IoT-Based Smart City Development using Big Data Analytical Approach M. Mazhar Rathore, Student Member, IEEE, Awais Ahmad, and Anand Paul, Member, IEEE 1 Abstract— To meet the needs of urban public and the city development smartly, the use of IoT devices, such as sensors, actuators, and smartphones, etc., and the smart system is the very fast and valuable source. However, interconnecting thousands of IoT devices while communicating with each other over the Internet to establish a smart system, results in the generation of huge amount of data, termed as Big Data. To integrate IoT services in order to get real-time city data and then processing such big amount of data in an efficient way aimed at establishing smart city is a challenging task. Therefore, in this paper, we proposed and developed a smart city system based on IoT using Big Data Analytics. We use sensors deployment including smart home sensors, vehicular networking, weather and water sensors, smart parking sensors, surveillance objects, etc. The complete architecture and implementation model is proposed, which is implemented using Hadoop ecosystem in a real environment. The system implementation consists of various steps that start from data generation and collecting, aggregating, filtration, classification, preprocessing, computing and finished at decision making. The efficiency in Big Data processing is achieved using spark over Hadoop. The system is practically implemented by taken smart systems as city data source to develop smart city. The evaluation results show that the proposed system is scalable and efficient. Index Terms—Smart City, Big Data, Hadoop. D I. INTRODUCTION UE TO the incorporation of ubiquitous and pervasive computing, the trend of living is now changing.one of the report notified that in 2050, seventy percent of the world population will live in cities [1]. Hence, a rapid increase has been seen in the transition of the population towards cities. Therefore, for urbanization, it is an utmost important factor to comprehend the demand for service profiling to enhance the efficiency and may bring the recent advancement in the city management. Presently, few organization are on the way with their IoT platforms for live monitoring, planning and gathering urban process parameters. For instance, Japan’s broadband access is providing the facility of communication between This study was supported by the Brain Korea 21 Plus project (SW Human Resource Development Program for Supporting Smart Life) funded by Ministry of Education, School of Computer Science and Engineering, Kyungpook National University, Korea (21A20131600005). M. Mazhar Rathore is with Kyungpook National University, Daegu, Korea (e-mail: rathoremazhar@gmail.com). Awais Ahmad is with Kyungpook National University, Daegu, Korea. (email: awais.ahmad@live.com). Anand Paul is Associate Professor in School of Computer Science and Engineering in Kyungpook National University, Daegu, Korea. (e-mail: paul.editor@gmail.com). 9781509011476/16/$31.00©2016IEE people, people and things, and things and things [2]. Similarly, S. Korea’s smart home enables their people to access things remotely [3]. Singapore next generation I-Hub [4] intentions to comprehend the next generation “U” type network through a secure and ubiquitous network [5]. Hence, The use of IoT for smart systems results in enhancing the number of things to be interconnected with each other, which results in the overwhelming amount of the heterogeneous data, referred to big data. Analyzing such data based on the user needs and choices, the cities would become even smarter. However, the discussed system works at a limited level, without considering the importance of Big Data generation and handling. Such activities should be followed by the amount of data collected, offline and real-time big data processing and analysis, and decision making. Usually, data collection and analysis techniques are difficult to achieve in such environment. Therefore, there is a need to incorporate smart technology that could efficiently collect the data, performed analysis, take real-time decisions and predict the future for better city planning and development. Having understood the feasibility and potential of the IoT and the smart systems, in this paper, we propel the concept of smart systems toward the smart city development based on big data analytics. In the paper, we proposed the complete architecture to the developed smart city and did urban planning using IoT-based Big Data analytics. The system provides the guidance to all the government authorities to make their cities smarter and intelligent in order to take a decision