#### What is in this article?:

- Passive Standards Aid Noise System Verification
- Verification Approaches
- Practical Considerations

Passive noise standards can be used to calibrate an validate the performance of impedance-tuner-based noise measurement systems.

Amplifier and oscillator designers can benefit from load-pull and noise test sets by using them to vary the load and source impedances to a device under test (DUT). Such test sets must be verified, however. Load-pull test sets are verified by using a “back-to-back” or “DeltaGt” verification method as an objective measure for accuracy. Noise test sets instead require some form of noise standard. Active noise standards, such as noise sources based on avalanche diodes, must be calibrated in specially equipped standards laboratories; one such facility is the National Institute of Standards & Technologies (NIST) in Boulder, CO.

**Related Articles**

• Filtering EMI And RFI Noise

• Cross-Spectral Phase Noise Is Measured On Terahertz Source

• REVEALED: Source Of Low-Frequency 1/f Noise

Passive noise standards are easier to use for verification, since passive components exhibit a noise figure and generate available power as white thermal noise. Almost any attenuator or termination can serve as a passive noise standard. In the case of either active or passive noise standards—namely, the four noise parameters—the small-signal electrical and physical properties must be precisely known so that its noise characteristics can be determined at a known and controlled temperature. These parameters are minimum noise factor (F_{min}), noise resistance (R_{n}), magnitude at optimum impedance match (|Γ_{opt}|), and phase at optimum impedance match (∠Φ_{opt}).

An avalanche diode is an example of an active device used as both a noise source and a noise calibration standard. Noise-figure measurements require a source of two noise power levels, which the diode can provide when biased on and off. It is a suitable active device for measuring noise figure (F) in amplifiers and other DUTs. It is typically teamed with some form of radiometer or measurement receiver—such as a spectrum analyzer or noise-figure meter—to perform the noise measurements.

Practical noise standards are those that users can produce themselves, created from known materials and with known dimensions. Attenuators with different values, such as 3 and 6 dB, have been used as passive noise standards for tuner-based four-noise-parameter test sets.

Noise measurements are never routine, whether making them in impedance-matched or unmatched conditions. Nevertheless, measurement accuracy is always vital, as is collecting adequate data from a noise measurement system for processing. In calculating the noise factor, F, the known noise of a noise standard can be used to measure the difference between the signal-to-noise ratio (SNR) of a DUT at the input and at the output of the DUT:

F = (S_{in}/N_{in})/(S_{out}/N_{out}) (1)

where:

S_{in} = the input signal level;

N_{in} = the input noise level;

S_{out} = the output signal level; and

N_{out} = the output noise level.

Focus Microwaves has developed a number of measurement system solutions for verification, with different types of matched networks (fixed attenuators) or mismatched networks used as passive noise standards. Of course, the ambient temperature must be known and the electrical RF parameters in form of scattering (S) parameters of such networks must be accurately measured, to be able to calculate the theoretical noise parameters and compare with the noise parameters then measured by the noise system.^{1}

Because of Eq. 1—and considering that a passive network produces the same available output thermal noise power, N_{out}, as it receivers at its input, N_{in}, or N_{out} = N_{in}—the noise factor, F, can be described as a ratio of available signal powers, F = S_{in}/S_{out}. And since only “available” power is considered, then F = 1/available gain = available loss, as given by Eq. 2^{2}:

F = [|1 - Γ_{S}*S_{11}|^{2} * (1 - |Γ_{0}|^{2})/(|S_{21}|^{2} * (1 - |Γ_{S}^{2})] (2)

where:

Γ_{0} = S_{22} + S_{12} * S_{21}/(1 - Γ_{S} * S_{11});

Γ_{S} = (1 - y_{s})/(1 + y_{s});

y_{s} = Y_{s}/ Y_{0}, with Y_{0} = 20 mS; and

S_{ij} = the S-parameters of the network.

It is therefore straightforward to compute the four noise parameters of a passive network by calculating F for at least four source impedances and solving the four equations for minimum noise factor (F_{min}), equivalent noise resistance (R_{n}), and optimum source reflection factor (Γ_{opt}).