Pragmatic Internet of Everything (IOE) for Smart Cities: 360-Degree Perspective

In today’s world, when everything around us is getting smart, be it the “phone” or “television”, there is an utter need and high time for the cities to get smart to solve the major problems of mankind. It is a futuristic approach to alleviate obstacles caused due to day by day increasing population of India. It will help the government as well as the common people to fight daily life problems such as water scarcity, waste management, and lack of interconnectivity in a city that can cause serious problems like delayed responses to emergency situations. This paper mainly focuses on the salient features of the smart city “Dholera” in Gujarat, India. The city uses futuristic technologies like Artificial Intelligence (AI) and the Internet of Things(IoT) to automate city resources. It has AI-based smart grids, transportation, water and waste management. This paper explains why Cholera is a smart, intelligent and fast-responsive city and how it is a boon for the masses.

The development and success of various smart cities is contingent on the multiple models of expertise they employ and execute like a functional and smart infrastructure to handle traffic chaos, a sustainable water recycling system, a smart administration task manager, or an efficient waste management plant. The development and efficacy of various smart cities are directly dependent on the efficiency of multiple models it employees. Thus, the presented study aims to review, analyse and document the various models that perform mundane tasks “smartly”, on the basis of key criteria namely: efficacy on the task in hand, power and time consumption, human interaction, upfront cost and operational task. These outcomes are then collated, and assayed by the application of various mathematical and statistical models to determine their performance as compared to the pre-existing non-technical approach of pursuing the same. Finally, the purpose of this study is to present a complete analysis of the performance of various models of smart cities to comprehend the profitability overall and provide suitable points to improve the same.

The development and implementation of smart cities can be a complex and challenging task as it becomes difficult to implement the notional concepts of smart cities that do not directly cater to the on-site problems. This study aims to identify and classify the various obstacles occurring in the development, planning & budget discussion of smart cities. It will deliberate wide purview of problems including standard difficulties like budget development that includes both capital upfront cost or Capex and operational cost or Opex, geographical & climatic challenges endemic to the city of development, employment & administration challenges like government permits, vendor availabilities, and technology troubles from IoT perspective. This study aims to categorize various issues into major five gradations namely: technical purview, location endemic purview, administration or employment troubles, pecuniary or financial issues and miscellaneous issues. Moreover, this study also provides a heads-up and detailed recommended steps to avoid the identified problems but presents specific case studies to show the significance and application of recommended solutions in multiple smart cities around the globe. 

While the development of smart cities is not an alien concept and in the last few decades, an increasing amount of investment both in terms of time and money is done, the pecuniary or financial benefit of this investment is still much uncharted. Like any other investment made, the return on investment plans and models are essential for further development. This research aims to explore, comprehend, analyse and further develop the return on investment in smart cities and various adjunct IoT and IoE-based models. The presented study primarily analyses and explores various possible financial benefits by deliberating multiple case studies of various smart cities and IoT-based projects worldwide. Additionally, a complete and detailed framework for each assayed case study is presented to comprehend the concept of RoI or return on investments which are distinct for every analysed case study contingent on numerous factors like investment, operational cost, area of application, etc. The presented study also aims to provide a comprehensive and comparative study between various plans of RoI for numerous case studies of smart cities and IoT-based projects.

Deep learning is an extension of Artificial Intelligence (AI) or cognitive learning that is used to optimize performance via the application of neural networks. And, big data analytics includes managing a plethora of continuous streams of data while obtaining valuable insights from them. Deep learning and Big Data analytics have been implemented in various avenues to obtain real-time optimized results, like biomedical applications, Computer Vision, and enhancing results for Internet of Things applications. This study aims to provide a deep insight into the application, performance, and values provided by Deep learning and Big-data analytics in the various intricacies of smart cities, smart governance and workflows in the same. Firstly, we provide applications or areas of smart cities that create Big-data, then provide techniques and literature where Big-data analytics is used to handle the same. Then, we present the different computing infrastructures used for IoT big data analytics, which include cloud, fog, and edge computing. Finally, we provide insights into various Deep learning modules that are successfully implemented in smart cities.

