Non-contact Abnormal Physiological Status Detection during Sport and Training

In sport and training, it is necessary to continue monitoring the physiological parameters of athletes to ensure that they can maintain a high level of competitive state. The previous monitoring physiological status methods mainly are contactable by sensors that are worn on body. This paper adopts a non-contact physiological parameter monitoring method by using imaging photoplethysmography (iPPG). In order to eliminate the noises in iPPG signals, the correlation energy entropy threshold adaptive denoising and variance characterization sereies are introduced to resist the noises from external conditions. The noises are remove by a threshold which is estimated by noise energy entropy. The constructed signals after denoising are used to estimate physiological parameters, such as heart rate and respiratory rate. The experimental results demonstrate that it estimates the physiological parameters better by usng iPPG based physiological parameter monitoring method than previous methods.

Early Conflict Checking in Two Phase Commit Protocol for Replicated State Machines

Representing an algorithmic workflow as a state machine is a frequently used technique in distributed systems. Replicating a state machine in a fault tolerant way is one of the main application areas under this context. When implementing a replicated state machine, a crucial problem is to maintain consistency among replicas that might handle various different requests arriving at each different replica. This problem requires maintaining a single consistent ordering of the distributed requests handled separately by replicas. Basic consensus protocols such as two phase commit (2PC), can be used to maintain consistency between replicas whenever a request is to be processed. In this study we modify 2PC protocol to take advantage of basic properties of a state machine and detect possible write conflicts earlier. Our experiments on distributed cloud environments show that our modified 2PC protocol increases the throughput and decrease wasted write operations by a significant amount.

Prediction of Breast Cancer using Extremely Randomized Clustering Forests (ERCF) Technique

cancer in breast indeed a significant public health concern in both developed and developing countries female population. It is almost one in three cancers diagnosed in all women. Data mining and pattern recognition applications in conjunction have been proven to be quite useful and relevant to extract the information useful for the medical purpose. This research work reflects the work based on Extremely Randomized Clustering Forests (ERCF) technique which is nothing but a type of pattern recognition technique that may be implemented as the prediction model for Breast Cancer (BC). The accuracy achieved through ERCF has also been compared with that of k-NN(Correlation) and k-NN(Euclidean) in this research work (where k-NN refers to k-Nearest Neighbours technique) and thereafter, final conclusions have been drawn depending upon the testing attributes. The results show that the accuracy of ERCF in the forecasting of breast cancer is so much larger than that of the exactness of k-NN(Correlation) and k-NN(Euclidean). Hence, ERCF, a randomized technique for pattern classification, is best

MQTT based QoS Model for IoT - M2M Critical Applications

The expeditious development of information technology provides opportunities for new remote and monitoring critical systems to be performed based on IoT technologies and M2M communications. This paper discusses important QoS issues in IoT systems and suggests a new QoS model for critical IoT applications, where each information must be delivered only once and in real-time. The proposal is based on the MQTT protocol with dynamic QoS handling, accordingly to the information importance. A prioritization scheme is adopted using different traffic classes, considering specific requirements for real-time communications and reliable operations while reducing end-to-end delay, packet loss, bandwidth, and energy consumption.

Distributed Recommendation Considering Aggregation Diversity

Recommender systems (RSs) are popular in e-commerce as they suggest different kinds of items for different users. Most existing research works focus on how to improve the accuracy of recommender systems. Recently, some recommendation ranking techniques have been proposed to obtain more diverse recommendations for all the users. In this paper, the authors propose design a distributed mechanism for improving the aggregated recommendation diversity and define three new metrics to evaluate the diversity of RSs. To avoid the disclosure of information to a central agency, a distributed mechanism is designed to collect user ratings. To increase the diversity of set recommendations, user-based and item-based weighted methods are proposed. The tasks of them are to deal with non-ratings by weighting the common ratings and calculating the weighted cosine similarities to predict the unknown ratings.

