606 BELL SYSTEM TECHNICAL JOURNAL 



the noise in a room by excluding street noise, using quieter office 

 equipment or sound absorbing material. Each of these measures will 

 involve a certain expense and will reduce room noise a certain amount. 

 Which of these measures will be of greatest benefit and most economi- 

 cal, i.e., give the greatest noise reduction per dollar expenditure? 



In another assumed case, the noise from a certain noise producer, 

 a piece of machinery, a ventilating system, etc., is known. Will this 

 apparatus be objectionable in the particular location for which it is 

 considered? 



These and similar questions can be answered by computation while 

 the project is still in the planning stage, whereas measurements can 

 be made only after the change has been made, i.e., after the money 

 for the project has been spent. 



The computation method, as these illustrations show, is useful in 

 specifying apparatus and in planning working or living quarters from 

 the noise standpoint, in studying the comparative effectiveness of 

 various noise reducing means, etc. The method has been used in 

 many practical problems in this way, with satisfactory results. In a 

 number of applications covering noise from 55 to 75 db sound level 

 the computed and measured absolute values agreed, on the average 

 within 1.0 db, and in the worst case within 2.0 db. Computations of 

 the effect resulting from modifications of the noise sources were checked 

 within closer limits. A few illustrations of applications are given at 

 the end of this paper. 



The Impulsive Character of Noises 

 Acoustical noise frequently is composed of sounds from a large 

 number of sources each of which produces a relatively small propor- 

 tion of the total noise. Usually these individual noise sources are 

 discontinuous, consisting of a series of individual impulses. Consider, 

 for instance, noise from a busy street. The hearers' first impression 

 is that of a general roar. After a period of listening, however, a 

 variety of individual sources may be distinguished, such as: The 

 movement of automobiles, squeaking of brakes, whistles, street car 

 wheels and bells, hammering and riveting from building operations, 

 footsteps and conversations of people, etc. Each of these sources has 

 a distinct time pattern and even those that appear most steady can 

 frequently be broken up into impulses. For instance, the noise from 

 an automobile passing down the street is composed of a series of impact 

 noises which depend on unevenness of the pavement, the driving gears, 

 number of cylinders in the engine, etc.; the hum of conversation of 

 people in the street is composed of individual syllabic speech sounds 

 from the different talkers. 



