IntroductionEar-Phone ArchitectureSystem ComponentsMobile Phone ComponentsSignal Processing ModuleCentral Server ComponentsComputing Long-term Equivalent Noise Level, LAeq,LT GPS, MGRS conversionsSignal Reconstruction ModuleImplementation and EvaluationSystem ImplementationMeasurement AccuracyResource UsagePower BenchmarksMemory and CPU BenchmarksPerformance EvaluationSimulationSimulation DesignSimulating Sensing StrategiesPerformance EvaluationRelated WorkConclusions and DiscussionAcknowledgementsReferencesEar-Phone: An End-to-End Participatory Urban NoiseMapping SystemRajib Kumar Rana†‡Chun Tung Chou†Salil S. Kanhere†Nirupama Bulusu�Wen Hu‡†School of Computer Science and Engineering, University of New South Wales, Sydney, Australia�Department of Computer Science, Portland State University, USA‡CSIRO ICT Centre Australia{rajibr,ctchou,salilk}@cse.unsw.edu.au, [email protected], [email protected] noise map facilitates monitoring of environmental noisep ollution in urban areas. It can raise citizen awarenessof noise pollution levels, and aid in the development ofmitigation strategies to cope with the adverse effects.However, state-of-the-art techniques for rendering noisemaps in urban areas are expensive and rarely updated(months or even years), as they rely on population andtraffic models rather than on real data. Participatoryurban sensing can be leveraged to create an open and in-expe nsive platform for rendering up-to-date noise maps.In this paper, we present the design, implementationand performance evaluation of an end-to-end partici-patory urban noise mapping system called Ear-Phone.Ear-Phone, for the first time, leverages CompressiveSensing to address the fundamental problem of recover-ing the noise map from incomplete and random samplesobtained by crowdsourcing data collection. Ear-Phone,implemented on Nokia N95 and HP iPAQ mobile de-vices, also addresses the challenge of collecting accuratenoise pollution readings at a mobile device. Extensivesimulations and outdo or experiments demonstrate thatEar-Phone is a feasible platform to assess noise pollu-tion, incurring reasonable system resource consumptionat mobile devices and providing high reconstruction ac-curacy of the noise map.Categories and Subject DescriptorsC.m [Computer Systems Organization]: Miscellaneous–Mobile Sensing SystemsGeneral TermsDesign, Experimentation, PerformancePermission to make digital or hard copies of all or part of this work forpersonal or classroom use is granted without fee provided that copies arenot made or distributed for profit or commercial advantage and that copiesbear this notice and the full citation on the first page. To copy otherwise, torepublish, to post on servers or to redistribute to lists, requires prior specificpermission and/or a fee.IPSN’10, April 12–16, 2010, Stockholm, Sweden.Copyright 2010 ACM 978-1-60558-955-8/10/04 ...$5.00.KeywordsNoise Pollution, Mobile Phones, Compressive Sensing,Participatory Sensing1. INTRODUCTIONAt present, a large number of people around the worldare exp osed to high levels of noise pollution, which cancause serious illnesses ranging from hearing impairmentto negatively influencing productivity and social behav-ior [12]. As an abatement strategy, a number of coun-tries, such as the United Kingdom [9] and Germany [10],have started monitoring noise pollution. They typicallyuse a noise map (a visual representation of the noiselevel of an area) to assess noise pollution levels. Thenoise map is computed using simulations based on in-puts such as traffic flow data, road or rail type, andvehicle type. Since the collection of such input data isvery expensive, these maps can be updated only aftera long period of time (e.g. 5 years for UK [9]). Toalleviate this problem, a recent study [20] proposes thedeployment of wireless sensor networks to monitor noisep ollution. Wireless sensor networks can certainly elimi-nate the requirements of sending acoustic engineers fortaking real measurements, but the deployment cost ofa dedicated sensor network in a large urban space willalso be prohibitively exp e nsive.In this paper, we instead propose an urban sensingapproach (also known in the literature as participatorysensing [6], people-centric sensing [11] or communitysensing [15]) for monitoring environmental noise, espe-cially roadside ambient noise. T he key idea in partici-patory sensing is to “crowdsource” the collection of en-vironmental data in urban spaces to people, who carrysmart phones equipped with sensors and location-providingGlobal Positioning System (GPS) receivers. The visionof participatory sensing is inspired by the success ofother online participatory systems, such as Wikipedia,online reputation systems, and human computation sys-tems such as the Google Image Labeler. Due to theubiquity of mobile phones, the proposed approach canoffer a large spatial-temporal sensing coverage at a smallcost. Therefore, a noise map based on participatorydata collection can be updated with a very small la-tency such as hours or days compared to months oryears, making information provided by such a noise mapsignificantly more current than that provided by tradi-tional approaches.It is non-trivial to build a noise pollution monitoringsystem based on mobile phones. Mobile phones are in-tended for c ommunication, rather than for acoustic sig-nal processing.1To be credible, noise p ollution data col-lected on mobile phones should be comparable in accu-racy to commercial sound level meters used to measurenoise pollution. Since a participatory noise monitoringsystem relies on volunteers contributing noise pollutionmeasurements, these measurements can only come fromthe place and time where the volunteers are present.Furthermore, volunteers may prioritize the use of themicrophone on their mobile phones for conversation. Orthey may choose to collect data only when the phonehas sufficient energy. Consequently, samples collectedfrom mobile phones are typically randomly distributedin space and time, and are incomplete. To develop auseful noise pollution monitoring application, we needto recover the noise map from random and incompletesamples obtained via crowdsourcing. In this paper, weaddress these challenges. Our main contributions are:1. We present the design and implementation of anend-to-end noise mapping system, called Ear-Phone,to generate the noise map of an area using partici-patory urban sensing. EarPhone consists of mobilephones
View Full Document
Unlocking...