Trends in Communication, Cloud, and Big Data

In wireless networks, the spectrum scarcity problem is migrated to much extent by cognitive radio network. Cognitive radio network has the capability to utilize the unutilized frequency spectrum. Due to change in data rates and channel availability, the task of routing in cognitive radio network has become a very challenging task. In this paper, we have given a review of different routing protocols that can be used in cognitive radio network before it we have given a brief description of cognitive radio network architecture that gives idea about things involved in cognitive radio network. We have also outlined various issues or challenges associated with routing in cognitive radio followed by different routing techniques. The last section of this paper gives brief idea about different approaches that can be used for reducing the effects of different routing issues in cognitive radio network. This paper helps in getting idea about cognitive radio network, their issues or routing protocols.

The multihop wireless network, especially mobile adhoc network, has gained popularity in various contexts that may include independent rescue system, connecting network at the edge in Internet and sub-network in upcoming global Internet of the cyber-physical system. The future-generation network systems will be a kind of software-defined network, where the system will have a possibility of choosing and adopting various routing protocol depending upon the context. This paper discusses the advantages of multihop wireless communication and investigates the performance behaviors of AODV and DSR with different Quality of Services parameters by proposing an enhanced energy model in the base node configuration.

Wireless mobile networks (WMNs) should be tested under various realistic conditions including transmission range, type and pattern of data traffic and movement pattern of the users. This paper is a review of mathematical modeling of mobility models that cater to the movement pattern of the users in IP-based WMNs. We describe five mobility models where movement pattern of the mobile user is estimated by some probability distribution function. The aim of this paper is to present some of mobility models based on classical theories in order to support researchers with options to choose when their protocol or architecture needs to be studied under various mobility patterns.

Depending on application type, cognitive radio wireless sensor network (CRWSN) needs energy-efficient data delivery mechanism that can ensure certain level of reliability. Game theory may be exploited for designing network protocols with different objectives. In this work, a game theory-based framework has been developed that can be used for energy-efficient and reliable data delivery in CRWSN. For this purpose, game theory-based channel allocation and routing protocols for CRWSN shall be developed, in the future. Implementation plan of the proposed framework has been outlined.

In recent time, Internet of Things (IoT) is one of the most popular subjects in various fields of development be it in academia or in industry. Machine learning can help in deriving knowledge from within the data available from millions of interconnected machines or devices. IoT networks are self-organized and decentralized, and such networks experience dynamic changes in their topology and node position. Hence, routing plays a very important role when data packets are transmitted between objects or things. Successful delivery of the data packets is to be ensured. Several nodes in IoT environment may undergo constant movements which result into breaking of interconnections between the devices. The change of topology that too frequently, is unavoidable in IoT networks. Routing of the data packets with efficiency and reliability is a great challenge that needs urgent attention. There has been effort in devising machine learning-based routing techniques for IoT environment. In this article, a survey on different routing techniques developed for IoT environment has been presented. Future research directions in the field of IoT have been outlined.

The book contributes to the subject area of remote sensing and Geographic Information System. It is focused on the study and analysis of automated cloud detection and removal of satellite imagery using the selection of thresholds value for various spectral tests in the perspective of RSGIS (Ramya, KarthiPrem, Nithyasri in IJIACS 3(2), [1], Rafael, Richard in Digital image processing. Prentice Hall, [2]). A significant obstacle of extracting information using satellite imagery is the presence of clouds. Removing these portions of image and then filling in the missing data is an important image-editing task. Traditionally, the objective is to cut the cloudy portions out from the frame and fill in the gaps with clear patches from similar images taken at different time. Remote sensing is providing opportunities in various branches of environmental research. The fields of application for multi-spectral remote sensing instruments in earth observation are monitoring the forests, oceans or urban areas over agricultural applications to the extent of natural resources. A significant prerequisite for analysis of earth observation data is the information that is free from external influences and disturbances. One possible cause of data loss is cloud cover of satellite imagery. Cloud cover is recognized as a significant loss of data and information quality by many scientific studies. The existence of cloud cover is the loss of meaningful data and information because they are a considerable source of uncertainty with regard to the application of any algorithm aiming for the retrieval of land surface (Zakaria, Ibrahim, Suandi in A review: image compensation techniques. pp. 404–408, [3], Sengee, Sengee, Choi in IEEE Trans Consum Electron 56(4):2727–2734, [4], Hardin, Jensen, Long, Remund in Testing two cloud removal algorithms for SSM/I [5]).

Big data analytics is one of the very popular technologies at the present moment. Ensuring security in the big data environment is a challenging task. Unique characteristics of big data environment make it very tricky to ensure security in it. There are unique challenges and requirements. Internet of Things and social networks are contributing in generating a huge amount of data and at the same time making the security scenario complex. In this article, various security threats, security challenges and security requirements in big data environment are reviewed.

The tremendous load on the rather archaic medical system in developing countries has necessitated the need to implement artificial intelligence-enabled automated systems to classify different kinds of electrocardiogram (ECG) traces. To this end, we are proposing a novel R-based open-source software with inherent capability to classify different kinds of automated geometric visualizations along with its categorization based upon similarity indices as measured by earth mover’s distance (EMD). This innovative automated software needs verification and validation by clinical practitioners/cardiologists before being implemented to classify large ECG databases to enhance its machine learning capabilities. We anticipate that integration of this robust automated classifier with divergent platforms such as mobile health applications would enable the subjects/patients to continuously monitor the heart rate themselves.

