unreasonable cost and space requirements. The system might be improved 

 by decreasing the wave guide attenuation or the distance between the 

 antenna and the receiver-transmitter unit. 



The relation between power output and range may be applied also to 

 the minimum discernible signal and the maximum range of detection. An 

 interesting relation can be derived for the absolute minimum discernible 

 signal obtainable. It has been shown [5~\ that the maximum range of de- 

 tection for a target per power output is a function of the receiver band- 

 width. The minimum perceptible signal is in part related to the thermal 

 noise; and because the thermal noise of a receiver is independent of the 

 receiver construction, and the internal noise is usually several times the 

 thermal noise, it is possible to derive the absolute minimum as follows. 

 Thermal noise is generated by the random motion of electrons in a con- 

 ductor and the rms thermal noise voltage which appears across terminals 

 of any circuit element is a function of the frequency interval (receiver 

 bandwidth) only; it is given by 



V n =V4kT-f>R 



where R is the resistance across which the noise voltage is measured, / 

 the bandwidth in cycles per second, T the absolute temperature, and k 

 Boltzmann's constant. If we assume, for the purpose of the limiting case, 

 that the receiver is without any internal noise and all of the noise is gen- 

 erated in the antenna which is at 17°C, the noise power is given by: 



P = 4X10- 15 / watts 



where / is now in megacycles. This means that the minimum detectable 

 signal cannot be less than 4X10~ 15 watts times the bandwidth. For the 

 receiver most used in the field work, the minimum discernible signal was 

 10~ 13 watts for the 20- and 40-mile scales and the bandwidth was 2 Mc/s. 

 Therefore the minimum discernible signal was about 11 decibels above 

 the absolute minimum of 8X10 -15 watts. To illustrate the importance of 

 maintaining a low minimum discernible signal, a target was examined 

 under two known minimum discernible signal values on the same radar 

 with the same power output within two minutes of time. These controlled 

 conditions were provided by making use of the two minimum discernible 

 signal values for the AN/SPS-23 radar; i.e., -91.9 dbm for the 8-mile 

 range, and —100.1 for the 20-mile range scales. A medium growler was 

 brought to within 2,000 yards of the Evergreen and two PPI photographs 

 were taken on either range setting. The growler, shown at the bottom of 

 the figure, was clearly painted on the 20-mile range setting (fig. 42A), but 

 was hardly discernible on the 8-mile setting. This difference in reception 

 for the 20- versus the 8-mile range settings has been observed on many 

 occasions. The propagation conditions at that time were subnormal. The 

 ratio of blips to the number of times the antenna scanned this target was 

 1.0 for the 20-mile range, but only 0.5 for the 8-mile range at 4,600 yards. 

 As the maximum range of this target on the 8-mile scale was 4,850 yards, 



93 



