392 THE RADAR RECEIVER 



must be linear ahead of the pad. Frequently it is not convenient to provide 

 an accurate 3-db loss in the receiver. An example is the case where the 

 preamplifier and main IF amplifier are contained in a single unit. In such 

 cases an arbitrary attenuation may be introduced by means of the manual 

 gain control. Measurements are made with two arbitrary output levels; 

 it is only necessary that the receiver have a linear transfer characteristic 

 to the noise at the selected levels. The method is as follows. 



1. Observe output deflection (d-c voltmeter or milliammeter at the 

 second detector) with no additional noise input. Let the deflection 

 be di. The noise is incident to Nrec- 



2. Introduce the noise source and adjust the noise power (A^i) applied 

 to the receiver to produce a deflection d^. The noise is incident to 



3. Insert attenuation a by means of the manual gain control so that 

 the noise A^i produces the deflection di. The noise is incident to 



4. Increase the output from the noise source (A^2) to produce the 

 deflection ^2. The noise is incident to (A^2 + Nrec)oi. 



From these observations, the noise figure can be determined from 



^ rec ^^ T ( T OT \ /-'TT'J 



where Ta = 290° K and T\ and T2 are noise temperatures corresponding to 



A^i and N2. 



In making a noise figure measurement with a dispersed signal source, 

 difficulty is experienced with spurious responses of the receiver. In broad 

 band receivers it is usual to add 3 db to the measured result to account for 

 beating at the image frequency. Because of the small available power from 

 the noise source it is necessary to couple directly to the antenna terminals 

 of the receiver rather than through a directional coupler. As a result the 

 noise figure is not usually measured with the transmitter operating in the 

 case of airborne radar sets. The measurements are also correct only if the 

 noise source has the same impedance as the antenna. 



Sensitivity. With the transmitter operating, additional noise may 

 appear which will degrade the performance. This is particularly the case 

 with high-PRFdoppler radar receiving systems. To determine the perform- 

 ance in detecting and tracking small signals a sensitivity measurement is 

 generally made; this is a measure of the least signal input capable of causing 

 an output signal having desired characteristics. 



In the case of a radar display it is a simple matter to determine the signal 

 power required to obtain a minimum discernible signal. The signal is 