at real-time based on current city scenarios. The complete system description is depicted in next sections. II. MOTIVATION Urban planning and development smart city applications have major impact area on the life of citizens [1]. This includes the effect on the citizen in terms of health and safety, disaster management, pollution control, and so on so forth. Different projects related to monitoring of cyclist, cars, public car parking, etc. are undergoing that utilizes sensors services for the collection of specific collection of data. Apparently, different other service domain applications are identified that utilizes smart city IoT infrastructure to provision operations in air, noise, pollution, and surveillance system in the cities. However, in the development of any metropolitan, the transport system has a key role. Even a country can only progress rapidly if his transportation system and the facilities of transportation to the citizen are remarkable. A good transportation system makes the task to be performed promptly. The smarter transportation has a lot of other benefits, such as reduction of pollution, facilitation of citizens, rapid development, economy improvement, and many more. The recent research consists of a very few research findings in the field of smart traffic as well as in the smart city. Moreover, in this electronic era, where billions of devices are connected on the internet generate millions of terabytes of Big Data. The big data generated by the various IoT systems is used to analyze different aspects of smart city. A similar concept is followed using the IoT paradigm and the big data concepts for urban planning. Thus, to analyze such amount of data and make intelligent decisions is a major challenge. Therefore, based on the need of the citizens and authorities, we focus on the building of city the more smartly by providing real-time information regarding the city for the citizens as well authorities. Therefore, the rationale behind our intentions is to enrich the vast deployment of ICT resources in developing the entire system. For this very reason, we come to know that the advancement of recent technology in the embedded system depicts the trends of ICT. Therefore, a system is required that could inhale all of the recent developments in the field of smart cities, due to which a remarkable growth can be seen in a near future. The design of this system requires all the capabilities of sensing the environment and analyzing the sensing information. Therefore, various real-time action could be welcomed due to these technological resources. Moreover, it can be seen that integrating a large amount of data to perform the efficient analysis are already performed at their best. However, with large scale environment, it is unavoidable that the huge portion of data is left disjoint. As a result, such data cannot provide us a better understanding of the situation so that we may plan for future. III. BACKGROUND AND RELATED WORK Big data and its analysis are at the verge of modern science and business, where author highlights the identity of number of sources on big data such as online transactions, emails, audios, videos, search queries, health records, social networking interactions, images, click-streams, logs, posts, search queries, health records, social networking interactions, mobile phones and applications, scientific equipment, and sensors [6]. The proposed model is using conventional database tools. The challenge is to capture, form, store, manage, share, analyze, and visualize the big data. Various studies have been performed on smart cities building in a limited way [7] and statistical features of the big data as well [8-12]. In the majority of the cases, we do not rely on their statistical measurements. With a view to manifest the statistical measurements of the big data sets, there is a need to represent them in some other forms, i.e., how to characterize big data sets as the majority of the data processing tasks rely on some appropriate data representation. For instance, in case of image processing scenario, the wavelet transform [13], in case of remote sensing scenario, multi-resolution representation, such as image segmentation [14], image denoising [15], image restoration [16], image fusion [17], change detection [18], feature extraction [19], and image interpretation. Hence, assessment of statistical measurements of big data in the wavelet transform domain in real-time or offline is a key challenging area. The publicity big data is used for transformation and governance of cities with the help of data flood that provide much more sophisticated, wider scale