7-15] THE SECOND DETECTOR (ENVELOPE DETECTOR) 385 



fore on the carrier level of the IF signal applied to the detector. With high 

 percentage of modulation, the negative peak modulation is distorted 

 incident to the nonlinearity of the detector at low levels. 



In receivers which provide considerable pre-detection integration (IF 

 bandwidths of a few kilocycles per second) it is feasible to obtain high 

 detection efficiency by use of large Rq and Co. When amplitude modulation 

 on the signal must be recovered in such receivers, it is required that Ri 

 and Ro satisfy the relationship 



^1 



Ro + R^ 



> m (7-38) 



where m is the highest modulation percentage that must be recovered 

 without distortion. Failure to satisfy this condition results in clipping of 

 the negative peaks of the modulation. 



When the signal-to-noise ratio of the IF signal is very small and the video 

 bandwidth is less than the IF bandwidth, signal suppression occurs in the 

 second detector.^ This is the result of noise-noise intermodulation at the 

 detector. An approximate expression for signal suppression is 



db suppression ^ - 7 + 20 logio {SIN)if. (7-39) 



It is desirable to provide as much filtering as possible prior to envelope 

 detection to minimize sensitivity loss caused by this signal suppression. 

 However, predetection selectivity is limited by the stability of the IF filters 

 and the tuning accuracy of the receiver. Some receivers, e.g. logarithmic 

 receivers, do not employ a diode envelope detector but obtain the envelope 

 by infinite impedance detection or plate detection in each of the IF stages. 



In monopulse receivers the IF detector which is employed to obtain 

 angular error signals is usually a balanced modulator. This may take the 

 form of either a phase detector or a synchronous detector. Such detectors 

 ideally produce an output only when both signals are applied. The output 

 is primarily dependent on one of the two signals present at the input 

 (provided one signal is much larger than the other). If one of the signals, 

 such as the sum signal in a monopulse receiver, is heavily filtered before 

 applying it to the demodulator, significant improvement in detected S \N 

 can be realized for low ^S" /A^ referred to the difference signal IF bandwidth. 

 Such filtering, however, requires time selection of the sum signal before it is 

 applied to the detector. Such a scheme is, in effect, a carrier reconditioning 

 and exaltation method of detection and, of course, reduces the information 

 rate of the radar. 



8S. O. Rice, "Mathematical Analysis of Random Noise," Bell System Tech. J. 23, 282-236 

 (1944), 24, 46-156 (1945); "Response of a Linear Rectifier to Signal and Noise," J. Acoust. 

 Soc. Am. 15, 164 (1944). 



