The Internet of Things (IoT) paradigm is gradually becoming more prevalent through numerous devices and technologies, including sensors, actuators, microcontrollers, cloud-enabled services, and analytics. IoT objects gain intelligence by integrating with wireless sensor networks (WSNs), mobile computing and communication, and others. With sensors, smart things can be enabled by monitoring and identifying environmental changes related to motion, temperature, humidity, pressure, light, vibration, etc. To timely keep track of state changes, researchers are considering developing a cyber replicator, denoted as Digital Twin (DT), of real physical systems as a way to visualize, model, and work with complex cyber-physical systems (CPS). In this paper, we first refine the dataset to a format that can be easily used for deep learning (DL) experiments, IoT data pipeline development, data modeling and simulation, data aggregation, etc. We then demonstrate that DT data can be used to determine space occupancy based on the ambient light sensor, which tends to indicate occupancy in particular spaces because the building has smart lighting that will switch off when rooms are unoccupied after a certain time. Given the apparent developments in machine learning technology, it is clear that machine learning-based prediction has the ability to enhance resource utilization and further forecast future events. Particularly, we use a DT-based dataset and Long-Short-Term Memory (LSTM) neural network architecture to forecast the campus building’s internal temperature.

With the rise and growing attention on Digital Twins (DT) as a way to provide integration between the Internet of Things (IoT) and data analytics, so does the need to consider how to address its challenges. To deal with these challenges, Named Data Networking (NDN) can be a possible solution. NDN has been rising in popularity due to its advancements over the traditional TCP/IP Internet architecture. In this paper, our approach begins with the framework that leverages an NDN-based DT architecture for data management. We then design two scenarios that focus on the performance of data querying in a small and large-scale simulated NDN-based DT architecture. Based on the designed scenarios, we conduct the performance evaluation of data query and DT performance to investigate the performance gap and determine whether an action needs to be taken.

Machine learning, as a viable way of conducting data analytics, has been successfully applied to a number of areas. Nonetheless, the lack of sufficient data is one critical issue for applying machine learning in Industrial Internet of Things (IIoT) systems. Insufficient data raises could negatively affect the accuracy of machine learning models. To tackle this issue, we design a framework to systematically investigate the impacts of insufficient data on model training. This framework employs the generative adversarial network (GAN) and continuous learning to generate and engage new data in model training, enabling us to study the security risks of introducing new data in the model training process and develop countermeasures to mitigate these risks. To validate the efficacy of our framework, we consider a representative IIoT scenario, in which a variety of industrial components needs to be recognized by convolutional neural networks (CNNs), and design and implement three evaluation scenarios that are based on a real-world IIoT data set. Our experimental results confirm that insufficient data can have a significant impact on the model accuracy, but that new data generated by GAN and continuous learning can greatly improve the model accuracy. Our experimental results also show that the data poisoning threat posed by the GAN can significantly reduce the model accuracy. However, our proposed defensive mechanism is capable of securing the model learning process. We conclude this article by discussing some emerging issues that need to be addressed in future work.

Internet of Things (IoT) communications include an exceptional number of Machine-to-Machine (M2M) devices to enable automation in smart-world systems. Given the explosion in number of distributed computing devices, traffic modeling and forecasting (TMF) schemes, which imitate and predict device traffic dynamics in M2M communications, become critical to providing useful guidance for constrained network resource planning and scheduling. In this paper, we explore the vulnerability of TMF, particularly focusing on data integrity attacks. Specifically, we consider a generic attack with adversaries impersonating M2M networks to send fake triggering messages to detached M2M devices. Thus, the number of attached M2M devices can be manipulated. We additionally conduct threat modeling and investigate attack impacts on TMF. Our experimental results show that data integrity attacks are effective in disrupting TMF to reduce goodness-of-fit and prediction accuracy.

Long-term Evolution (LTE)/LTE-Advanced (LTE-A) can be a viable network infrastructure for Internet-of-Things (IoT) as it provides broadband wireless connections and wide-area coverage. However, effectively allocating limited spectrum resources remains a challenging problem due to the massive number of mobile and IoT smart devices competing for limited network resources. In this paper, we assess the performance of transmitting IoT traffic over LTE/LTE-A and discuss strategies for supporting distinct IoT-based systems. We design six cases by considering two data load patterns (i.e., small data load and large data load) from IoT devices and three transmission environments (i.e., free space, suburban, and urban). Through extensive simulation, we evaluate network performance with respect to bandwidth efficiency, throughput, packet loss ratio, and delay. Our experimental results confirm that a narrower bandwidth achieves higher bandwidth efficiency in all cases, but a narrow bandwidth could not maintain acceptable network performance when large data loads need to be transmitted over the network.