understanding and control of urbanity. To the best of our knowledge, there is no such industrial definition of big data, but a survey of emerging trends signify a number of key features, such as volume indicates a lot of data to be aggregated for processing and analyzing while velocity refers to the high-speed processing and analysis (e.g., online streaming, real-time remote sensing, social websites data, e-health data, and so forth.). On the other hand, the term Variety refers to the vastly varied structures (e.g., Machine-to-Machine, Internet of Things, Wireless Sensor Network, and so forth). Interested readers are referred to [20] [21] [22] [23] [24]. In case of collaboration interactions within scientific communities through different knowledge fields, results in new products. Since the role of communities and networks as an essential condition of novelties is discussed in [1]. The IoT is expected to be a substantial support for the IT infrastructure of smart cities [25]. A scheme is also presented that deploys sensors in a cloud that provide IoT services are presented in [26]. This work discussed different commercial cloud-assisted remote sensing platforms and highlights their capabilities. The nature of sensors is to generate a massive volume of big data. Thus, the distribution of context to sensor data for originating intelligible and persuasive information plays a critical role [25]. Also, context-aware computing has proven to be successful in understanding sensory data. Various other works about context aware in term of IoT perspective is discussed in [27]. Due to the fact that the deployment of a maximum number of sensors along with numerous heterogeneous devices and systems, and due to unreliable nature of the majority of objects, the Quality of Service could be reduced. To solve such problem, the cognitive management framework for IoT is proposed [28]. In this scheme, intelligent and autonomous implementation of different application are enabled by mapping cognition and proximity with related objects for the application. Due to the fact that urban planning and development applications can be benefited from a smart city IoT capabilities can be grouped into impact areas [1]. This includes the effect on the citizen in terms of health and safety, the transportation system in terms of mobility and pollution, and so on so forth. Different projects related to monitoring of cyclist, cars, public car parking, etc. are undergoing that utilizes sensors services for collection of specific collection of data. Apparently, different other service domain applications are identified that utilizes smart city IoT infrastructure to provision operations in air, noise, pollution, vehicle mobility, and surveillance system in the cities. The recent research consists of a very few research findings in the field of smart city as well as in urban areas. Similarly, a compact system is not yet built which is more scalable and efficient. The big data is used to analyze different aspects of the smart city and then uses the knowledge obtain from the past generated data for the betterment of cities. A similar concept is followed using the IoT paradigm and the big data concepts for urban planning. Thus, we tried to come up with a solution that is applicable in the smart city and as well as in the urban areas. The proposed system is implemented and tested on the Hadoop framework with a spark to get the real time effects in the case of real-time smart city decision. Moreover, Hadoop and MapReduce is used for large historical data for urban planning and future enhancements. placed on the front screen of the car. We get the IV. EASE OF USE IOT-BASED SMART CITY A. System Overview The primary concept of the smart city is to get right information at the right place on the right device at the right time to make the city related decision with easiness and to facilitate the citizens more quick and fast ways. To develop the IoT-based smart city, we deployed several wireless and wired sensors, surveillance cameras, emergency buttons in streets, and other fixed devices. The main challenge in this regard is to achieve smart city system and link smart system generated data at one place. We do this by placing the main data hub linking all smart system to have all smart system data at a central place. Figure 1 shows the sensors and smart system deployment in order to generate data using a central hub for building the smart city. In order to get Real-time city data, we proposed to deploy many sensors at different places to achieve smart homes, smart parking, weather and water systems, vehicular traffic, environment population and