The benefit of IoT devices is that they allow for automation; nevertheless, billions of connected devices connected with one another waste a substantial amount of energy. IoT systems will have difficulty in wide adoption if the energy requirements are not adequately managed. This study proposes a solution for IoT devices to regulate their energy consumption. Both hardware and software aspects are taken into consideration. Using a mobile computer or smartphone with Internet connectivity to interact with actual scenarios has grown more prevalent as technology has advanced over the years. An intelligent home automation system based on android applications has been developed to save electricity and human energy. This study aims to create comprehensive Energy optimization through intelligent home automation utilizing widely available mobile applications and Wi-Fi technologies. The devices are turned on and off using Wi-Fi. Intelligent home, in the area of electronics, automation is the most purposely misused term. Numerous technological revolutions have occurred as a result of this demand for automation. These were more essential than any other technologies due to their ease of use. These can be used in place of household current switches, resulting in sparks and, in rare instances, such as fires. A unique energy optimization system was developed to control household appliances while taking advantage of Wi-Fi benefits.

The main purpose of this application, in addition to boosting the vision of a smart city, is to reduce humankind's effort and resources while simultaneously enhancing the vision. The squashing of the dustbin will occur on a regular schedule. It will be possible to manage waste more efficiently when these sensible garbage bins are implemented on a large scale and replace our previously designed dumpster, as it eliminates the need for waste to be piled up on the roadside in the first place. When managing the trash containers, wireless sensor systems in networks (WSN) in connection with the IoT technologies are used. On the other hand, sensors are used to monitor the container contents in real-time, with results displayed on the website, and the sensed contents are then evaluated to determine the optimal container distribution. This allows for the processing of a variety of waste types depending on the needs of the customer. As a result of installing ultrasonic sensors in each bin, garbage levels are continuously checked. As a result of this notification, the bin will be cleared. 

Solar energy is becoming more and more incorporated into the global power grid. As a result, enhancing the accuracy of solar energy projections is crucial for effective power grid planning, control, and operations. A fast, accurate and advanced estimation method is desperately needed to prevent PV's detrimental consequences on electricity and energy networks. For the optimum integration of solar technology into existing power systems, which benefits both grids and station operators, accurate prediction of solar production is crucial. The purpose of this research is to test the effectiveness of the machine learning model for projecting PV solar output. Using ANN in this research, weather parameters with the Power Generation for the next day appear to have been predicted. The evaluation findings suggest that the models' accuracy is sufficient to be employed with existing works and their approaches. Machine learning was shown to be capable of accurately predicting power while removing the difficulties associated with predicted solar irradiance data in this study.

The majority of today's weather forecasting studies have been focused on complex physical models. These models are usually run on hundreds of nodes in a High-Performance Computing system, which consumes a lot more power. Despite the employment of these costly and complex tools, projections are frequently incorrect due to inaccurate beginning conditions, measurements or a lack of understanding of atmospheric dynamics. Furthermore, solving complex models like this often takes a long time. The Internet of Things has helped any field that deals with technology. Using an IoT device, a prototype based on a machine learning approach is proposed in this study with an efficient framework, and implementation of an automated weather prediction system based on Artificial Neural Network algorithms was designed and developed. This system includes a technologically advanced irrigation system for our convenience. Using ANN in this research, the weather for the next day appears to have been predicted. The evaluation findings suggest that the model’s accuracy is sufficient for existing works and their approaches. 

Web sensing devices capture and transmit data from the physical environment to a central place using rapid advances in software, hardware, and IoT technologies. Depending on the source, the overall count of web-connected devices is estimated to be between 50 and 100 billion by 2025. The amount of data released will increase as the population expands and technology improves, which is already happening. The Internet of Things (IoT) technology connects and interacts with the physical and virtual worlds. A gadget linked to the Internet is called an IoT. Intellectual data handling and investigation are required to construct smart IoT requests. This article gives knowledge about the Machine learning (ML) algorithms available for dealing with IoT data challenges, using smart cities as the primary use case. This article looks at common IoT diagnostic applications. This research compares and evaluates the predicted precision and understandability of supervised and unattended ML models. These technologies are briefly addressed in desktop, mobile, and cloud computing settings.