With an unparalleled number of Machine-to-Machine (M2M) devices being deployed to support a variety of smart-world systems powered by Internet of Things (IoT) technologies, the heterogeneity, uncertainty, and complexity of M2M communications have increased enormously. Thus, how to conduct network resource planning (NRP) has become a challenging issue. In this paper, we propose a novel time series framework (TSF) to model and forecast timescale network traffic dynamics in M2M communications that is capable of providing useful guidance for effective NRP. Specifically, our TSF utilizes the statistical techniques INGARCH(p,q) (integer valued generalized autoregressive conditional heteroskedasticity) and βARMA(p,q) (beta autoregressive moving average) to accurately capture both the internal and external impact factors of the asynchronous and synchronous M2M traffic dynamics over a large time scale, and produces forecasts for multiple upcoming time points by leveraging conditional maximum-likelihood estimators (CMLE). Through a combination of theoretical analysis and extensive simulation, we have validated the modeling and forecasting efficacy of TSF. Our experimental results demonstrate that TSF achieves superior performance with respect to goodness-of-fit and prediction accuracy.

Unmanned Aerial Vehicles (UAVs), also known as drones, have become a new paradigm to provide emergency wireless communication infrastructure when conventional base stations are damaged or unavailable. In this paper, we propose new schemes to enable the 3D deployment of drones, which can provide network coverage and connectivity services for users located in uneven terrain. We formalize two models, including optimal coverage model and optimal connectivity model, which belong to NP-hard. To be specific, we first consider both the quality of service (QoS) requirements of users and the capacity of drones. We then formalize the problem and design a heuristic scheme, called Particle Swarm Optimization (PSO) algorithm to achieve a cost-effective solution. We also address the optimal connectivity problem in a scenario, in which a number of isolated local networks have been established by users through ad hoc communication and/or device-to-device (D2D) communication. We further develop the cost-effective heuristic algorithm to effectively minimize the total number of required drones. Via extensive performance evaluation, our experimental results demonstrate that the proposed schemes can achieve the effective deployment of drones for users in uneven terrain with respect to the number of required drones.

In this paper, we address the issue of Android malware detection by implementing a deep learning environment and fine-tune parameters to determine optimal settings for the classification of Android malware from extracted permission data. By determining the optimal settings, we demonstrate the potential performance of a deep learning environment for Android malware detection. Specifically, we conduct an extensive study of various hyper-parameters to determine optimal configurations, and then carry out a performance evaluation on those configurations to compare and maximize detection accuracy in our target networks. Our results achieve approximately 95 % detection accuracy, with an approximate F1 score of 93 %.

Communication networking leverages emerging network technologies such as topology management schemes to satisfy the demand of exponentially increasing devices and associated network traffic. Particularly, without efficient topology management, Machine-to-Machine (M2M) communications will likely asymmetrically congest gateways and eNodeBs in 3rd Generation Partnership Project (3GPP) Long-Term Evolution (LTE) and Long-Term Evolution Advanced (LTE-A) networks, especially when M2M devices are massively deployed to support diverse applications. To address this issue, in this paper, we propose a 3D Topology Optimization (3D-TO) scheme to obtain the optimal placement of gateways and eNodeBs for M2M communications. By taking advantage of the fact that most M2M devices rarely move, 3D-TO can specify optimal gateway positions for each M2M application, which consists of multiple M2M devices. This is achieved through global optimization, based on the distances between gateways and M2M devices. Utilizing the optimization process, 3D-TO likewise determines optimal eNodeB positions for each M2M application, based on the distances between eNodeBs and optimal M2M gateways. Our experimental results demonstrate the effectiveness of our proposed 3D-TO scheme towards M2M communications, with regard to throughput, delay, path loss, and packet loss ratio.

