The propagation of spin waves and their interaction with the spin solitons like skyrmions, domain walls and vortex are one of the promising ways for designing nanoscale spintronic devices. Magnetic skyrmion, a particle-like nanoscale object has potential applications in next-generation spintronic devices. In this paper, the unidirectional motion of the skyrmion under the influence of spin wave is studied using micromagnetic simulations. Here, two different magnetic anisotropies are considered on a nanotrack that creates an energy gradient. As a result, the repulsive forces act on the skyrmion and is responsible for the motion of the skyrmion in one direction. The spin wave driving force leads the skyrmion to move in forward direction and the anisotropy gradient is responsible to prevent the skyrmion motion in reverse direction. The skyrmion moves from higher perpendicular magnetic anisotropy region to lower energy region, leading to a unidirectional transport of the skyrmion. This proposed device has less Joule heating and is more energy efficient as compared to other spin transfer torque (STT) and spin orbit torque (SOT) driven techniques. This is due to the fact that spin wave can generate a flow of magnetic momentum without generating an electron flow. This spin wave driven skyrmionics device is a promising pathway towards the development of a complete non-charge based magnetic devices.
The development of energy-efficient and ultrafast neuromorphic computing based on the dynamics of the ferromagnetic (FM) skyrmion on the nanotrack has attained considerable interest. In this work, FM skyrmion based artificial neuron device is proposed. The perpendicular magnetic anisotropy (PMA) gradient is created on a thin film ferromagnetic (FM) layer by voltage control-PMA effect (VC-PMA). The anisotropy is directly co-related with the strength of 𝑚𝑧 that affects the size of skyrmion meaning that in the region with larger PMA, the skyrmion size is smaller and hence, more energy. However, the skyrmions have the tendency to move in the direction to minimize the energy. Hence, the skyrmion move towards the lower PMA. This behavior of skyrmion on a nanotrack with PMA gradient corresponds to the leaky-integrate-fire (LIF) functionality of the neuron device. Hence, the suggested energy-efficient artificial neuron opens up the path for developing for energy-efficient neuromorphic computing.
Neuromorphic computing (NC) has emerged as one of the leading computation methodologies with low power consumption. Conventional CMOS based implementation of a complex computing system leads to power hungry, unscalable and inefficient designs. The present technologies are inefficient for computation intensive tasks such as neural networks. Hence, there is a need for unconventional devices and techniques. Nano-scale devices with the ability to mimic biological computation efficiently have been explored. Non-volatile memories such as resistive random access memory (RRAM) and phase change memory (PCM) are used for mapping synaptic functionality. Spintronic devices are among the suitable candidate to implement NC with low power consumption. Neuron as well as synapse functionality for neural networks are mapped using spin devices such as spin transfer torque magnetic random access memory (STT-MRAM) and spin orbit torque (SOT-MRAM). This paper compares the performance metrics of spin devices with other non-volatile memories for the implementation of neural network architecture with a single hidden layer. Spin device architecture consumes less area and much lower leakage power while to achieve the same level of accuracy as other devices. The MNIST dataset image classification achieved 1.17X reduction in leakage power, 9.42X reduction in latency and 1.08X reduced area consumption using SOT-MRAM device as compared to RRAM, PCM and conventional static random access memory (SRAM) respectively.
Magnetic skyrmions are chiral spin textures which are topologically stable entities. These are the promising candidates for data storage and logic devices owing to nanoscale size and remarkably low current density required to drive them in contrast to the magnetic domain walls. Antiferromagnetic (AFM) skyrmions are spin textures with equal and opposite topological charge in each sub-lattice leading to cancellation of Magnus force in contrary to ferromagnetic (FM) skyrmions that exhibits skyrmion (SkHE) Hall effect. FM skyrmions also lacks behind the AFM skyrmions by virtue of their sensitivity towards the stray fields. The formation of skyrmions in magnetic materials possessing broken inversion symmetry is subject to material parameters. Therefore, in this paper, we have demonstrated the nucleation of AFM skyrmions by injection of spin polarized current and effect of tuning the magnetic anisotropy and Dzyaloshinskii-Moriya interaction (DMI) constant on skyrmion nucleation. The variation of velocity w.r.t. the applied current density in a nanotrack is also investigated.
Currency classification is an important task in computer vision. Traditional models extract relevant features (brightness, shape, colour etc.) through complex mathematical calculations. A deep learning approach towards fiscal classification by automatically learning higher order feature representations of the dataset is presented. A family of Resnet models is trained to minimize the effect of distortions in currency dataset. The classifier achieves a peak test set performance of 98.09% and an ensembled accuracy of 98.3%. Finally, an optimization method is introduced to allow the models initialized with pretrained weights to converge faster and achieve better accuracy in certain cases.
The growth in number of transistors on a semiconductor chip has followed the Moore's law over the years and simultaneously, the size of transistor has reached equivalent to atomic size. Hence, a novel approach is needed for replacement in CMOS technology. The existing technologies are unable to meet the complex computing challenges due to their inability of inherent parallel processing and adaptability. Hence, there is a need of unconventional methodologies such as quantum computing and neuromorphic computing; and technologies like spintronics to support the realization of these methodologies. Skyrmions are the primary components of skyrmionics, the next generation of spintronics. These are the promising candidates for neuromorphic computing due to their significant features such as topological stability, low driving current density and nanoscale size. In this paper, spintronics based neuromorphic computing has been reviewed and the performance is evaluated in comparison to conventional techniques.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.