This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.
The devices at the edge of the Internet are consumer centric mobile sensing and computing devices, such as smart phones, music players, and in-vehicle sensors. These devices will fuel the evolution of the Internet of Things as they feed sensor data to the Internet at a societal scale. In this paper, we will examine a category of applications that we term mobile crowd sensing, where individuals with sensing and computing devices collectively share data and extract information to measure and map phenomena of common interest. We will present a brief overview of existing mobile crowd sensing applications, explain their unique characteristics, illustrate various research challenges and discuss possible solutions. Finally we argue the major areas & architecture of Mobile Crowd Sensing.
The gather of the sensing devices & computing devices at the edge of the internet will result a embedded internet. Typically IoT objects embedded with sensors. Which are lack of communicating & communication capabilities. When consumer centric mobile sensing and computing devices are connected to the internet we consider them as edge devices in evolution of internet. Mobile Crowd Sensing comes when we consider community sensing. Mobile Crowd sensing means the integration of sensors that can be used for gathering materialistic or non-materialistic information. For that we use Mobile crowd sources. The research area depend on mobile crowd sensing is how much this are integrating from mobile crowd sources. A number of successful applications, such as
wikipedia, iReport, reCAPTCHA, Amazon Mechanical Turk (MTurk), and others can be categorize under the topic of crowd sourcing systems.
Using the mobile phone we can distribute the sensing. Processing & communication can be done through common sensors in smart phones. Such examples include Urban Sensing, MetroSense, Nokia Sensor Planet, and others. Mobile Crowd Sensing comes when we consider community sensing. Also mobile phone sensors are extend to the wide area domain such as healthcare, entertainment, social networking, gaming, transportation and citizen science. We can divide those sensing as two parts. Which are personal & community sensing. Mobile Crowd Sensing comes when we consider community sensing. Mobile Crowd sensing means the integration of sensors that can be used for gathering materialistic or non-materialistic information.
In this review paper I hope to discuss the major areas of mobile crowd sensing & application based on these areas. Also what is a architecture of MCS & how to integrate it. Then how I hope to expand my research in future & what I did so far. Next section I basically tend attention to what is a MCS & what are the challenge it face in real world.
Basically crowd sensing is a specific format of acquiring sensor data from multiple Smartphone. Crowd sensing is a capability by which a requestor can recruit Smartphone users to provide sensor data to be used towards a specific goal or as part of a social or technical experiment. Crowd-sensing is a form of networked wireless sensing. but it is different from networked. Because of two reasons. some sensing actions on the Smartphone may require human intervention. Crowd sensing is differ from crowd sourcing. crowd-sourcing system like the Amazon Mechanical Turk which permits requestors to task participants with human intelligence tasks like recognition or translation, while a crowd-sensing system enables requestors to task participants with acquiring processed sensor data. Crowd Sensing is differ from wireless sensing. Also crowd sensing supports privacy of sensor data.
2.1 Challenges of MCS
Mobile phone resources are limited. Sensor resources are also limited. But introducing new aspects instead of traditional sensor data this problem can be overcome to little bit.
Energy, Bandwidth & Computation Resources are limited. That is a one of major challenge. The set of devices that are collecting sensor data are highly dynamic in availability and capabilities. Due to this highly dynamic nature, modeling and predicting the energy, bandwidth requirements to accomplish a particular task is harder than traditional sensor networks. Second, when there are a large number of available devices with diverse sensing capabilities which identifying and scheduling sensing and communication tasks among them. GPS, WiFi and GSM are the techniques use to identify route. GPS consume battery rather than others. Solution of this problem is a low duty cycling to reduce energy consumption of high quality sensors. A mobile device can be sampling various sensors (e.g. GPS, accelerometer, air quality) on behalf of different applications.
Privacy, Security & Data Integrity is a another challenge of MCS. MCS applications collect sensor data individual. For example GPS can be used to collect the individual route information. It means his home, works & other route information. Using this sensor measurements can get the data from traffic in given city.
3 Major Research area of MCS
MCS applications has two application specific components. one on the device (for sensor data collection and propagation) and the second in the backend (or cloud) for the analysis of the sensor data to drive the MCS application. The architecture of the MCS is shown in figure. There is a no common component. But each application faces a
number of common challenges in data collection, resource allocation and energy conservation.
