404 



SUBMARINE TARGET STRENGTHS 



reduce the usual fluctuations enough to enable an 

 evaluation of target strength as a function of range. 



The indirect measurements at MIT, at selected as- 

 pect angles, and at full-scale ranges of about' 250 and 

 630 yd, gave about the same values as those at a full- 

 scale range of 190 yd." This result is consistent with 

 the theoretical computations shown in Figure 16, 

 where the maximum change in target strength is only 

 about 3 db at beam aspect, as the range changes from 

 200 to 1,000 yd. Since for these ranges the target 

 strength was found to be independent of range, it 

 was apparent that the intensity of the light reflected 

 from the models and measured at the receiver varied 

 with range at the same rate as that from a sphere, 

 rather than that from a cylinder or plane, in other 

 words, inversely as the fourth power of the distance. 

 The shortest range was approximately twice the 

 length of the submarine. It is likely that this relation 

 would not hold at much closer ranges where target 

 strengths would be expected to depend strongly on 

 the range. 



Target strength was found to depend on the range 

 in the acoustical model experiments at Mountain 

 Lakes, for full-scale ranges of 85, 170, and 250 yd. 

 The submarine model behaved as a cylinder, not as a 

 sphere or plane. For reflection from a cylinder, the 

 echo level should decrease 9 db when the range is 

 doubled as long as the range is not much greater than 

 the length of the cylinder (see Section 20.4.3); for .a 

 sphere the same increase in range causes a drop of 

 12 db. It was found that the echo level at beam as- 

 pect actually dropped 8.8 db as the full-scale range 

 was doubled, from 85 to 170 yd. Thus the hull at 

 beam aspect and at short ranges behaves as a cyl- 

 inder, as might be expected from its large radius of 

 curvature in the horizontal plane. 



In general, the indirect measurements agree with 

 the theoretical predictions of the dependence of tar- 

 get strength on range. This predicted dependence 

 should be most marked at ranges less than 200 yd; 

 experimentally, it was observed and verified only at 

 ranges less than 200 yd. However, too much impor- 

 tance cannot be attached to these results, since the 

 measurements were made only at three particular 

 ranges in each of the two indirect measurements, and 

 since experimental errors were so large. 



23.5 DEPEiNDENCE ON PULSE LENGTH 



When short pulses are used instead of continuous 

 sound, target strength may depend on the lengths of 



these pulses. Sections 19.3 and 20.7 discussed in an 

 elementary way the effect of pulse length on meas- 

 ured target strengths. For short pulses — signals 

 whose length in the water is less than the length of 

 the target in the direction of the sound beam — the 

 echo level and therefore the target strength will de- 

 pend on the signal length. The exact variation of 

 target strength with signal length, however, depends 

 on whether peak echo intensities or average echo 

 intensities are used in computing target strengths. 



23.5.1 



Theory 



In most direct measurements of target strength, 

 peak amplitudes are measured from the oscillograms 

 rather than average amplitudes, because echo pro- 

 files are so irregular that the average amplitudes over 

 the length of the echo would be difficult to measure. 

 A simple analysis given in Section 19.3.1 shows that 

 for long pulses the average echo iatensity is inde- 

 pendent of signal length, while the echo length varies 

 with the signal length. For short pulses on the other 

 hand (see Section 19.3.2), the average echo inten- 

 sity is approximately directly proportional to the 

 signal length, while the echo length remains con- 

 stant. This analysis applies to square-topped pulses 

 striking an extended single target. 



It is arbitrary whether peak amplitudes or aver- 

 age amplitudes are used to compute target strengths. 

 Peak amplitudes are easier to measure. In addition, 

 most other imderwater sound measurements, such 

 as those undertaken in the investigation of recogni- 

 tion differentials, are based on peak amplitudes. It 

 may be, however, that the ear, or the sound range 

 recorder, or other detection devices may respond to 

 the average echo intensity instead of the peak echo 

 intensity, or to the total energy in the echo. There- 

 fore a comparison of average and peak echo ampli- 

 tudes and their variation with signal length might be 

 a profitable study. 



The change of average and peak echo intensities 

 with pulse length has been investigated theoreti- 

 cally,^* as described in Section 21.6.4. First it is as- 

 sumed that the length of each individual peak in an 

 echo is approximately equal to the signal length. 

 Then it is assumed that these peaks are statistically 

 independent, or distributed at random throughout 

 the length of the echo. By assuming that the echo 

 is essentially a group of rectangular peaks, each of 

 which follows the Rayleigh distribution for succes- 

 sive echoes and each of which is independent of the 