OCEDS

Many organizations require handling a massive quantity of data. The rapid growth of data in size leads to the demand for a new large space for storage. It is impossible to store bulk data individually. The data growth issues compel organizations to search novel cost-efficient ways of storage. In cloud computing, reducing an execution cost and reducing a storage price are two of several problems. This work proposed an optimal cost-effective data storage (OCEDS) algorithm in cloud data centres to deal with this problem. Storing the entire database in the cloud on the cloud client is not the best approach. It raises processing costs on both the customer and the cloud service provider. Execution and storage cost optimization is achieved through the proposed OCEDS algorithm. Cloud CSPs present their clients profit-maximizing services while clients want to reduce their expenses. The previous works concentrated on only one side of cost optimization (CSP point of view or consumer point of view), but this OCEDS reduces execution and storage costs on both sides.

Performance Analysis of Hadoop YARN Job Schedulers in a Multi-Tenant Environment on HiBench Benchmark Suite

Big data processing technology marks a prominent place in today's market. Hadoop is an efficient open-source distributed framework used to process big data with fewer expenses utilizing a cluster of commodity machines (nodes). In Hadoop, YARN got introduced for effective resource utilization among the jobs. Still, YARN over-allocates the resources for some tasks of a job and keeps the cluster resources underutilized. This paper has investigated the CAPACITY and FAIR schedulers' practical utilization of resources in a multi-tenancy shared environment using the HiBench benchmark suite. It compares the above MapReduce job schedulers' performance in two scenarios and proposes some open research questions (ORQ) with potential solutions to help the upcoming researchers. On average, the authors found that CAPACITY and FAIR schedulers utilize 77% of RAM and 82% of CPU cores. Finally, the experimental evaluation proves that these schedulers over-allocate the resources for some of the tasks and keep the cluster resources underutilized in different scenarios.

An Empirical Cluster Head Selection and Data Aggregation Scheme for a Fault-Tolerant Sensor Network

In order to improve the sensor network, the nodes resources should be used in a well-organized way. The new cluster-based routing protocols and data aggregation approach have helped to increase the lifespan of the network. The methods of data aggregation eliminate the network's redundant data packets, which extends the lifetime of the network. A fault tolerant cluster head selection and data aggregation scheme (FT-CHSDA) that performs node clustering and data aggregation in the network is demonstrated in this study. The suggested method uses the energy level of the node to pick the most energy-efficient node as the head of the cluster and executes data aggregation to reduce redundant data packets. In addition, the use of a concept called backup node in a cluster has implemented a novel method to make the network accessible and run without any interruption. In the NS2 simulator, the simulation of the proposed scheme (FT-CHSDA) is being discussed. Using different performance metrics to assess its effectiveness, the proposed scheme (FT-CHSDA) is contrasted with existing proto-cols.

A Survey

Agriculture occupation has been the prime occupation in India since the primeval era. Nowadays, the country is ranked second in the prime occupations threatening global warming. Apart from this, diseases in plants are challenging to this prime source of livelihood. The present research can help in recognition of different diseases among plants and help to find out the solution or remedy that can be a defense mechanism in counter to the diseases. Finding diseases among plant DL is considered to the most perfect and exact paradigms. Four labels are classified as “bacterial spot,” “yellow leaf curl virus,” “late blight,” and “healthy leaf.” An exemplar model of the drone is also designed for the purpose. The said model will be utilized for a live report for extended large crop fields. In this exemplar drone model, a high-resolution camera is attached. The captured images of plants will act as software input. On this basis, the software will immediately tell which plants are healthy and which are diseased.

On Detecting Abnormal Access for Online Ideological and Political Education

With the development and spread of networks, online education has become a new way in education. The online education platform encounters a large number of concurrent visiting, while the system must guarantee network security in the process of online education. The network visiting requests are real-time and dynamic in online education. In order to detect network intrusion and abnormal access in real time and adapt to the dynamic changes of network visiting requests, this paper adopts a data stream-based network intrusion detection method to monitor and manage online education visiting. First, a knowledge library is constructed by normal visiting mode and abnormal visiting mode. Second, the dissimilarity between data point and data cluster is used to measure the similarity between normal mode and abnormal mode. Lastly, the knowledge library is updated to reflect the changes of network in online education system by re-clustering. The proposed method is evaluated on a real dataset.