Ubiquitous Communications and Network Computing

Significant increase of crimes against women in recent years and the advent of smart phone and wearable technologies have accelerated the need for personal safety devices and applications. These systems can be used to summon for help during the emergency situations. While several mobile applications that sends emergency help requests are available they need to be manually activated by the victim. In most of the personal emergency situations the victim might not be in a position to reach out for the Smart phone for summoning help. In this research we address this issue by implementing a system that automatically senses certain personal emergency situations, that summons for help with minimal or no user intervention. Summoning of help gets triggered when the smartphone sensors senses an abnormal events such as unusual movement and voice. This system also profiles the spatial information using the crawled web data and provides the contextual information about the risks score of the location. By using sensors and context awareness our system summons for emergency help with minimal/no intervention by the user.

Urban flooding is a common occurrence these days due to many reasons. Providing timely and adequate help to the victims is challenging. Enlisting the citizens to help themselves using their smartphones to provide real-time status updates and ensure timely delivery of needed help is a winning proposition. This paper describes a novel crowd-sourcing approach to urban flood management addressing the inherent challenges using smartphone applications and services that can be deployed by a variety of entities – government agencies, NGOs, social networks, etc. It enables sharing of real time information on the status of flooding, rescue and relief requests and responses, etc. It also collects data from various smartphone sensors in the background which is analyzed and synthesized to track the location and movement of people and to assess the integrity of structures such as bridges. It can also be a valuable resource for future city planning.

Data aggregation process can extract relevant information from raw data obtained from various sources using certain mathematical functions. Aggregation reduces the transmission of redundant data. A protocol named DP_AODV is implemented in this paper. Aggregator nodes (cluster head) are identified using the positional information. Routes are established between these aggregator nodes using efficient routing techniques. Data is aggregated along the path to the destination conserving additional energy. The aggregation process involves averaging the data if it is within the threshold range, else, only the data part along with the positional information is appended to the payload. Size of the Data packet varies dynamically based on the number of nodes having co-related data at that particular instance. The common header occupies a substantial part of the packet. Avoiding multiple transmission of common part of the header saves energy.

There is a large difference between the power draw from the mains and the actual utilization of commercial devices like CPU, as some amount of the power is bound to be wasted. In this paper, the technique that can be implemented for measuring the difference between the actual power draw and the utilized power draw, the steps to reduce the amount of power draw and hence saving the energy costs is mentioned. TVAKSHAS is an energy analytical device that measures the efficiency and performance of CPU and consists of a real time power measurement circuit that reads the real time power drawn from the mains by the CPU and sends it to the wireless sensing and communicating device, i.e. sensor node, TelosB in our project. It also consists of a software based power measurement tool that measures the performance of the CPU in terms of Watts. Hence by analysis of these two data, the difference between power obtained and power utilized can be obtained on base station in the wireless sensor network.

The alliance with mobile device cloud computing technology promises new ways of developing business application. Using web and cloud technology, it is possible to transfer a small part of secured business data from cloud to the small storage mobile devices. However, it is challenging to keep secure data in offline mode when the web application is unable to connect to the cloud and sync those offline data at online mode. The paper discusses architecture of cloud-based mobile application. The proposed architecture helps us to develop cloud-based mobile application. Special consideration of low memory and network connectivity on the mobile device is taken into account. We present various challenges to design a cloud-based mobile application, store the data in a secured manner at offline mode and sync those at online mode. This would greatly improve enterprise productivity even when users are working offline.

Mobile commerce environment pertains to multiple transactions with the purpose of providing goods and services to customers. The customers are using wireless devices like smartphone, mobile phone and PDA to shop for a broad range of products and information related services. The success of commercial business models for mobile commerce environment depends on the method of providing value to the customers. Modeling the specific characteristics of each business participant is a challenging task in dynamic business environment. In this paper, we propose a formal description of business model by identifying the specific characteristics of mobile participant and illustrate with examples in a trading scenario. The proposed analytical concepts assist in developing innovative user friendly interfaces and also to model e-business domain.

Blind signature is an interesting cryptographic primitive which allows user to get signature on his document from signatory authority, without leaking any information. Blind signature is useful in many e-commerce applications where user’s anonymity is the main concern. Since the Zhang et al., was the first to propose the identity based blind signature, many schemes based on bilinear pairing have been proposed. But the computational cost of pairing operation on elliptic curve is around 20 times the point multiplication on an elliptic curve. In order to save the running time, we present a new Identity-Based Blind Signature (ID-BS) scheme whose security is based on elliptic curve discrete logarithm problem (ECDLP). Performance comparison shows that proposed scheme reduces the cost of computation. Security analysis shows that proposed scheme is secure against the adversary and achieves the property of blindness and Non-forgeabillity. At the end; we propose an e-cash payment system based on our ID-based blind signature scheme.

