Smart Cities Performability, Cognition, & Security

Polar codes, recently adopted in 5G standard due to their excellent performance at a very low complexity compared to other competitive schemes in the literature, are deemed to be a strong candidate for the Internet of Things (IoT) applications as well due to meeting their requirements. However, since polar codes construction is naturally channel-dependent, there has recently been an increasing interest in addressing the challenge of making polar codes work in realistic fading environments as they do in a binary symmetric channel (BSC). Recent studies on polar codes for fading channels have mainly focused on constructing new specific polar codes suitable to particular fading channels. This results in a non-universal code structure, leading to continuous changes in the code structure based on the channel, which is not desirable in practice. To address this problem, we develop and propose a novel transceiver architecture which enables using the polar coding design of a binary input additive white Gaussian noise (BI-AWGN) channel for multipath fading channels without causing any change in the structure of the encoder and decoder sides. This is made possible by eliminating the channel fading effect so that a net AWGN channel can be seen at the input of a simple successive cancelation decoder (SCD). The novelty of the proposed solution lies in using a channel-based orthonormal transformation with optimal power allocation at the transmitter and another transformation at the receiver to make the net, effective channel seen by the SCD very similar to the AWGN. Simulation results show that the proposed design makes the bit error rate (BER) performance of polar codes over a frequency selective fading channel as same as that over an AWGN channel.

Although we have reached new levels in smart city installations and systems, efforts so far have focused on providing diverse sources of data to smart city services consumers while neglecting to provide ways to simplify making good use of them. In this context, one first step that will bring added value to smart cities is knowledge creation in smart cities through anomaly detection and data annotation, supported in both an automated and a crowdsourced manner. We present here LearningCity, our solution that has been validated over an existing smart city deployment in Santander, and the OrganiCity experimentation-as-a-service ecosystem. We discuss key challenges along with characteristic use cases, and report on our design and implementation, together with some preliminary results derived from combining large smart city datasets with machine learning.

Wireless local area networks (WLANs) are widely deployed for Internet-centric data applications. Due to their extensive norm in our day-to-day wireless-enabled life, WLANs are expected to play a vital role for Internet of Things (IoT). It is predicted that by 2020, about 50 billion devices (things) will be connected via IoT. Consequently, WLANs need major improvements in both throughput and efficiency for such a massive device deployment in applications like smart offices, smart train stations, and smart stadiums for smart cities in IoT. New technologies continue to be introduced for WLAN applications for this purpose. The IEEE 802.11ac standard is the currently implemented amendment by the IEEE 802.11 standard working group that promises data rates at gigabits per second. The main features of the IEEE 802.11ac standard are adopting increased bandwidth and higher order modulation than the previous standards, and multiple-input, multiple-output (MIMO) and multiuser MIMO transmission modes. These features are designed to improve the user experience. In addition to technologies that enhance the efficiency of the WLAN, the IEEE 802.11ax High Efficiency WLAN (HEW) standard is also investigating and evaluating advanced wireless technologies to utilize the existing spectrum more efficiently.

The next-generation dense WLAN, HEW is expected to confront ultradense user environments and radically new applications for smart cities. HEW is likely to provide four times higher network efficiency even in highly dense network deployments. However, the current WLAN itself faces huge challenge of efficient channel access due to its temporal-based MAC layer resource allocation (MAC-RA). WLAN uses a carrier sense multiple access with collision avoidance (CSMA/CA) procedure to access the channel resources, which is based on a binary exponential backoff (BEB) mechanism. In BEB, a random backoff value is generated from a contention window (CW) to obtain channel access. The CW size is doubled after every unsuccessful transmission and reset to its minimum value on successfully transmissions. However, this blindness when increasing and resetting the CW induces performance degradation. For a dense network, resetting the CW to its minimum size may result in more collisions and poor network performance. Likewise, for a small network environment, a blind increase in CW size may cause an unnecessarily long delay while accessing the channel. To satisfy the diverse requirements of dense WLANs, it is anticipated that prospective HEW will autonomously access the best channel resources with the assistance of sophisticated wireless channel condition inference in order to control channel collisions. Such intelligence is possible with the introduction of deep learning (DL) techniques in future WLANs.