surveillance system. These systems are used by the authorities to make an intelligent decision based on the real-time data to establish the smart city. In a smart home, the home is continuously monitored by sending data generated from the sensors, e.g., the smoke and temperature to detect a fire at real-time, the electricity and gas consumption to effectively manage the power, gas, and water consumption to the houses and different areas of the city. The smart parking helps in managing the vehicles coming and going out of different car parking zones. Thus by smart car parking analysis, the need for new car parking at specific areas can be identified. In our system, the citizens easily get the information of the nearest free slot and suitable place for parking at real-time. Weather and water system provides the weather related data like temperature, rain, humidity, pressure, wind speed and water levels at rivers, lakes, dams, and other reservoirs. In the world, most of the flood occur due to the rain and similarly few by snow melting and dam breakage. Therefore, we use rain measuring sensors and snow melting parameters in order to predict the flood earlier. We also predict about the water reservoirs in advance in order to meet the need of the water to the citizens. Vehicular traffic information is the most significant source of a smart city. Through this type of data source and with useful real-time analysis the citizen and as well as government can get more benefits. In our smart city system, we are getting the traffic information by GPRS, vehicular sensors, as well as the sensors Fig. 1. Sensors Deployment location of each vehicle the number of vehicles between two pairs of sensors placed at the various location of the city. Moreover, if any accident happens, the front screen will be damage and the sensor will send the alert to the police, traffic authorities, and hospital. Moreover, a city can never be smart with unhealthy citizens. Therefore, while designing smart city, we put a separate module to get environment data which includes gases information such as particular metals, carbon monoxide sulfur dioxide, ozone, and noise as well as. The citizens are alerted when any of the toxic gas is more in the air. Last but not least, the most important thing for the people of the smart city is the security concerns. Security is achieved by the proposed system by continuous monitoring the video of the whole city. However, it is very difficult to analyze all city videos and detect any mishap with anyone at real time by the system. To overcome this limitation, we propose new scenarios, which increase the security of the system of the whole city. We put various emergency buttons including microphones at different places of the city with surveillances cameras. When any mishap happens with anyone like robbery, car stolen, etc. He can just push the emergency button at any near place, and it will send the message to the nearest police station etc. Thus, the police or security agencies can start monitoring the nearby locations through surveillance cameras and can quickly locate the imposter. The Analytical Architecture and Implementation Model Based on the needs of the smart city, we proposed an architecture and the implementation to analyze IoT-based smart systems to establish smart city by real-time analysis. The system complete architecture and implementation model is given in Figure 2. It shows the full details of all the steps performed from data generation to decision making and applications. Initially, every system will generate their data, such as smart home generated data, vehicular data, smart parking data, etc. At every system, there is relay node, collecting data from all sensors and then sending to the Intelligent City Building through GW, The Internet, and the central data-Hub. Intelligent City Building is the main analysis system that is responsible for all activities from data collection, filtration to Decision making. It contains various servers equipped with many advanced analytical algorithms at various levels. At data preprocessing level, the data is collected by fast network cards and driver, so that it cannot lose any packets. As the sensors have a lot of metadata, and the sensors also generate the heterogeneous type of data. Therefore, all the unnecessary metadata and the redundant data are discarded at filtration server. The classification server classifies the incoming data from the various system by the message type and the identifier. After classification, the classified data is converted to the organized form, i.e., understandable to the Hadoop ecosystem, such as sequence file or