Context-aware computing is one of the key research issues in smart business applications, IoT paradigm and it is evident that it is successful in understanding the data sensed from trading environment. The success of trading depends on providing the proper personalized services to the trader based on his current context. The challenge here is to design and develop a context aware model which extracts meaningful information from raw data generated from underlying physical world. When there exists right information at right time it is possible to make effective decisions. This paper, proposes a C-IOB (Context-Information, Observation, Belief) hierarchical graphical structure model which signifies a causal relationship between low level context and high level context to reason about the customer and vendor context based beliefs. The proposed model provides accurate service with minimum computation and also reduces solution search space since the context information of a business entity is synthesized in the form of beliefs. Our research findings indicate the importance of the context based belief model in developing innovative Forex applications.

This paper considers classification of diverse traffic types in Internet of Things (IoT) based on importance of data rate, packet size and proposes a priority-based probabilistic packet scheduling strategy for efficient packet transmission. Reduction of peak resource usage, dynamic control of service rate corresponding to arrival rate and QoS buffer management are few main factors considered to develop this strategy. By calculating percentage of link bandwidth required for prioritized traffic in each cycle, we provide quality of service (QoS) to real time traffic in IoT and non-IoT applications. Different experiments including MPEG traffic traces and Poisson traffic are conducted to verify the proposed scheduler. Also, performance of scheduler for both IoT and Non-IoT applications is compared for different data rates. We observe that the proposed packet scheduler satisfies QoS requirements for both IoT and non-IoT traffic.

Big Data plays a major role in every field in recent days. Analyzing, storing and visualizing the varied and complex data collected are important tasks for which Big Data tools are used. Big Data handles data in a more efficient manner. So, Big Data is preferably used in enterprises, organizations, companies, business etc. Henceforth, there are various fields such as healthcare/medical, business, sports, education, stock market, web and entertainment etc., which use big data tools. The motive of this paper is to give an insight into different types of analytics that are used in various fields and also to give a detailed study of various organizations and the extent they use the analytics in their respective fields.

Internet of Things (IoT) paradigm is the interconnection of machines, intelligent devices and location aware analytics platforms that collectively enable us to have smart world around us. As the billions of already connected devices and newly added devices grow this network, IoT pose the most complex operational and information technology challenges to the way networks are designed and operated. With the emerging technologies like SDN, SD-WAN, NFV, IXP evolving into standards, researchers are proposing new communication platforms to deliver secure and scalable networks for Internet of Things (IoT). In this paper, we discuss major security challenges in IoT networks and present the notion of security architecture for IoT based on programmable and virtualization technologies SDN/NFV, explain the architectural choices and its applications for IoT. We review prior works in this area and discuss our future work to solve security and privacy challenges of heterogeneous systems and networks in IoT.

The traditional networks, with the control and data plane integrated into the same network devices, do not provide a global view of the network performance like degree of congestion, bandwidth utilization, etc. Software defined network (SDN) is an approach towards this problem which separates the control plane of the switch from its data plane and provide a centralized control plane so as to get a global view of the network performance and thus make decisions of how to regulate flows. In SDN, network monitoring can be achieved more efficiently than traditional networks using OpenFlow statistics. SDN controller can keep track of available bandwidth on each link and thus estimate end-to-end available bandwidth of a path simply by composing individual link bandwidths thus avoiding end-to-end probing. We have made two contributions in this paper: (i) proposed and validated a method to estimate end-to-end available bandwidth on any given path by composing link-wise available bandwidths (ii) proposed a method to measure link capacity using OpenFlow protocol. We compared our results to the ones obtained using the state-of-the-art bandwidth measurement tool, Yaz.

In SDN based networks, for network management such as monitoring, performance tuning, enforcing security, configurations, calculating QoS metrics etc. a certain fraction of traffic is responsible. It consists of packets for many network protocols such as DHCP, MLD, MDNS, NDP etc. Most of the time these packets are created and absorbed at midway switches. We refer to these as raw packets. Cumulative statistics of sent and received traffic is sent to the controller by OpenFlow compliant switches that includes these raw packets. Although, not part of the data traffic these packets get counted and leads to noise in the measured statistics and thus, hamper the accuracy of methods that depend on these statistics such as calculation of QoS metrics.