The potential applications of DL to the MAC layer of IEEE 802.11 standards have already been progressively acknowledged due to their novel features for future communications. Their new features challenge conventional communications theories with more sophisticated artificial intelligence-based theories. DL has been extensively proposed for the MAC layer of WLANs in various research areas, such as deployment of cognitive radio and communications networks. Deep reinforcement learning (DRL) is one of the DL techniques that are motivated by the behaviorist sensibility and control philosophy, where a learner can achieve an objective by interacting with the environment. In this chapter, a DRL-based intelligent paradigm is developed for channel access in dense WLANs in smart cities.

One of the DRL models, Q-learning (QL), is used to propose an intelligent QL-based resource allocation (iQRA) mechanism for MAC-RA in dense wireless networks. iQRA exploits channel observation-based collision probability for network inference to dynamically and autonomously control the backoff parameters (such as backoff stages and CW values). The simulations performed in network simulator 3 (ns3) indicate that the proposed DRL-based iQRA paradigm learns diverse WLAN environments and optimizes its performance, compared to conventional non-intelligent MAC protocol, BEB. The performance of the proposed iQRA mechanism is evaluated in diverse WLAN network environments with throughput, channel access delay, and fairness as performance metrics.

Our cities face non-stop growth in population and infrastructures and require more energy every day. Energy management is the key success for the smart cities concept since electricity is one of the essential resources which has no alternatives. The basic role of the smart energy concept is to optimize the consumption and demand in a smart way in order to decrease the energy costs and increase efficiency. Among the variety of benefits, the smart energy concept mainly enhances the quality of life of the inhabitants of the cities as well as making the environment cleaner. One of the approaches for the smart energy concept is to develop prediction models using machine learning, ML algorithms in order to forecast energy demand, especially for daily and weekly periods. The upcoming chapter describes thoroughly what is behind the deep learning concept as a subset of ML and how neural networks can be applied for developing energy prediction models. A specialized version of the recurrent neural network (RNN), e.g., long short-term memory (LSTM), is described in detail. In addition, the chapter tries to answer the question as to why the LSTM is a state-of-the-art ML algorithm in time series modeling today. To this end, we introduce ANNdotNET, which provides a user-friendly ML framework with capability of importing data from the smart grids of a smart city. By design, the ANNdotNET is a cloud solution which can be connected by other Internet of Things, IoT, devices for data collecting, feeding, and providing efficient models to energy managers in a bigger smart city cloud solution. As an example, the chapter provides the evolution of daily and weekly energy demand models for Nicosia, the capital of Northern Cyprus. Currently, energy demand predictions for the city are not as efficient as expected. Therefore, the results of this chapter can be used as efficient alternatives for IoT-based energy prediction models in any smart city.

The editor has retracted this chapter [1] because a significant part of the text and Fig. 5.3 overlap with a previously published conference paper by Wagih et al. [2]. Akanksha Pal agrees with this retraction. Abhishek Singh Rathore has not responded to correspondence about this retraction.

Detecting road obstacles is a major challenge in autonomous vehicles for smart transportation and safe cities. They have always been a major problem, as they increase car accidents leading to mechanical/electrical problems, injuries, and even deaths. Several methods exist to detect these road abnormalities, allowing for an earlier control of the car dynamics. While these methods rely on different sensors and multiple viewpoints, the one applied in this work is based on edge detection within a single view acquired from a conventional digital camera. In addition, as the fractional calculus has shown good results in several engineering domains, this mathematical technique is deployed for the already developed edge detection techniques. Hence, the novelty in this chapter consists of detecting road abnormalities using fractional calculus techniques. Results show very good identification of road obstacles, especially concerning humps, bumps, and cushions. In addition, fractional methods show better results compared to conventional methods in terms of response time and edge sharpness.

