A. M. Jehad Sarkar

A Framework for Supervising Lifestyle Diseases Using Long-Term Activity Monitoring

Activity monitoring of a person for a long-term would be helpful for controlling lifestyle associated diseases. Such diseases are often linked with the way a person lives. An unhealthy and irregular standard of living influences the risk of such diseases in the later part of ones life. The symptoms and the initial signs of these diseases are common to the people with irregular lifestyle. In this paper, we propose a novel healthcare framework to manage lifestyle diseases using long-term activity monitoring. The framework recognizes the users activities with the help of the sensed data in runtime and reports the irregular and unhealthy activity patterns to a doctor and a caregiver. The proposed framework is a hierarchical structure that consists of three modules: activity recognition, activity pattern generation and lifestyle disease prediction. We show that it is possible to assess the possibility of lifestyle diseases from the sensor data. We also show the viability of the proposed framework.

A New Data Mining Scheme Using Artificial Neural Networks

Classification is one of the data mining problems receiving enormous attention in the database community. Although artificial neural networks (ANNs) have been successfully applied in a wide range of machine learning applications, they are however often regarded as black boxes, i.e., their predictions cannot be explained. To enhance the explanation of ANNs, a novel algorithm to extract symbolic rules from ANNs has been proposed in this paper. ANN methods have not been effectively utilized for data mining tasks because how the classifications were made is not explicitly stated as symbolic rules that are suitable for verification or interpretation by human experts. With the proposed approach, concise symbolic rules with high accuracy, that are easily explainable, can be extracted from the trained ANNs. Extracted rules are comparable with other methods in terms of number of rules, average number of conditions for a rule, and the accuracy. The effectiveness of the proposed approach is clearly demonstrated by the experimental results on a set of benchmark data mining classification problems.

Energy-Efficient Scheduling Algorithms for Data Center Resources in Cloud Computing

A significant amount of energy is consumed to render high-level computation tasks in large scale cloud computing applications. The state-of-the-art energy saving techniques based on centralized job placement approaches reduce the reliability of operation due to a single point of failure. Moreover, the existing works do not consider energy consumption cost for communication devices and network appliances which contribute a lot. In this paper, we have proposed a mechanism for cluster formation based on network vicinity among the data servers. We have developed two distributed and localized intra-cluster and inter-cluster VM scheduling algorithms based on energy calculation, resource requirement and availability. Our proposed scheduling algorithms manage VMs to reduce the energy consumption of both the servers and networking devices. Simulation results show that our proposed distributed VM scheduling algorithms can conserve significant amount of energy compared to state-of-the-art works.

An energy-efficient multiconstrained QoS aware MAC protocol for body sensor networks

One of the most challenging jobs in designing a Medium Access Control (MAC) protocol for Body Sensor Networks (BSNs) is to achieve QoS requirements for heterogeneous traffics generated from various sensors. Increasing the energy-efficiency of the network should be kept in mind while doing this. Physiological data monitoring applications generate different types of traffic including multimedia data packets. These heterogeneous traffics should be treated differently by an underlying communication protocol, allowing the transmission schedule of these traffic types based on their priorities. In this paper, we present an energy-efficient multiconstrained QoS aware MAC protocol, namely eMC-MAC, wherein the medium access control is designed based on traffic prioritization. We have redefined the superframe structure in such a way that the critical data packets are transmitted earlier than other packets. In our proposed eMC-MAC protocol, we have introduced minislots during CFP (Contention Free Period), where requests for urgent packets are collected to the coordinator node in an energy-efficient way. We also develop an energy-efficient algorithm for preempting allocated data transmission slots to facilitate transmission of packets with higher priority. Thus, the proposed eMC-MAC protocol delivers emergency data packets to the coordinator with reduced delay. We have evaluated the effectiveness of our eMC-MAC protocol through extensive simulations in ns-3. The simulation results have shown that it outperforms a number of state-of-the-art MAC protocols for BSNs.

Reconciliation of Ontology Mappings to Support Robust Service Interoperability

Information on web and in use of web services is increasing enormously even on hourly bases. On semantic web the information is represented in ontology. For system and services to share the information, a sort of mediation (i.e., mappings) is required. Mappings are established between the ontologies (information sources) of web services for resolving the terminological and conceptual incompatibilities. However, with the discovery of new knowledge in the field and accommodating the knowledge in domain ontologies makes the ontology to evolve from one consistent state to another. This consequently makes existing mappings between ontologies unreliable and staled due to the changes in resources. So there is a need for mapping evolution to eliminate discrepancies from the existing mappings. To re-establish the mappings between dynamic ontologies, existing systems restart the complete mapping process which is time consuming. The approach proposed in this paper provides the benefits of mapping reconciliation between the updated ontologies. It takes less time as compared to the existing systems. It only considers the changed resources and eliminates the staleness from the mappings. This approach uses the change history of ontology to drastically reduce the time required for reconciling mappings among ontologies. Experimental results with four different mapping systems using standard data sets are provided that validate our claims.

Prioritized Medium Access Control in Cognitive Radio Ad Hoc Networks: Protocol and Analysis

Cognitive radio (CR) technology enables opportunistic exploration of unused licensed channels. By giving secondary users (SUs) the capability to utilize the licensed channels (LCs) when there are no primary users (PUs) present, the CR increases spectrum utilization and ameliorates the problem of spectrum shortage. However, the absence of a central controller in CR ad hoc network (CRAHN) introduces many challenges in the efficient selection of appropriate data and backup channels. Maintenance of the backup channels as well as managing the sudden appearance of PUs are critical issues for effective operation of CR. In this paper, a prioritized medium access control protocol for CRAHN, PCR-MAC, is developed which opportunistically selects the optimal data and backup channels from a list of available channels. We also design a scheme for reliable switching of a SU from the data channel to the backup channel and vice-versa. Thus, PCR-MAC increases network throughput and decreases SUs’ blocking rate. We also develop a Markov chain-based performance analysis model for the proposed PCR-MAC protocol. Our simulations, carried out in

Hidden Markov Mined Activity Model for Human Activity Recognition

NS-3

ERANN: An Algorithm to Extract Symbolic Rules from Trained Artificial Neural Networks

NS-3, show that the proposed PCR-MAC outperforms other state-of-the-art opportunistic medium access control protocols for CRAHNs.