In this paper, we propose a method to estimate the fraction of the network traffic that consists of raw packets in Software Defined Networks. The number of raw packets transferred depends on the number of switches and hosts in the network and it is a periodic function of time. Through experiments on several network topologies, we have estimated a way to find a cap on the generated raw packets in the network, using spanning tree information about the topology.

In order to meet QoS demands from customers, currently, ISPs over-provision capacity. Networks need to continuously monitor performance metrics, such as bandwidth, packet loss etc., in order to quickly adapt forwarding rules in response to changes in the workload. The packet loss metric is also required by network administrators and ISPs to identify clusters in network that are vulnerable to congestion. However, the existing solutions either require special instrumentation of the network or impose significant measurement overhead.

Software-Defined Networking (SDN), an emerging paradigm in networking advocates separation of the data plane and the control plane, separating the network’s control logic from the underlying routers and switches, leaving a logically centralized software program to control the behavior of the entire network, and introducing network programmability. Further, OpenFlow allows to implement fine-grained Traffic Engineering (TE) and provides flexibility to determine and enforce end-to-end QoS parameters.

In this paper, we present an approach for monitoring and measuring online per-flow as well as per-port packet loss statistics in SDN. The controller polls all the switches of the network periodically for port and flow statistics via OpenFlow 1.3 multipart messages. The OpenFlow compliant switches send cumulative statistics of sent and received packets to the controller that includes raw packets (control, non-user generated packets responsible for network management); which, although not being part of the end-to-end data traffic, get counted and act as noise in the statistics. The proposed method takes into account the effect of raw packets and thus, hamper the accuracy of methods.

Other implementations propose approaches for per-flow packet loss only. We also take into account the effect of raw packets (control, non-user generated packets) which makes our packet loss estimation more accurate than other implementations. We also present a study of extrapolation techniques for predicting packet loss within poll interval.

Mobile IPv6 is the widely acknowledged technology that supports mobility in networks. A single home agent in the network suffers from the issue of single point of failure and consequently focuses on the deployment of multiple home agents in the network. The load sharing mechanism in most of the exiting methods is passive and centralized in approach. Moreover, the failure detection and recovery mechanism uses the concept of periodic messaging updates which results in signaling overhead. Hence, a new method of active load sharing that is distributed in nature is proposed in this paper. The proposed method contributes a load balancing mechanism at the registration time itself using the concept of preferred home agent. The paper investigates the existing methods and presents the comparative analysis with the proposed method. The advantages of our proposed load sharing are active and distributed approach, less signaling overhead and better throughput.

Emotion recognition systems are in huge demand to understand the emotion of human towards human, animals, computers, machines and systems. This has influenced such systems to be employed for applications in various domains such as website customization, study of audience reaction in theaters, gaming, software engineering, education and many more. In this paper, a real-time emotion recognition system is designed which is capable of recognizing six human emotions of any person in front of the camera, without any prior information about the person. This is achieved using a combination of both geometric and appearance based features. In order to assess and enhance the performance of the proposed design, the system is tested using standard CK+ datasets too. The system designed is evaluated using three different classifiers and their results are reported. Maximum accuracy of 98.73% is achieved by the system. The proposed system is designed using open source software and can be used for various IoT based applications too.

The menace of counterfeit drugs has prompted the regulatory authorities to mandate trace and track systems to verify the provenance of drugs as it travels across the supply chain. Here, we propose a highly scalable and enhanced trace and track system for pharma supply chain. Our novel idea of using an IoT framework known as GDP (Global Data Plane) integrated with blockchain helps in communication and management of data between untrusted parties. The blockchain maintains an immutable record of drugs each party holds and transactions the parties make amongst themselves. This makes it hard to introduce counterfeit drugs into the supply chain.

Internet of things (IoT) is surrounded by heterogeneous entities such as sensors, mobile devices and actuators in a constrained environment which are running on very low power and lossy networks. These entities are also of very small memory and can handle small computational overhead. To this end, complete IoT system and different devices which are working in home network management system will be presented in this paper. Applications for home network are considered under different architectures and their designs are discussed. Conventional simple network management protocol (SNMP) is generally applied for network management but it is not optimal due to lack of flexibility in configuring devices and lack of capabilities in managing operations in an IoT network. A design of smart washing machine device using the IoT design methodology is being discussed using network configuration protocol (NETCONF) and yet another next generation (YANG) data modeling language to illustrate how it would be a better alternative for managing home network.

