipment configuration for measuring noise figure using the continuous wave method Power meter GB 11299.5-89
Appendix A
Noise Bandwidth Measurement
(Supplement)
The noise bandwidth (B) of the device under test can be obtained by numerical integration. The corresponding measuring equipment configuration is shown in Figure A1. When the method of Figure A1(a) is used, a spectrum analyzer operating in linear mode is required. The measurement steps are as follows:
a. The X·Y recorder uses linear chart paper, the device under test is not connected to the spectrum analyzer, and a zero signal baseline is drawn on the record. b. The output of the device under test is connected to the spectrum analyzer and a linear amplitude/frequency characteristic is drawn. This characteristic curve should extend from both sides of the center frequency until it intersects the baseline.
Use the counter in the sweep generator to establish two frequency reference points on the baseline in order to determine the frequency increment per division. For each division (frequency scale) on the baseline, record the amplitude scale (A:) and its square (A). d,
To find the noise bandwidth, first calculate the sum of the squares of all amplitudes, divide by the square of the maximum amplitude, and then multiply this value by the frequency increment per division (A) to obtain the noise bandwidth (B).MAN
Figure A1(b) shows an alternative measurement equipment configuration. The measurement process is the same as above, except that the input frequency is manually adjusted and the RF voltage is recorded on a linear scale. The frequency increment should be selected so that approximately 100 readings are taken between 30 dB down points on either side of the center frequency of the DUT passband.
In both measurement equipment configurations, the DUT must be terminated with a nominal load impedance, and it must also be ensured that the input level to the DUT is constant at each test frequency
It should be noted that the first method cannot be used when frequency changes occur in the DUT. RF
Swept-frequency signal generator with frequency counter
RF output
RF input
X output,
Spectrum analyzer
(linear mode)
RF signal
Amplifier
Y output
Counter
Splitter
Equipment under test
Recorder
Equipment under test
Figure A1 Measurement of noise bandwidth
Additional remarks:
This standard was drafted by the 54th Institute of the Ministry of Electric Industry. RF
Voltmeter
Tip: This standard content only shows part of the intercepted content of the complete standard. If you need the complete standard, please go to the top to download the complete standard document for free.