Summary form only given. The complete presentation was not made available for publication as part of the conference proceedings. Smart systems such as smart grid, smart transportation, and smart city are typical applications supported by Internet-of-Things. Particularly, the smart grid is the integration of renewable energy resources, information and communication technologies into the electricity grid to achieve a reliable, cost efficient, sustainable, and environment-friendly power grid. To deal with uncertainties raised by components such as renewable energy resources in the smart grid systems, our research team at Towson University has developed a modeling and simulation framework to assess techniques, enabling the effectiveness of smart grid operations. With the development of smart sensors, smart vehicles, and vehicular communication technologies, the smart transportation system is considered to be the future of transportation critical infrastructure: improving traffic efficiency and safety. To this end, our research team has been working on the investigation and evaluation techniques that seek to mitigate traffic congestion and improve traffic efficiency. Furthermore, while smart city concept holds great promise of boosting living standards through effective management and utilization of scarce resources, unavailability of real-world datasets and ideal test environments to evaluate algorithms and models has slowed research progress. A group of our research team has been working on the reviews of major research endeavors and projects ongoing in the field. Our research group has been trying to integrate information communication technology with physical infrastructures for a more effective resource management geared towards improving city living standard.

While smart city concept holds great promise of boosting living standards through effective management and utilization of scarce resources in cities, the unavailability of realworld datasets and test environments to evaluate designed models and techniques have slowed research progress. In this paper, we review existing research endeavors and develop a tool for extracting real-time smart city related data. We also conduct some simulations and evaluations in smart energy, which will be an important application in smart cities.

In this paper, we propose a Relay Nodes Position and Selection approach to obtain an optimal set of relay nodes for mobile nodes, which has a large amount of streaming data to be delivered in real time. Based on the selected set of relay nodes, data between mobile nodes and the core network can be transmitted via multiple paths. In our approach, we first determine the number of relay nodes, above which the performance of network cannot be further improved. We then propose a Centerof-Gravity (CoG) mechanism to properly position all the available relay nodes based on the density of mobile nodes. Finally, an optimal subset of relay nodes are selected for each mobile node to minimize the overall distance between mobile nodes and relay nodes in the network. We implement our designed approach in MATLAB and conduct an intense emulation in Common Open Research Emulator (CORE) based on the results generated in MATLAB. The experimental data confirms that our approach can improve the network performance with respect to network capacity, delay, and energy consumption.

The Internet of Things (IoT) has emerged as the key networking paradigm for supporting the connectivity of massively distributed objects and numerous simultaneous applications. The constrained application protocol (CoAP) is designed to meet the requirements for IoT data transmission among constrained nodes. The lean design of CoAP enables it to additionally meet the needs of the data transmission in dynamic network environments. In this paper, we conduct an emulation-based quantitative performance assessment of CoAP in comparison with HTTP, assessing data transmission based on key characteristics of dynamic network environments and the designed scenarios. We also designed scenarios and evaluate the performance of investigated protocols using real-world IoT datasets. Our experimental results demonstrate that CoAP performs better than HTTP for data transmission in the dynamic network environments with respect to delivery rate, delay, and overhead. In addition, we analyze the impact of features in dynamic network environments on the performance of data transmission protocols with respect to success rate, delay and overhead, as well as discuss some further extensions for future research.

The usage of well-tried software design patterns and application frameworks is often encountered in Mission and Safety Critical Applications development due to the high stakes involved in the case of failures. To increase reliability, some frameworks attempt to separate the implementation of business logic and low level implementation details and move the latter inside of framework-implementation in order to allow the developers to focus as much as possible on the problem to be solved while providing the necessary infrastructure into easy to use API's. In this paper we present a framework for message processing which takes advantage of the newer C++11 features to enforce separation of concerns, perform dead-lock avoidance, and encourage unit testing.

Traffic efficiency and safety are major hallmarks of Intelligent Transportation Systems (ITS). To accurately validate and investigate the effectiveness of traffic efficiency and safety application of ITSs, realistic studies are highly demanded [1]. In this paper, using real-world traffic and simulation data, we developed a realistic ITS test bed and a mobile application known as the Incident Warning Application (IWA) with the view of answering the following question: what is the traffic efficiency and safety benefits of Vehicle-to-Infrastructure (V2I) communications in a realistic ITS environment? Our real-world dataset consists of six weeks road traffic data of the Maryland (MD)/Washington DC and Virginia (VA) areas from August 8th, 2012 to September 27th, 2012. Our evaluation data shows that vehicles running our IWA application show improvements in almost all of the performance metrics evaluated. Specifically, our data shows that improvements in travel time (139.89%), fuel consumption (11.77%), and environmental emissions - carbon dioxide [CO2] (11.77%), etc. can be achieved through V2I communication.