Using these areas there are different ways to deploy the MCS applications. It's hard to program an application. To write a new application, the developer has to address challenges in energy, privacy, and data quality in an ad hoc manner. He has to develop different analytics to run the application based on different OS. But this approach is inefficient. Applications
performing sensing and processing activities
independently without understanding the consequences on each other will result in low efficiency. There may be a duplicate processing among multiple applications. Traffic sensing, air and noise pollution all require location information.
Figure Typical functioning of MCS applications.
4 Mobile Crowd Sensing Application
In this section I hope to discuss existing mobile crowd sensing applications & what are challenges they face. We can categorized MCS application based on the area they are measured. Those are,
MCS environmental application based on the natural environmental. Such as measuring pollution levels in a city, water levels in creeks, and monitoring wildlife habitats. Such application use to measure large scale environmental phenomena based on the common man. An example prototype deployment for pollution monitoring is common sense. Common sense use specialized air quality sensing devices that communicate with mobile phones (using
Bluetooth) to measure various air pollutants (e.g. CO2, NOx). These measurements are done by using large area. Also someone can use Microphone on mobile to measure noise level.
Infrastructure applications involve the measurement of large scale phenomena related to public infrastructure. Examples include measuring traffic congestion, road conditions, parking availability, outages of public works (e.g. malfunctioning fire
hydrants, broken traffic lights). Early MCS applications measured traffic congestion levels in cities, examples of which include MIT's CarTel  and Microsoft Research's Nericell . CarTel utilizes specialized devices installed in cars to measure the location and speed of cars and transmit the measured values using public WiFi hotspots to a central server. This central server can provide information about traffic & route delay or not. Another example is ParkNet , an application that detects available parking spots in cities using ultrasonic sensing devices installed on cars combined with smart phones.
social application share sense information themselves. As an example people can share their works data (e.g. day today works) and compare their works levels with the rest of the community. So they can use this comparison to improve daily works routine. Just like BikeNet  and DietSense . In BikeNet, measure location and bike route air quality(e.g. CO2 content on route, bumpiness of ride) and aggregate the data to obtain "most" bikeable routes. In DietSense, individuals take pictures of what they eat and share it within a community to compare their eating habits. So this help to watch people what are the diabetics people eat & prevent that disease. To summarize the MCS application Figure help to identify the components which are associated.
Figure- Existing MCS applications take an "application silo" approach
5 Future Enhancement
Smart phones technology is growing rapidly. Also the amount of people who are using smart phones is growing. So Phone application developers have to think about the application they make. Because It must be attractive & efficiency. Then The applications have market.
Sensors of mobile phone act special role when we are talking about smart phones. Because without sensing application that smart phone is not consider as a smart phone. People use smart phones to easy their works. So sensing application are made based on this concepts. They address three areas. Which are environmental, Infrastructure & Social. But these applications have problem which are regarding resources they consume, Security & other sort of things.
So we must create the sensing application to address these areas & overcome to the problem now we faced. For that we must expand our research from Mobile Crowd Sensing To Urban Sensing & other sort of sensing.
In conclusion, we have identified a category of IoT applications that use to data collection from large number of mobile sensing devices such as smartphones, which we called mobile crowdsensing (MCS). We presented several MCS applications, such as CarTel, Nericell, ParkNet, BikeNet, and DietSense. Also we mentioned the major three areas application depends. Which are environmental, Infrastructure & Social. We then identified the unique characteristics of MCS & area of MCS & architecture. Also presented several research challenges of MCS and discussed their solutions briefly. Overcome these identified problem & challenges the sensors data can integrated in new way. Making sensors application which required minimum resources of mobile device.
Mobile crowd sensing is a path to go beyond the world. Because people can give message to the world through this. Making crowd sensing application they can do it. But those application must be pleasant to the environment. So I thought selecting this topic I can do something for social benefit. First I search about topic & I got those details. Then I thought this is a one I can learn new thing.
First I gathered information about this topic using research paper which address this topic. then I search what are the sensing data & sources. How we implement them in real world. Then I found Smart phones. Basically they are target smart phones. Then I found what are the applications regarding sensors data. What are the challenges currently mobile crowd sensing face & how to overcome them.
Then I wrote review paper using those details & I am searching more details regarding this topic.
First and foremost, I would like to thank my supervisor Mr. Shalinda Adikari for his consistent supervision, encouragement, cooperation and guidance he gave during the time I spent writing this review paper. Also, I am indebted to a number of individuals who, directly or indirectly, assisted in the preparation of the review paper.