The Energy hole problem, a common phenomenon in wireless sensor networks, significantly decreases the lifetime of any deployed network. Some of the popular techniques to minimize such problems are using mobile sinks instead of static sinks, extending the transmission range dynamically, and deploying redundant sensor nodes near the base station/sink. The major drawback to these techniques are that energy holes may still be created at some point due to their static nature of deployment, despite having the overall residual energy very high. In this research work, we adopt a new approach by dividing the whole network into equiangular wedges and merging a wedge with its neighboring wedge dynamically whenever individual residual energy of all member nodes of a wedge fall below a threshold value. We also propose an efficient Head Node (HN) selection scheme to reduce the transmission energy needed for forwarding data packets among Head Nodes. Simulation results show that WEMER, our proposed WEdge MERging based scheme, provides significantly higher lifetime and better energy efficiency compared to state-of-the-art Power-Efficient Gathering in Sensor Information Systems (PEGASIS) and contemporary Concentric Clustering Scheme (CCS), and Multilayer Cluster Designing Algorithm (MCDA).

This research work proposes a novel priority aware schedule based charging algorithm that uses wireless power transfer (WPT) technique in order to charge embedded sensor nodes (SNs) in a wireless body area network (WBAN). Implanted sensor nodes in WBANs require energy for both information extraction and data transmission to the remote controller unit. Thus, energy shortage of these SNs deteriorates due to the data transmission process of the patient health monitoring system. However, continuous operation by means of electromagnetic induction for energy harvesting, obtained from ambient sources, reduces the overall efficiency of the primary unit. With this paradigm in sight, an algorithm demonstrating the modeling of a priority-based mechanism is proposed in order to ensure proper sensor voltage level and to reduce the transmission losses. Medium access control (MAC) protocols are used for inductive powering from the primary unit to the secondary unit in a collision-free centralized scheduling scheme. Therefore, the proposed wireless charging algorithm for implanted SNs in WBAN is designed as per carrier sense multiple access with collision avoidance (CSMA/CA) technique. Because of this, the overall power consumption of SNs for certain operation periods, successful charging probabilities for multiple SNs, and instantaneous power requirements are considered as key performance measures of analysis. It is assumed that proper energy storage in both transmitters and receivers can handle channel interference and traffic contention. Simulation results verify that a significant reduction in power consumption for the proposed priority aware algorithm will maintain almost similar output. For this reason, saturating class—C as well as class—E driver circuits have been used to justify the performance in two different circuit topologies. Effects of priority with respect to the full charge period have also been observed for the multi-node system. Furthermore, from performance analysis, it has been demonstrated that the scheduling scheme causes both single MOSFET composed saturating class—C and L choke modeled class—E associated driver circuits to be considerably more loss efficient than corresponding existing ones.

No abstract

Spiking neural networks (SNN) offer a power efficient, biologically plausible learning paradigm by encoding information into spikes. The discovery of the memristor has accelerated the progress of spiking neuromorphic systems, as the intrinsic plasticity of the device makes it an ideal candidate to mimic a biological synapse. Despite providing a nanoscale form factor, non-volatility, and low-power operation, memristors suffer from device-level non-idealities, which impact system-level performance. To address these issues, this article presents a memristive crossbar-based neuromorphic system using unsupervised learning with twin-memristor synapses, fully digital pulse width modulated spike-timing-dependent plasticity, and homeostasis neurons. The implemented single-layer SNN was applied to a pattern-recognition task of classifying handwritten-digits. The performance of the system was analyzed by varying design parameters such as number of training epochs, neurons, and capacitors. Furthermore, the impact of memristor device non-idealities, such as device-switching mismatch, aging, failure, and process variations, were investigated and the resilience of the proposed system was demonstrated.

Multiple Sclerosis (MS) is a neuroinflammatory demyelinating disease that affects over 2,000,000 individuals worldwide. It is characterized by white matter lesions that are identified through the segmentation of magnetic resonance images (MRIs). Manual segmentation is very time-intensive because radiologists spend a great amount of time labeling T1-weighted, T2-weighted, and FLAIR MRIs. In response, deep learning models have been created to reduce segmentation time by automatically detecting lesions. These models often use individual MRI sequences as well as combinations, such as FLAIR2, which is the multiplication of FLAIR and T2 sequences. Unlike many other studies, this seeks to determine an optimal MRI sequence, thus reducing even more time by not having to obtain other MRI sequences. With this consideration in mind, four Convolutional Encoder Networks (CENs) with different network architectures (U-Net, U-Net++, Linknet, and Feature Pyramid Network) were used to ensure that the optimal MRI applies to a wide array of deep learning models. Each model had used a pretrained ResNeXt-50 encoder in order to conserve memory and to train faster. Training and testing had been performed using two public datasets with 30 and 15 patients. Fisher’s exact test was used to evaluate statistical significance, and the automatic segmentation times were compiled for the top two models. This work determined that FLAIR is the optimal sequence based on Dice Similarity Coefficient (DSC) and Intersection over Union (IoU). By using FLAIR, the U-Net++ with the ResNeXt-50 achieved a high DSC of 0.7159.