tabular for with parameters information. Since we are dealing with a large amount of data (termed as big data). Therefore, we need a system that could efficiently process a large size of real-time (smart city) as well as offline (urban planning) data. To meet these requirements, we used Hadoop ecosystem, which contains Master nodes, and various data nodes under the Master node. The Hadoop ecosystem has HDFS file storage, which divides the data into an equal amount of chunks and stored them in multiple data nodes. Later, the parallel processing is performed on these chunks using MapReduce system. Hadoop basically used for batch processing, However, in order to use it for real-time analytics, we used Spark over the Hadoop system. All the processing calculations, results generation are done at Hadoop ecosystem. Finally, the decision making is performed based on the results generated by Hadoop ecosystem. The decision-making approach uses machine learning, pattern recognition, soft computing and decision models. The generated results are used for many smart city activates such as shown as applications in Figure 2. B. System Implementation and Data Generation? We implement the whole system on Hadoop ecosystem taking it as the Intelligent City Building. All the smart systems are connected to the main system, which collects the real-time data. All the remaining activities are performed by the Hadoop ecosystem. For the city vehicular network data collection, we developed a hardware-based vehicular network with two vehicles and a base station, as shown in Figure 3, to generate real data such as location, speed, time, etc. The base station is attached to the system by USB port that transmits the vehicular information to the system. For the other smart systems, we used existing smart systems datasets, which we replayed from another computer system to make the real-time environment. The real-time traffic is processed by Hadooppcap-lib, Hadoop-pcap-scr-de libraries and converted into sequence file to make it capable of processing on Hadoop. Each facility of the smart city is implemented as a separated class or sub-module. Citizens have limited access to the results of these modules, and the government has full access to them. Fig. 2. System Architecture and Implementation Model. Fig. 3. The developed vehicular network As it is very difficult at this time to implement all the smart systems, therefore we take existing datasets of the mentioned smart systems from various reliable resources. The details of each dataset including the datasets size, the area information, description, the number of parameters, and the reference are given in Table 1. V. DATA ANALYSIS AND IMPACT ON SOCIETY The data analysis using the developed system. The analysis results can be used for many purposes and benefited the society from common citizen to businessman and the government authorities. Here, we are highlighting some of the impacts and benefits gains by the citizens and the government by the real data analysis. Figure 4 are the graphs generated to show the average speed of the vehicles on the road when the intensity of the vehicles is low and high respectively. Which shows the higher speed when the intensity of the traffic is low and lower average speed when then the intensity of the traffic is higher. Similarly, Figure 5 shows the intensity of the traffic between two points at a various time. Figure 6 is the graph generated by the system to show the estimated time to reach to the other point based on the average speed of the vehicles and the intensity of the vehicles. The system also checks out the over speed vehicles. The over speed area of the Madrid Highway with respect to a number of vehicles are shown in figure 7. From this analysis, A citizen can decide which is the suitable routes to reach the destination depending on the on the current traffic scenario. Moreover, the government can control traffic and make optimize plan at runtime to decrease the intensity of the traffic on the crowded road. Open the blocked road due to any incident at the run time. Similarly, the government can do many things from current traffic analysis. S# 1 2 3 4 5 Datasets Water Usage TABLE I DATASETS DETAILS Area Description Surrey, Canada Madrid traffic Madrid Vehicular Cologne, Mobility Traces Germany Parking lots Aarhus, Denmark pollution Aarhus, Denmark 6 City traffic 7 Weather 61263 houses meter reading Location, speed Mobility, 400 km2, 24 hours, 700 cars Eight parking lots Fig. 6. Est. time to reach the destination depending on the traffic intensity No. of attr. Ref. 11 [29] 5 5 [30] [31] 7 [32] 449 sensors, Ozone, 8 NO,NO2, particle matters, etc. Aarhus, Sensors b/w 2 points. 