Pneumonia is a common lung infection in which an individual’s alveoli fill up with fluid and form a cloudy-like structure. Pneumonia is of two types: (a) bacterial and (b) viral, but both the X-rays have a very similar pattern. The accurate identification along with how much extent the person is infected is still a challenge for doctors. In this paper, the use of EMD to correctly identify infected pneumonia lungs from normal non-infected lungs is shown. EMD, also known as Earth Mover’s Distance is the distance of two probability distributions over some region D. First, we preprocessed the images to just have the images of lungs, and then we did some re-scaling, rotation, and normalization of intensity so that we will have a set of uniform size/shape of lungs X-rays, and then, we calculated EMD and compared the results.

In the present times, IoT devices have become ubiquitous for collecting huge volumes of data in real time. These data are of immense value for conducting various types of analyses, critical predictions, and for deriving previously undiscovered insights. Occasionally, because of factors beyond control or errors in the IoT device, anomalies or discrepancies get introduced in the data generated. Therefore, the entire amount of data generated by IoT devices is not always usable; only a part of it is usable. Concepts related to the pre-processing of incomplete data are elaborated in this paper. Also, a number of contemporary stochastic techniques which can be used for forecasting of IoT time series data are discussed in this paper.

If X is sender of a message and Y is the receiver of the same, then X encrypts the same with Y’s public key and then sends the encrypted data to Y. It confirms about authenticity and authorization of receiver. If X sends message and Y receives, X encrypts the message with X’s private key and sends the encrypted data to Y. It confirms authenticity and authorization of the sender [1]. Such a schema could work properly. It involves the usage of a message digest or hash. Hash is a fingerprint or the summary of the message. It carries similar concept of cyclic redundancy check (CRC). Integrity of the data is verified with this process. This process actually confirms that the data should not be damaged between the path of sender and receiver [2]. Hashing confirms few things like (i) complexity of calculation of hash value of a message should be decreased; (ii) it follows a one-way encrypting procedure or technique; and (iii) different hash values should be generated by a particular hashing technique for any two different messages.

Although the current version of this research work might not be fool-proof as this kind of works are currently being researched upon and yet to be successfully implemented at large. Popular software such as WinRAR makes use of AES to encrypt the data and also compresses it at the same time but only one algorithm is used to compress the data [1]. It is sometimes seen that using different lossless compression algorithms multiple times can achieve a better compression ratio and may even achieve better security [2]. Our plan is to achieve such kind of compression ratios while achieving confidentiality with a model that we have devised ourselves.

Landslides are movement of rocks, debris or earth along a slope. There are seven states into which the activities of a landslide can be classified into; they are dormant, relict, suspended, reactivated, active and stabilized. Landslide susceptibility map is very crucial as it provides significant data and information that is required for planning of landslide-prone areas as it provides crucial information about spatial probability regarding the occurrence of landslide which is very important in the planning of land use. This paper provides a review of various studies and experiments conducted in this domain of research. The focus area of the study is Sikkim Himalayan region as it shares some similar type of parameters.

Transportation is the key infrastructure of a country. A country’s economy status depends upon how well the country is served by its roads, railways, airports, ports, pipelines and shipping. The rate at which a country’s economy grows is very closely linked to the rate at which the transport sector grows. As road transport gives personal mobility to persons, the vehicle ownership rate has been increasing at a fast rate round the world. Due to increase in car ownership, the problem of parking is becoming more and more acute day by day (Singh and Sharma in application of advanced parking management system techniques—a case study. Int Organ Sci Res J Mech Civil Eng 3(2):24–28, 2012, [1]). Parking space utilization is one of the major problems in crowded cities. Even after coming up with regulations and dedicated parking spaces, with the number of increasing vehicles on road each day, it is hardly making any difference in the existing vehicle parking problems. Yes, we see a huge improvement in the efficiency of space utilization while parking the vehicles, but there is more that can be done to improve it even further without building an entirely new parking space. Even in the most efficient vehicle parking spaces, where the vehicles are parked in a horizontal row with only a little space (for the driver to get inside/outside of the vehicle) between each vehicle, the little space between each vehicle collectively makes up a lot of space that is wasted in the name of driver’s comfort to get inside/outside the vehicle. This project aims toward developing a solution to eliminate these small spaces between each vehicle without compromising with the driver’s comfort in getting inside/outside the vehicle, in fact offering more convenience in doing so and truly making an efficient use of the parking space. There are two types of car parking systems: automated and traditional. In the long term, automated car parking systems are more cost effective when compared to traditional parking garages. IoT is the best option for smart parking system (Osmani et al. in research paper on smart city parking system. Int J Adv Res Innov Ideas Educ 2(3), 2016, [2]). This project incorporates IoT level 1 technology and uses the concepts of bi-directional rotating conveyor belts using DC motors and Arduino microcontroller, NFC tagging and automation. This report depicts the model of the entire system and can be used to build the physical prototype and also build the technology over it.

In ITS, radar plays an important role. Detection of the target and characterization is very important. Similarly, in dynamic channel condition, target image quality degrades considerable due to sidelobe enhancement and defocusing. The clean algorithm is introduced to eliminate sidelobes and significantly improves the target detection performance of radar signals. The experimental results demonstrate the effectiveness of the CLEAN algorithm.

The paper presents an intelligent method to generate list of songs online for listening, download according to age factor of users. Context-aware is a process used in smartphone or computer system that can sense their physical environment and adopt their behavior correspondingly. It is a Web-based application software that recommended the different songs depends upon the listener choice based on their age group from the music library and also classify the unknown songs in the same cluster depends on the review of user.