نشریه مهندسی برق و مهندسی کامپیوتر ایران
سال پانزدهم شماره 4 (پیاپی 45، زمستان 1396)

This paper aims to present a cloud-based learning agent-oriented approach for verification of the pacemaker behavior by monitoring and heart rate adjustment of an arrhythmic patient. In case of the pacemaker failure or inappropriate heart rate generation, the patient is put at risk. Using the proposed approach, one can directs the pacemaker rate to correct one when it is incorrect. Using a learnable software agent, the proposed approach is able to learn un-predefined situations and operates accordingly. The proposed approach is cloud based meaning that it sends a message through cloud in case of a critical situation. After determining the patient heart rate by pacemaker, the proposed method verifies this rate against the predefined physician suggestion and automatically corrects it based on a reinforcement learning mechanism if there is some conflict. The proposed method was implemented and installed on a tablet as a patient mobile device for monitoring the pacemaker implanted in the patient chest. The contrast between results of our approach and expected results existing in the dataset showed our approach improved the pacemaker accuracy until 13.24%. The use of the software agent with reinforcement learning is able to play a significant role in improving medical devices in case of critical situations.

Several routing algorithms have been presented for multicast and unicast traffic in MPSoCs. Multicast protocols in NoCs are used for clock synchronization, cache coherency in distributed shared memory on-chip multiprocessors, replication and barrier synchronization. Unicast routing algorithms are not useful for multicast. Indeed, when unicast routing algorithms are employed to realize multicast operation, high traffic, congestion and deadlock are imposed to the network. To prevent from these problems, Tree-based and path based techniques have been proposed for multicast in multicomputers (and recently NoCs). In this paper, we present a new multicast routing method to decrease power consumption and multicast message latency based on a reconfigurable NoC architecture. In this line, we benefit from simple switches in our reconfigurable architecture instead of routers; we then divide the network to smaller partitions to make better trees for conducting multicast packets. Our evaluation results reveal that, for both real and synthetic traffic loads, the proposed method outperforms the baseline tree-based routing method in a reconfigurable mesh, and reduces message latency by up to 51% and power consumption by up to 33%.

Nowadays, with the advent and proliferation of high-dimensional data, the process of feature selection plays an important role in the domain of machine learning and more specifically in the classification task. Dealing with high-dimensional data, e.g. microarrays, is associated with problems such as increased presence of redundant and irrelevant features, which leads to decreased classification accuracy, increased computational cost, and the curse of dimensionality. In this paper, a hybrid method using ensemble methods for feature selection of high dimensional data, is proposed. In the proposed method, in the first stage, a filter method reduces the dimensionality of features and then, in the second stage, two state-of-the-art wrapper methods run on the subset of reduced features using the ensemble technique. The proposed method is benchmarked using 8 microarray datasets. The comparison results with several state-of-the-art feature selection methods confirm the effectiveness of the proposed approach.

Since in wireless sensor networks, selection of next hop is critical in attack avoidance and lowering the power consumption, a method based on fuzzy logic is proposed in this paper considering status and report transmission of the nodes. In this method, the next hop is selected considering four factors, based on fuzzy logic system. These factors, indicating four optimized parameters; i.e., degree of node proximity to the shortest path, degree of node proximity to the sink, residual energy ratio of each node, and the number of false filtered messages. This method leads to an increase in energy level as well as maintaining security level in comparison with LEAP protocol. Meanwhile, it is possible to identify safe paths. Comparing with other related methods, it is shown that this method leads to significant reduction in energy consumption level and consequently the life-time of the network is increased. Meanwhile with selecting the appropriate next hop, packet drops are reduced as well.

Nowadays, Internet of Things (IoT) has emerged as an important field in information and communication technologies. Despite the progress of networks and communication technologies, the development of IoT has encountered some challenges mainly with regard to computation power, battery lifetime and memory of mobile devices. In order to overcome these challenges, mobile cloud computing has been raised which uses the cloud storage space and computation power to extend the capabilities of mobile devices. In this regard, some of application’s components are selected to be offloaded to the cloud in order to optimize the execution time and energy consumption of application. Since the mobility has an important effect on the acquired condition of the access network and the quality of the connection, the mobility should be considered while selecting components for offloading. Although a number of mobility-aware offloading approaches has been already proposed, these works suffer from the lack of an appropriate mobility-model, ignorance of the fault-tolerance capability and use of only coarse-grain offloading. In order to address these issues, we propose a mobility-aware offloading scheme which uses the user mobility Markov chain and the fault tolerance capability in order to optimize the offloading decision. Evaluation results show that our proposed method significantly outperforms the existing alternatives, reaching respectively up to 75 and 65 percent enhancement in terms of the execution time and the energy consumption.

Seam carving is one of content aware image retargeting techniques. In this method, a path of pixels with lowest energy, called seam, crossing from top to bottom or from left to right in an image is extracted. By removing or inserting seams, size of the image can be changed. Speed and quality are two main parameters in seam carving. In this paper a new method for speed enhancement of seam carving is proposed. The input image is decomposed into odd and even subimages and searching for seams is performed in parallel in these subimages. Compared to the original seam carving, the proposed method improves the speed at least by two times while maintain image’s quality unchanged. Previous seam searching algorithms can be utilized in our method or it can be combined with other parallel processing schemes. Finally, image quality of the proposed seam carving is improved.

In this paper, An IPTV packet classification algorithm is introduced with adaptive adjustment property and the objective of reducing packet loss ratio in Ethernet passive optical networks. The proposed algorithm, improves weight allocation for WRR scheduling using proper classification and prioritization of arriving packets to OLT. Based on simulation results with NS2 simulator, the packet loss ratio of proposed algorithm, shows a 65% reduction compared to existing methods which leads to perceptible user quality of experience improvement.

The point-set covering is one of the important problems in computational geometry, which has many applications. In this problem, the given points should be covered by at least one geometric shape. A variant of the problem is the point-set separation, in which there are at least two different kinds of points which are colored by different colors. The geometric shapes, which are called separators, should only cover the points of the same color. In this paper, separation of blue and red points by a double-wedge of a given angle θ is considered. The proposed algorithm reports all separator θ angle double-wedges in optimal time O(nlogn).