9 Denmark Avg.speed, time to reach. Aarhus, temperature, humidity, 7 Denmark rain, pressure, etc. [32] Fig. 7. Location of Speed Violation on Madrid highway [32] [32] Fig. 8. Free Spaces at various parking lots at different times Fig. 4. The speed of vehicles at low intensity of traffic between two points Fig. 9. Usage of various parking lots at different times Fig. 5. No. of vehicles b/w two sources and destination at various time Fig. 10. Water usage of various areas of Surrey City Fig. 11. Pollution Level at different time of the day The car parking analysis in the city are shown in Figure 8 and Figure 9 that describe the free and consumed slots of several parking lots on various timeslots. From such type of analysis results, the citizens are updated to select the best suitable parking lot near their location. He can save his fuel without manually searching the free car garage. In a nutshell, It makes the profit equilibrium between the low as well as high-profit merchants by diverting customer to free parking lots. Moreover, The government can make urban planning build more parking areas near the places where most of the people normally go. From smart home data, the government can provide a lot of services to the citizens and can also make smart real-time decisions at run time such as fire detection and control, energy management, etc. Here we are just taking one case of water management, which benefits the government regarding proper control and planning of water usage. Figure 10 shows the consumption of water for various calculated by the system from each home consumption. The developed system continuously monitors the environment for toxic gases such as ozone (O3), carbon monoxide, sulfur dioxide (SO2), nitrogen oxide, and particulate environment for toxic gases such as ozone (O3), carbon monoxide, sulfur dioxide (SO2), nitrogen oxide, and particulate matter. The monitoring graph for a short time, when some of the gases exceed, is shown in Figure 11. The system generates alerts when any of the gas exceeds the dangerous threshold limit. It alters the people especially the patients with the particular diseases snot to go out. The government can also do urban planning, i.e., plan for traffic, city and industrial building and shifting to other places based on the city pollution. VI. SYSTEM EVALUATION We implemented the system using several (4-5) Hadoop data nodes. However, for evaluation purpose, we take single node setup on UBUNTU 14.04 LTS coreTMi5 machine with 3.2 GHz processor and 4 GB memory. Since the system is based on big data analytics, therefore, the systems is evaluated with respect to efficiency and response time using offline traffic as well as real-time traffic. The processing time results with respect to offline datasets size and the throughput analysis result is shown in Figure 12 and Figure 13 respectively. It is evident in the graph that when the data size is increased the processing time proportionally increased, both Fig. 12. Processing time corresponding to the data size Fig. 13. Throughput corresponding to the data size data size and processing time are directly proportional to each other. Moreover, throughput is also directly proportional to data size because of the parallel processing nature of Hadoop system, which is the major achievement of the system. VII. CONCLUSION The smart city has a major impact on country’s economy. Strong and smart city system helps in taking a quick and intelligent decision. This paper focuses on the implementation of the smart city by the use of the IoT-based smart system. Various smart systems are used to get real-time city data to make a decision. The Hadoop ecosystem is used to process Big Data generated by all the smart systems deployed in the city. The System is practically implemented and tested on real data. In future we are planning the actual deployment of all Smart systems, testing the accuracy of the system, considering security issues. ACKNOWLEDGMENTS This research was supported by Kyungpook National University Bokhyeon Research Fund, 2015. This study was also supported by the Brain Korea 21 Plus project (SW Human Resource Development Program for Supporting Smart Life) funded by Ministry of Education, School of Computer Science and Engineering, Kyungpook National University, Korea (21A20131600005). This work was also supported by the IT R&D Program of MSIP/IITP. [10041145, SelfOrganized Software Platform (SoSp) for Welfare Devices]. REFERENCES [1] Jin, Jiong, Jayavardhana Gubbi, Slaven Marusic, and Marimuthu Palaniswami. "An information framework for creating a smart city through Internet of things." Internet of Things Journal, IEEE 1, no. 2 (2014): 112-121. [2] Srivastava, Lara.