Urban Noise Computing for Event Discovery Prajakta Joglekar, Vrushali Kulkarni Department of Computer Engineering, Maharashtra Institute of Technology, Pune, Savitribai Phule Pune University, Maharashtra. somanprajakta@gmail.com, vrushali.kulkarni@mitpune.edu.in Abstract—Urban computing is a wide area for research, which has picked up momentum after the introduction of Smart City vision. Smart City initiative aims to improve the quality of life of its citizens by building a strong and resilient framework for the city. Noise pollution is one of the main problems affecting citizens all over. Citizen science, when combined with Mobile crowdsensing and Data mining can help researchers and town planners to collect urban noise data at a massive spatio-temporal scale, and analyse it to understand urban dynamics. This work is an effort to use urban noise computing to detect anomalies and interesting events happening in the city. The paper introduces the readers to the basics of Urban noise computing. It covers a comprehensive literature survey in the area of Urban noise com- puting and Anomaly detection. This work also includes problem statement, problem modelling and partial implementation details. Index Terms—Anomaly detection, Event discovery, Mobile crowdsensing, Smart city, Urban noise computing. I. I NTRODUCTION Urban computing is an interesting, multi-disciplinary area of research which has gained momentum with the advent of ’Smart Cities’ concept. In India, around 100 cities have been selected for development under this initiative. Efforts are being taken to improve the quality of life of the citizens by providing smart and sustainable solutions. The pillars of a Smart City include transport, healthcare, environment, economy, governance and citizens. Urbanization has given rise to job opportunities and higher educational facilities. At the same time, it is facing many challenges such as global warming, pollution, increased energy consumption, traffic congestion and health issues related to sedentary lifestyle of people. Urban noise problem or Noise pollution is one such challenge which affects all the pillars of a smart city directly or indirectly. Noise pollution is one of the six major types of pollutions. It is also one of the leading causes of ailments in humans as well as the ecosystem. People suffer from high blood pressure, hearing loss, stress and psychological disorders due to continuous exposure to high noise levels [1]. The World Health organization (WHO) has provided the guidelines of community noise citing different health problems caused due to different types of noise [2]. It is necessary for the citizens and the government to measure urban noise, understand dif- ferent causes that produce noise, model the noise situation in the city and come up with abatement measures. According to WHO, noise measurements may also be done to identify emerging trends and patterns in the city. We hypothesize that the study of urban noise can help us to understand the urban dynamics and discover interesting events in the city. Our research work is an effort to support this hypothesis. The paper is organized as follows: It first states the motivation behind this work. Concepts related noise pollution are explained. The paper presents a comprehensive Literature Survey in the area of Urban Noise Computing as well as Event Detection in a city. It states the observations and gap analysis and formulates the problem statement. The proposed system is expressed using set theory and data analysis. A suitable architecture is proposed for such an Event Discovery system. The report also gives current implementation status with partial results and concludes by stating the future work. II. MOTIVATION Our city, Pune, has been selected as one of the cities under ’Smart Cities’ mission by the Indian Government. Lot of efforts are being put in the direction of transportation, air pollution problems, garbage management, developing cycle tracks etc. to build a resilient environment[3]. Unfortunately, there is no sound plan in place to curb noise pollution or record ambient noise levels on regular basis. As per the latest Environmental Status Report submitted by Pune Mu- nicipal Corporation [4], noise levels in industrial, commercial, residential as well as silence zones are said to exceed the respective threshold values set by the Noise Pollution Control Body. This has motivated us to select ”Urban Noise” as an area of research. Further, involvement of citizens is very important in urban computing as they are the generators of myriads of data being generated every day. We can exploit the capabilities of ’human as a sensor’ to capture urban noise levels using smart phone. This has motivated us to use ’Mobile Crowdsensing’ paradigm in our work. Finally, we believe that we can discover interesting events in the city by modelling the urban noise data. Much work has been done to detect anomalies and festive events using GPS traces from taxi data, SMS and mobile call records, tweets and USA’s 311 complaint data [5, 6, 7]. To the best of our knowledge, no author has yet used the urban noise levels to detect interesting events and anomalies in an urban region. Sixth Post Graduate Conference for Computer Engineering (cPGCON 2017) Procedia International Journal on Emerging Trends in Technology (IJETT) IJETT | ISSN: 2455-0124( E) | 2350-0808 (P) | (IF: 0.456) | Volume 4 | Special Issue July - 2017 | 8224 Peer-review under responsibility of organizing committee of cPGCON 2017 © IJETT www.ijett.in