×

System for detecting the siren of an approaching emergency vehicle

  • US 5,287,411 A
  • Filed: 08/11/1992
  • Issued: 02/15/1994
  • Est. Priority Date: 07/27/1990
  • Status: Expired due to Term
First Claim
Patent Images

1. A system for detecting the presence of sound signals varying in frequency in a predetermined manner over a predetermined spectrum and for determining the direction from which said sound signal are emanating, said variation in audio frequency defining a sound pattern, said system including in combination(a) a plurality of directional sound pickup means each connected to a sound selection input and each facing in a different selected direction to produce input signals representative of the volume and frequency of said sound signals;

  • (b) signal selection means for monitoring said sound selection inputs and for at any given time producing digital output signals derived from only one of said sound selection inputs and consisting of a stream of counts;

    (c) first counting means responsive to said digital output signals to(i) detect sound frequencies by continuously repeatedly taking time samples which indicate the time required for said first counting means to receive a pre-selected number of the counts in said digital output signals, and(ii) produce frequency output signals representing the time measured during each of said time samples,said time measured during each of said time samples indicating the time elapsed from initiation of counting until the instant the required number of counts is received by said first counting means;

    (d) second counting means responsive to said frequency output signals of said first counting means to obtain at least four (4) successive frequency samples;

    (e) means for storing said successive frequency samples;

    (f) random pattern recognition means for recalling said successive frequency samples and, when said samples include at least two successive increasing samples each having a frequency randomly falling in said predetermined audio spectrum followed by at least two decreasing frequency samples each having a frequency randomly falling in said predetermined audio spectrum, for producing a pattern recognition output signal, said random sound pattern recognition means producing said pattern recognition output signal without requiring that said successive frequency samples(i) correspond to and fit a pre-selected series of specific reference values defining a specific pre-selected sound pattern,(ii) each fit within one of a pre-selected stepped series of specific frequency bands within said predetermined audio spectrum,(iii) correspond to a pre-selected sequence of frequencies each lying within a pre-selected stepped series of specific frequency bands within said predetermined audio spectrum.

View all claims
  • 0 Assignments
Timeline View
Assignment View
    ×
    ×