“Japan’s ubiquitous mobile information society”. info, vol. 6, no. 4, pp. 234-251, 2004. [3] Giroux, Sylvain, and Hélène Pigot. From Smart Homes to Smart Care: ICOST 2005, 3rd International Conference on Smart Homes and Health Telematics. Vol. 15. IOS Press, 2005. [4] Han, Sun Sheng. "Global city making in Singapore: a real estate perspective." Progress in Planning 64, no. 2 (2005): 69-175. [5] O'droma, Mairtin, and Ivan Ganchev. "The creation of a ubiquitous consumer wireless world through strategic ITU-T standardization." IEEE Communications Magazine 48, no. 10 (2010): 158-165. [6] Xia, Feng, Laurence T. Yang, Lizhe Wang, and Alexey Vinel. "Internet of things." International Journal of Communication Systems 25, no. 9 (2012): 1101. [7] Kyriazis, Dimosthenis, Theodora Varvarigou, Anna Rossi, Douglas White, and Joshua Cooper. "Sustainable smart city IoT applications: Heat and electricity management & Ecoconscious cruise control for public transportation." In World of Wireless, Mobile and Multimedia Networks (WoWMoM), 2013 IEEE 14th International Symposium and Workshops on a, pp. 1-5. IEEE, 2013. [8] Lu, Jianguo, and Dingding Li. "Bias correction in a small sample from big data." Knowledge and Data Engineering, IEEE Transactions on 25, no. 11 (2013): 2658-2663. [9] A. Kleiner, A. Talwalkar, P. Sarkar, and M. I. Jordan, “The big data bootstrap.'' in Proc. 29th Int. Conf. Mach. Learn. (ICML), 2012, pp. 1_8. [10] Li, Runze, Dennis KJ Lin, and Bing Li. "Statistical inference in massive data sets." Applied Stochastic Models in Business and Industry 29, no. 5 (2013): 399-409. [11] Cormode, Graham, and Minos Garofalakis. "Histograms and wavelets on probabilistic data." Knowledge and Data Engineering, IEEE Transactions on22, no. 8 (2010): 11421157 [12] Yang, Qiang, Yuqiang Chen, Gui-Rong Xue, Wenyuan Dai, and Yong Yu. "Heterogeneous transfer learning for image clustering via the social web." In Proceedings of the Joint Conference of the 47th Annual Meeting of the ACL and the 4th International Joint Conference on Natural Language Processing of the AFNLP: Volume 1-Volume 1, pp. 1-9. Association for Computational Linguistics, 2009 [13] Portilla, Javier, and Eero P. Simoncelli. "A parametric texture model based on joint statistics of complex wavelet coefficients." International Journal of Computer Vision 40, no. 1 (2000): 49-70. [14] Mallat, Stephane G. "A theory for multiresolution signal decomposition: the wavelet representation." Pattern Analysis and Machine Intelligence, IEEE Transactions on 11, no. 7 (1989): 674-693. [15] Shah, Vijay P., Nicolas H. Younan, Surya S. Durbha, and Roger L. King. "Feature identification via a combined ICA– wavelet method for image information mining." Geoscience and Remote Sensing Letters, IEEE 7, no. 1 (2010): 18-22. [16] Liu, Peng, Fang Huang, Guoqing Li, and Zhiwen Liu. "Remote-sensing image denoising using partial differential equations and auxiliary images as priors." Geoscience and Remote Sensing Letters, IEEE 9, no. 3 (2012): 358-362. [17] Liu, Peng, and Kie B. Eom. "Restoration of multispectral images by total variation with auxiliary image." Optics and Lasers in Engineering 51, no. 7 (2013): 873-882. [18] González-Audícana, María, José Luis Saleta, Raquel García Catalán, and Rafael García. "Fusion of multispectral and panchromatic images using improved IHS and PCA mergers based on wavelet decomposition." Geoscience and Remote Sensing, IEEE Transactions on 42, no. 6 (2004): 1291-1299. [19] Celik, Turgay, and Kai-Kuang Ma. "Unsupervised change detection for satellite images using dual-tree complex wavelet transform." Geoscience and Remote Sensing, IEEE Transactions on 48, no. 3 (2010): 1199-1210. [20] Bruce, Lori Mann, Cliff H. Koger, and Jiang Li. "Dimensionality reduction of hyperspectral data using discrete wavelet transform feature extraction." Geoscience and Remote Sensing, IEEE Transactions on 40, no. 10 (2002): 2331-2338. [21] Boyd, D., & Crawford, K. (2012). Critical questions for big data. Information, Communication and Society, 15(5), 662– 679. [22] Dodge, M., & Kitchin, R. (2005). Codes of life: Identification codes and the machine-readable world. Environment and Planning D: Society and Space, 23(6), 851–881. [23] Marz, N., & Warren, J. (2012). Big data: Principles and best practices of scalable realtime data systems. Manning: MEAP edition. [24] Mayer-Schonberger, V., & Cukier, K. (2013). Big data: A revolution that will change how we live, work and think. London: John Murray. [25] Srikanth, S. V., P. J. Pramod, K. P. Dileep, S. Tapas, Mahesh U. Patil, and