PhD Studentship in Computer Science: Urban Noise Mapping using Artificial Intelligence: Combining Propagation Models and Empirically-derived Machine Learning Predictions

Overview

Noise pollution has long been an issue in cities and is linked to a multitude of health issues. Attempts to map noise pollution have largely been based on annual traffic flow data, primarily from roads. However, city noise is more complex than just traffic noise and more realistic maps may be obtained by including all sources of noise within the urban environment. This may be achieved by using actual recorded sound data, either from static sensors deployed along roads, or by using mobile surveys.

Building on Alvares-Sanches et al. (2021), this project will investigate a hybrid approach to modelling urban noise pollution by coupling propagation models and empirically measured sound levels from mobile surveys. By recording sound across the full frequency spectrum, the student will be able to segment the data into appropriate frequency bands and also develop deep learning models to identify sound sources (e.g. traffic, bird song, speech). This approach will provide far more detail than the usual A-weighted noise summaries. The key research question is whether combining the strengths of propagation models and mobile surveys recording full-spectrum sound data leads to better, and more transferable, urban noise models.

Number Of Awards

1

Start Date

September 2024

Award Duration

3.5 years

Sponsor

School of Computing

Supervisors

Dr Tatiana Alvares-Sanches, Professor Philip James and Professor Jeff Neasham

Eligibility Criteria

You must have, or expect to gain, a minimum 2:1 Honours degree in Computing, Engineering, quantitative Geography/Environmental Science, or a related discipline with a strong numerical component. For this role, the candidate will need to carry out field data collection, as well as coding, spatial data modelling and signal processing. An interest in environmental pollution and impacts on human health and biodiversity would be an advantage. Ability to think and work independently, and strong communication skills, are essential. 

The studentship covers home fees (UK and EU applicants with pre-settled/settled status and meet the residency criteria). International applicants will be required to cover the difference between Home and International fees.  

Applicants whose first language is not English require an IELTS score of 6.5 overall with a minimum of 5.5 in each subsection.  

International applicants may require an ATAS (Academic Technology Approval Scheme) clearance certificate prior to obtaining their visa and to study on this programme. 

How To Apply

Use Apply to Newcastle Portal 

Once registered select ‘Create a Postgraduate Application’. 

Use ‘Course Search’ to identify your programme of study: 

• Search for the ‘Course Title’ using programme code: 8050F 
• Research Area: Computing Science  
• Select PhD Computer Science as programme of study 

 You need to provide the following in ‘Further Details’ section: 

• ‘Personal Statement’ - upload a document/written statement directly into application form 
• When prompted - select ‘Write Proposal’. Type the title of the research project from this advert. You don’t need to upload a research proposal. 
• Degree transcripts and certificates and, if English is not your first language, a copy of your English language qualification if already completed. 
• Studentship code COMP2159 in the ‘Studentship/Partnership Reference’ field 

In the ‘Supporting Documentation’ section upload:   

• your CV 

Contact Details

Tatiana Alveres-Sanches

Advert information