Both opportunistic routing and inter-flow network coding are useful mechanisms for improving the performance of wireless networks. Both of them exploit the broadcast nature of the wireless medium and the spatial diversity of multi-hop wireless networks. In this paper, we aim at incorporating interflow network coding into opportunistic routing for further improving the performance of wireless mesh networks (WMNs). The main issue in designing such a scheme is candidate set selection and prioritization based on a proper metric for opportunistic routing. To this end, in this paper, we first present a new metric to determine the prioritization of the forwarders in the set of candidates and then design an Inter-flow Network Coding-based Opportunistic Routing (INCOR) scheme using the defined metric. Our proposed INCOR scheme can integrate the characteristics of inter-flow network coding and opportunistic routing effectively to make full use of the broadcast nature of the wireless medium. We carry out extensive simulations to evaluate the effectiveness of the INCOR method. Our data shows that INCOR outperforms both opportunistic routing and inter-flow network coding schemes.

Our research group at Towson University has been working on the P-adic theory and its implementation. Based on the Chinese Remainder theorem and the Hensel code a new data type, called Multiple P-adic Data Type, has been established to realize rational calculation. With this data type all rational number operations are converted to integer calculation, and the fast integer multiplication of modern computer architectures can be fully taken advantage of. This data type can be significantly effective in the parallel and cloud computing environment due to its independent computation at each node during the calculation process. Experimental results are given to illustrate computational efficiency.

A P-adic Exact Scientific Computational Library (ESCL) for rational matrix operations has been developed over the past few years. The effort has been focusing on converting all rational number operations to integer calculation, and fully taking advantage of the fast integer multiplication of modern computer architectures. In this paper, we report our progress on parallel implementation of P-adic arithmetic by means of a multiple modulus rational system related to the Chinese remainder theorem. Experimental results are given to illustrate computational efficiency.

Network-based crime has been increasing in both extent and severity and network-based forensics encapsulates an essential part of legal surveillance. A key network forensics tool is trace back, which can be used to identify true sources of suspects. Both accuracy and secrecy are essential attributes of a successful forensic trace back. In this paper, we present a class of hopping based spread-spectrum techniques for forensic trace back, which fully use the benefits of the spread spectrum approach and preserves a greater degree of secrecy. Our proposed techniques, including Code Hopping-Direct Sequence Spread Spectrum (CHDSSS), Frequency Hopping-Direct Sequence Spread Spectrum (FH-DSSS), and Time Hopping-Spread Spectrum (TH-DSSS), operate to randomize the effects of marking traffic through both the time and frequency domains. Our simulation study validates these techniques in terms of accuracy and secrecy.

Independent Component Analysis (ICA) has found its application in face recognition successfully. The goals are to estimate the components from raw image data. These components are then used to extract features of face images on which face classification is conducted. The components play key role in face recognition system. However these separated components are not equally important in terms of contribution to the feature extraction. ICA components are un-ordered. We do not know which component is more valuable than others. In order to improve ICA performance it is highly desired to select most discriminative components that are most effective. It is of great significance for ICA face recognition to find methods for optimizing independent components (ICs). In this paper we explored two methods for this purpose. One is ICA Component Subspace Optimization, the other is Sequential Forward Floating Selection (SFFS).

In this paper, we proposed a joint spatial and temporal ICA method for face recognition, and compared the performances of different ICA approaches (spatiotemporal ICA and spatial ICA). In our study, two face datasets collected by AcSys FRS discovery system were used. One face dataset involves less variation in terms of face expression and head movement, while the other encompasses much more change. The experimental results led to the following conclusions: 1) the number of features affects the recognition rate; 2) spatiotemporal ICA outperforms spatial ICA in every scenarios; 3) the type of classifier is also a fact that affects the recognition rate. These findings justify the promise of spatiotemporal ICA for face recognition

Micro-motions, such as vibrations or rotations of an object or structures on the object, induce additional frequency modulations on returned radar signal, which generates sidebands about the object's Doppler frequency, called micro-Doppler by V.C. Chen et al. (2002). In this paper, we investigated statistical classification methods for target classification using their micro-Doppler signatures. At this stage only simulated data are studied, and two models are used to generate simulation data: point scatter model and RCS model. Both models are tested and compared for their performance on target classification

SAR (Synthetic Aperture Radar) can produce target images in range and cross-range with sufficient resolution for recognition. In this paper, we did an experimental test on three different feature extraction techniques (Principle Components Analysis PCA, Independent Components Analysis ICA, and Hu moments) by using different target SAR images taken from the MSTAR database. The performance of these techniques is analyzed. A number of classification techniques, such as Linear (LDC), Quadratic (QDC), K-nearest Neighbor (K-NN), and Support Vector Machine (SVM) are tested and compared for their performance on the target classification. Our experimental results provide a guideline for selecting feature extracting techniques and classifiers in automatic target recognition using SAR image data.