Chandra Babu N. Sarat. "Design and implementation of a prototype smart PARKing (SPARK) system using wireless sensor networks." In Advanced Information Networking and Applications Workshops, 2009. WAINA'09. International Conference on, pp. 401-406. IEEE, 2009. [26] N. Komninos, “The architecture of intelligent cities: Integrating human, collective and artificial intelligence to enhance knowledge and innovation,” 2nd IET Int. Conf. on Intell. Environments, vol. 1, 2006, pp. 13-20. [7] S. Abdelwahab, B. Hamdaoui, M. Guizani, and A. Rayes, “Enabling smart cloud services through remote sensing: an Internet of everything,”IEEE Internet Things J., vol. 1, no. 3, pp. 276-288, 2014. [27] C. Perera, A. Zaslavsky, P. Christen, and D. Georgakopoulos, “Context aware computing for The Internet of Things: A Survey,” IEEE Commun. Surveys Tuts., vol. 16, no. 1, pp. 414-454, 1st Quart. 2014. [28] P. Vlacheas, R. Giaffreda, V. Stavroulaki, D. Kelaidonis, V. Foteinos, G. Poulios, P. Demestichas, A. Somov, A. R. Biswas, and K. Moessner, “Enabling smart cities through a cognitive management framework for the Internet of Things,” IEEE Commun. Mag., Jun. 2013. [29] http://data.surrey.ca/dataset/water-meters, accessed on June 30, 2015 [30] Vehicular Networks on Two Madrid HighwaysMarco Gramaglia, Oscar Trullols-Cruces, Diala Naboulsi, Marco Fiore, Maria Calderon, IEEE SECON 2014, 3 July, Singapo [31] S. Uppoor, O. Trullols-Cruces, M. Fiore, J.M. BarceloOrdinas, Generation and Analysis of a Large-scale Urban Vehicular Mobility Dataset, IEEE Transactions on Mobile Computing, Vol.13, No.5, May 2014 [32] Stefan Bischof, Athanasios Karapantelakis, Cosmin-Septimiu Nechifor, Amit Sheth, Alessandra Mileo and Payam Barnaghi, "Semantic Modeling of Smart City Data", Position Paper in W3C Workshop on the Web of Things: Enablers and services for an open Web of Devices, 25-26 June 2014, Berlin, Germany. Muhammad Mazhar Ullah Rathore received his Master’s degree in Computer and Communication Security from the National University of Sciences and Technology, Pakistan in 2012. Currently, he is pursuing his Ph.D. with Dr. Anand Paul at Kyungpook National University, Daegu, South Korea. His research includes Big Data Analytics, Network Traffic Analysis and Monitoring, Remote Sensing, IoT, Smart City, Urban Planning, Intrusion Detection, and Computer and Network Security. He is an IEEE student member. He is also a nominee of Best Project Award in 2015 IEEE Communications Society Student Competition for his project ‘‘IoT based Smart City’’. He is serving as a reviewer for various IEEE, ACM, Springer, and Elsevier journals Awais Ahmad received his BS (CS) from the University of Peshawar and Masters (Telecommunication and Networking) from Bahria University, Islamabad Pakistan in 2008 and 2010 respectively. During his Master’s research work he worked on energy efficient congestion control schemes in Mobile Wireless Sensor Networks (WSN). Currently, he is pursuing his Ph.D. at Kyungpook National University, Daegu South Korea. There he got research experience on Big Data Analytics, IoT/SIoT, 4G/5G, Machine-to- Machine Communication, and Wireless Sensor Network. He also received three prestigious awards: (i) IEEE Best Research Paper Award: International Workshop on Ubiquitous Sensor Systems (UWSS 2015), in conjunction with the Smart World Congress (SWC 2015), Beijing, China, (ii) Research Award from President of Bahria University Islamabad, Pakistan in 2011, (iii) Best paper nomination award in WCECS 2011 at UCLA, USA, and (iv) Best paper award in the 1st Symposium on CS&E, Moju Resort, Korea in 2013. Anand Paul, (SM’**) received the Ph.D. degree in Electrical Engineering from the National Cheng Kung University, Tainan, Taiwan, in 2010. He is currently working as an Associate Professor in the School of Computer Science and Engineering, Kyungpook National University, South Korea. He is a delegate representing South Korea for M2M focus group and for MPEG. His research interests include Algorithm and Architecture Reconfigurable Embedded Computing. He is IEEE Senior member and has guest edited various international journals and he is also part of Editorial Team for Journal of Platform Technology, ACM Applied Computing review and Cyber---Physical Systems. He serves as a reviewer for various IEEE /IET/Springer and Elsevier journals. He is the track chair for Smart human computer interaction in ACM SAC 2015, 2014. He was the recipient of the Outstanding International Student Scholarship award in 2004---2010, the Best Paper Award in National Computer Symposium and in 2009, and International Conference on Softcomputing and Network Security, India in 2015.