Machine learning (ML)-based parameter estimation and classification have been receiving great attention in data modelling and processing. The Gaussian mixture model (GMM) is a probabilistic model that represents the presence of subpopulations, which works well with a parameter estimation strategy. In this chapter, maximum likelihood estimation based on expectation maximization is used for the parameter estimation approach; the estimated parameters are used for the training and testing of medical images for normality and abnormality. The mean and the covariance, considered to be the parameters, are used in GMM-based training for the classifier. Support vector machine (SVM), a discriminative classifier, and the GMM, a generative model classifier, are the two most popular techniques. The classification strategy performances of both classifiers have better proficiency than other classifiers. By combining the SVM and GMM, it is possible to improve classification because estimating the parameters through GMM has very limited features; hence, there is no need to use any feature reduction techniques. The features extracted were used for the training of the classifiers. The testing of medical images for normality was performed with respect to the features that were trained.

Increasing traffic, population, and public safety are major issues of cities. Many cities face social and environmental sustainability challenges such as pollution and environmental deterioration. One challenging application area of big data analytics and machine learning that has huge potential to enhance our lives is smart cities. Intelligent services connected with smart devices and sensors deployed all around the area can provide efficient and sustainable cities. This idea is challenged with the fact that the amount of data that needs to be processed has increased to a level that new algorithms and techniques are required. Besides, constantly changing nature of cities requires adaptable architectures and learning systems. In this chapter, we present some open source tools for handling huge amounts of data that can be used to create solutions for smart cities. Also, we discussed some open research issues and enabling technologies such as energy consumption models, heterogeneous networks, and security.

Biometrics suggests a smart solution to keep the city safe. Installing a biometrics app on a mobile device facilitates identity recognition and verification instantaneously. Current work explores an authentication algorithm to address requirements of such memory restricted apps. A potential portion of periocular region, known as lower central periocular region, is examined to attain unconstrained authentication coupled with benefits of reduced template size. A novel computationally efficient feature extraction approach is employed over the region of interest using an efficient variation of conventional local binary pattern. The technique computes texture patterns over a dominant bit-plane, alternative to employing entire intensity image itself. Construction of the dominant bit-plane prior to feature extraction significantly simplifies operations required for texture pattern computations. The proposed methodology is tested on benchmark UBIRISv2 database and periocular images retrieved from high and low resolution imaging devices. Experimental results show an attainment up to 99.5% authentication accuracy in an unconstrained environment.

The recent public disclosure of mass surveillance of electronic communication, involving powerful government authorities, has drawn the public’s attention to issues regarding Internet privacy. For almost a decade now, there have been several research efforts towards designing and deploying open-source, trustworthy, and reliable systems that ensure users’ anonymity and privacy. These systems operate by hiding the true network identity of communicating parties against eavesdropping adversaries. TOR, acronym for The Onion Router, is an example of such a system. Such systems relay the traffic of their users through an overlay of nodes that are called Onion Routers and are operated by volunteers distributed across the globe. Such systems have served well as anti-censorship and anti-surveillance tools. The implementation of TOR allows an individual to access the Dark Web, an area of the Internet said to be of a much larger magnitude than the Surface Web. The Dark Web has earned a connotation as a sort of immense black market, associated with terrorist groups, child pornography, human trafficking, sale of drugs, and conspiracies, and hacking Dark Web research has received significant national and international press coverage. However, to date, little or no research has been conducted on the illicit usage TOR usage by South Africans. To date, there has yet to be a study that characterizes the usage of a real deployed anonymity service. Observations and analysis obtained are presented by participating in the TOR network. The primary goal is to elicit TOR usage by South Africans. In particular, interest is focused in answering the following question:

In sampling the results, it reflects that the main usage of TOR in a South African Context is to access social media websites.

Encryption is a cryptographic technique to secure information that is being transmitted over the network. In the present era, most of the applications hosted on cloud are affected by enumerable number of threats. Any web application hosted on cloud has several challenges for data privacy and security. In this work, a mobile bill payment application has been designed and developed in Java programming language for the purpose of secure bill payment over the cloud. Google Cloud Platform, Google App Engine, is used for the deployment of mobile bill payment application. The proposed system encapsulates two phases: In the first phase, affine cipher encryption technique is used to encrypt the original information. Affine cipher technique is improved using dynamic key, which encapsulates geographic coordinates of the user and its results are compared with rail fence cipher technique. In the second phase, generated cipher text has been further translated into two Indian languages: character by character. The result of two-phase encryption makes the original data stronger and secure.