Acoustic monitoring of urban noise and biodiversity for green future using IoT-Sound-Radar and AI for event classification

Excessive traffic noise pollution is the second most important environmental cause of health problems in Europe. The European Union has launched the ""Zero pollution Action Plan"" as part of the European ""Green Deal"" to decrease the proportion of people chronically disturbed by traffic noise by 30%. At the same time, the decline in biodiversity has become a critical issue as species and populations disappear from ecosystems at the rate of mass extinction event. Environmental noise is one of the most significant influencing factors, as evidenced by the Horizon Europe Call: ""Impact of light and noise pollution on biodiversity"". A literature survey on environmental noise, biodiversity, and urban noise pollution clearly indicates that the lack of suitable and affordable hardware and software remains one of the most significant obstacles for providing relevant data in biodiversity monitoring and in urban noise pollution. Methods for monitoring biodiversity based on bioacoustics lack simultaneous localization, identification, and classification of animals. The literature survey also revealed the urgent need for a new robust methodology for noise source localization, identification, and classification that can promptly distinguish among different noise sources, enabling spatiotemporal distribution in urban environments. This project addresses these challenges by proposing a new methodology based on a network of IoT microphone arrays that will allow for the localization of each sound event every 50 milliseconds and its classification every second. The proposed methodology will enable tracking and counting of selected species that produce sounds within a designated space. It will also allow for the counting of the number of different animals that produce sound within the observed area. The use of a low-cost network of IoT microphone arrays with integrated signal feature extraction will enable the synthesis of advanced dynamic noise maps. These maps will provide more information for evaluating noise impacts on people and for developing more effective noise abatement/control measures. The proposed methodology based on IoT microphone arrays addresses both issues, i.e., monitoring of biodiversity and noise pollution. Advanced Sound Maps with identified sound sources will be synthesized from measured parameter called Immission Directivity on each IoT microphone array. By combining the spatial sound feature Immission Directivity with psychoacoustic features of sound events, AI algorithms will be able to classify different sound sources, providing valuable information to policymakers to identify and address specific noise sources and their impact on human health and well-being. The proposed methodology has a high potential to be a game-changer in addressing urban noise pollution and monitoring biodiversity. This project proposal aligns with the guidelines set forth by the EU, as well as Slovenian strategy 2030 and objectives of this tender. The proposed topic represents the cutting-edge of current scientific research and aligns with the: 1) Horizon Europe Call: ""Impact of light and noise pollution on biodiversity"", 2) BIODIVERSA+ call: ""Innovation and harmonization of methods and tools for the collection and management of biodiversity monitoring data"", 3) ""Environmental noise guidelines"", by the WHO and 4) EU commission ""Zero pollution action plan"". Our team comprises of leading researchers from top institutions in Slovenia, ensuring that the project is backed by a abundance of knowledge and expertise. As a result, we are confident that the project will succeed and believe that it is an excellent candidate for financing.