The quality of health care has been drastically improved with the evolution of Internet. Electronic health records play a major role in interoperability and accessibility of patient’s data which helps in effective and timely treatment irrespective of the demographic area. The proposed model is to ensure and monitor maternal health during pregnancy and to create awareness alerts (options include messages, voice alerts or flash the system) based on the individual health record. The system aims to prevent maternal death due to medical negligence and helps to make recommendations to prevent future mortality based on medical history and take appropriate action. Authentication is a critical aspect considering the trade-off between usability and security whereas data breach and related cybercrime are major concerns in health care. The proposed model uses DNA based authentication techniques to ensure usability and confidentiality of electronic data, Aadhaar to prevent unauthorized access to patient’s data in case of emergency without affecting availability.

There is a serious mismatch between the growing traffic volume and the availability of resources to support the traffic. Some of the important reasons for this mismatch are the rapid development of our economy, increased affordability of our society, multiple vehicles per family, and so on. We believe that the mismatch will continue to grow and adversely affect our traffic infrastructure unless efficient traffic management solutions that include system integration, design, prediction, safety verification, validation, and security are developed and deployed. Security has appeared as an important issue for Intelligent Transportation Systems (ITS). Some security threats become more challenging task with the emergence of Vehicle-to-Vehicle (V2V) communication and Vehicle-to-Roadside (V2R) communication in vehicular networks. Addressing the security issues in smart vehicular communication systems requires new effective and efficient algorithm that encompass considerations of new security techniques, safety things, communication related resource limitations, and other related new performance metrics. In this paper, we recommend a unified framework and new metrics that combines integrated modelling, system integration and optimization, official certification and validation, and automatic synthesis approaches for analysing the security and safety of ITS and booming out design space investigation of both in-vehicle electronic control systems and vehicle-to-vehicle communications. This integrated framework will facilitated the system integration and optimization and enable validation of various design the new metrics for vehicular networks such as timing, direction, reliability, speed, security and performance.

Diverse performance requirements of the emerging 5G cellular systems and their deployment challenges motivate researchers to explore high frequency millimetre wave (mmWave) spectrum as a potential solution. The allocation and utilization of mmWave spectrum in cellular communication is a new frontier. In this paper, we investigate a directional mmWave small cell outdoor propagation channel by exploiting its deterministic nature; using ray tracing method. Estimates on specific attenuation measurements and free space propagation parameters reveal that directional transmission is inevitable and would result in a channel model divergent from that of an omnidirectional propagation model. In addition, we examine the effect of antenna tilting in the access links to establish highly directional adaptive link. Capacity of the sparsely faded channel is also analyzed. Results exemplify that, beyond 50 m propagation range, the received signal strength in mmWave small cells employing base stations of height 5 m as opposed to macro cells with base station height of 20 m is independent of beam steering. A simplified geometry devoid of the complexity of adaptive beam steering is hence proposed to provide uniform signal strength in an outdoor small cell channel to affirm low latency.

We present a systematic simulation study of dimension-induced photonic band-gap tuning in two-dimensional (2-D), hexagonal lattice photonic crystals, consisting of air-holes in dielectric slabs. Photonic crystals are interesting candidates for application in various fields e.g. communication ranging from optical to THz regime, sensing, spectroscopy, imaging etc., using their property to trap and harness light and to produce high-Q resonances by the principle of localization and photonic bandgap formation. The insensitivity towards launched light wavelength shown herein by the bandgap response of a given 2-D planar photonic crystal is promising for enabling cheaper visible or NIR light sources to produce desired response in Mid-IR wavelengths with ease. The structures and material studied lie within the range of popular fabrication methodology. The results show that silicon photonic crystals, operated at 1.55 µm, can produce sharp resonances and large band transmission in Mid-IR wavelengths (3–5 µm) as well.

For a stated core, the test time changes in a staircase pattern with the width of Test Access Mechanism (TAM). The core test time cannot decrease all time with increase in TAM width. However, the test time can always be diminished with increasing the test clock speed but clock speed cannot be increased beyond power limits. Here, a new method is proposed to reduce the Network on Chip (NoC) test time, by differing the test clock frequency such that it doesn’t cross the predefined power limit. The power dissipation, test clock frequency and overall test time is the three trade off. In the proposed method, the clock frequency is optimized to minimize the total test application time (TAT) considering the power limits. Experimental results show an reduction of 48% over existing solution for the benchmark system on chip (SoC) D695, P93791 and P22810.