Modelos computacionais para distribuição dos tempos de silêncio em Soundscapes
The soundscape, or acoustic landscape, is a concept that has received a great attention from ecologists, animal behaviorists and urban planners. The understanding and modeling of soundscape measurements in physics is still incipient. The duration of silence times in a soundscape is defined using...
Na minha lista:
| Autor principal: | |
|---|---|
| Outros Autores: | |
| Formato: | Dissertação |
| Idioma: | pt_BR |
| Publicado em: |
Universidade Federal do Rio Grande do Norte
|
| Assuntos: | |
| Endereço do item: | https://repositorio.ufrn.br/handle/123456789/52629 |
| Tags: |
Adicionar Tag
Sem tags, seja o primeiro a adicionar uma tag!
|
| Resumo: | The soundscape, or acoustic landscape, is a concept that has received a
great attention from ecologists, animal behaviorists and urban planners. The
understanding and modeling of soundscape measurements in physics is still
incipient. The duration of silence times in a soundscape is defined using a microphone and a typical sound intensity threshold; above the threshold, sound is perceived, below the threshold, silence is defined. In many soundscapes it was observed that the statistics of silence durations does not depend on
the used threshold and follows a power law. In this disertation, we build a
null model for the statistics of silences: a microphone is placed in the center
of a square matrix and sound emitters are randomly drawn in the matrix.
Given a threshold, the sound is heard or not, depending on the sound intensity reaching the microphone. Assuming a squared decay of intensity and having all emitters the same power, only emitters within a critical radius are detected at the microphone. Under these conditions, the proposed model
resembles a Bernoulli experiment, in which success consists of sound detection, and failure consists of inaudible sound at the microphone. The ratio between the area of the circle and the matrix defines the success probability of Bernoulli’s experiment. Furthermore, the statistic between two successes
follows a geometric distribution. Computer simulations confirm the theoretically established geometric distribution. Alternative tests with non-constant power have been performed, but the observed distribution in simulations is still a exponential type, and not a power law type. |
|---|