System Noise Temperature

System noise temperature is the equivalent input temperature representing all noise sources that enter a receiver at a stated reference plane. It is usually written T_sys and measured in kelvin. Unlike component noise figure, system noise temperature includes the antenna environment, losses before the receiver and the receiver’s own equivalent noise temperature.

The metric is central in satellite, microwave, telemetry, radar and weak-signal radio links because it connects physical noise sources to G/T, carrier-to-noise density and link margin. A small temperature error can move a link budget by several decibels when the margin is already narrow.

Component Temperatures

Receiver equivalent noise temperature is related to noise factor:

where T_0 is commonly:

For a passive loss with linear loss L at physical temperature T_p, the input-referred noise contribution is:

If the loss is before the receiver, it also scales the receiver noise temperature when referred to the antenna input.

System Reference Plane

For an antenna followed by a lossy feeder and a receiver, a common antenna-input screen is:

where:

• T_ant is antenna noise temperature from sky, ground, atmosphere, radome or background;
• L is feeder loss as a linear ratio greater than one;
• T_p is physical temperature of the lossy element;
• T_rx is receiver equivalent input noise temperature after the loss.

The formula is a reference-plane statement. Moving the reference plane changes which terms are included and how losses are applied.

Worked Example

A microwave receiver has antenna noise temperature:

The feeder loss before the low-noise receiver is:

so:

The feeder is near:

and the receiver equivalent noise temperature is:

The system noise temperature referred to the antenna input is:

If antenna gain is:

then:

Using this in a C/N0 budget with:

and:

gives:

Boundary With Noise Figure

Noise figure is convenient for components and cascades at a reference temperature. System noise temperature is often better when antenna temperature, sky noise, rain, radome loss, feed loss, cryogenic receivers or hot backgrounds matter. The two are connected, but they answer different review questions.

A low receiver noise figure does not guarantee a low T_sys if feeder loss is high or the antenna sees a hot environment. Conversely, a modest receiver may be acceptable when antenna temperature dominates the budget.

Placement matters as much as the component data sheet. Loss before the first low-noise stage both adds its own thermal noise and multiplies the receiver contribution when referred to the antenna. Loss after sufficient low-noise gain may still hurt level and linearity, but it usually has a smaller effect on input-referred system temperature. This is why feedline, waveguide, radome and switch placement are part of receiver-noise design, not only mechanical layout.

Validation Evidence

A defensible system-noise-temperature statement includes frequency, reference plane, antenna pointing, sky or ground contribution, atmospheric and rain condition, radome or feed loss, physical temperature of lossy elements, receiver noise temperature or noise figure, calibration method, uncertainty and whether values are measured, vendor specified or assumed.

Field validation may use hot/cold load tests, sun or sky dips, calibrated noise sources, modem C/N0, receiver noise-floor measurements or comparison between predicted and measured link margin. The evidence should match the receiver state used in service.

Common Mistakes

Common mistakes include mixing dB loss and linear loss inside temperature equations, using receiver noise figure alone as system temperature, ignoring feeder loss before the first low-noise stage, treating clear-sky antenna temperature as valid in rain, moving the reference plane without recalculating terms, and reporting G/T without the antenna gain and temperature basis.

The practical rule is to state the reference plane first, then account for every noise source and loss that exists before the decision point.