Amplitude‐Controlled Electromagnetic Pulse Switching Using Waveguide Junctions for High‐Speed Computing Processes

Abstract

In his 1965 paper, [1] Gordon Moore described the relationship between the density of transistors on a silicon chip and the operational speed of that chip. In this seminal work, which is now known as Moore's law, he envisioned that the density of these devices would double every 18 months. In the years since its publication, this prediction has been found to be remarkably accurate. [2] This rapid advancement has been possible by the tremendous scalability of semiconductor devices such as metal-oxide-semiconductor field-effect transistors (MOS-FETs). [3,4] However, Moore recognized that this fast rate of advancement would someday become difficult to continue as there are limits to how small transistor-based devices can be made, [5,6] limits that many academics believe we are now reaching. [7,8] The ability to control the flow of energy via switching techniques can be considered as a fundamental process by which computation takes place. For instance, in classical computing devices, the flow of current is controlled by such semiconductor-based devices (transistors) working as switching elements. However, intrinsic parasitic capacitances are difficult to avoid in such devices. These unwanted reactive elements are charged and discharged during the dynamic switching process performed by transistors, adding further delays to the overall computations which restrict the speed and efficiency of the devices. [9] To overcome this, new paradigms on computing are needed. Different scenarios have been recently proposed such as spintronics, [10] biological computing, [11] computing with electromagnetic (EM) waves, [12,13] and optical solitons, [14][15][16][17][18] among others. EM wave-based computing has become a hot research topic worldwide as the information can be transferred at the speed of light in the medium where the wave propagates. In this context, such fully EM wave-based computing systems that do not require charging/discharging processes have the potential to open new avenues for future high-speed computing. [6] In this realm, metamaterials (MTMs) and metasurfaces (MTSs), as their 2D version, have been recently applied to the field of computing using EM waves. [19][20][21] MTMs are artificial materials that exhibit EM responses not always easy to find in natural media such as negative or near-zero permittivity values. [22][23][24] MTMs have successfully been applied in multiple scenarios such as sensing, [25,26] antennas, [27,28] and imaging, [29,30] demonstrating their ability to arbitrarily control fields

Time derivatives via interconnected waveguides

“…1 B. As in our previous works 16 , 18 , 19 we consider two types of waveguide junction with parallel plate waveguides connected in either series or parallel configuration. The full details of the splitting and superposition of signals at these junctions can be found in 16 , 18 , 19 , here we present the basic concepts for completeness.…”

“…As in our previous works 16 , 18 , 19 we consider two types of waveguide junction with parallel plate waveguides connected in either series or parallel configuration. The full details of the splitting and superposition of signals at these junctions can be found in 16 , 18 , 19 , here we present the basic concepts for completeness. When the characteristic impedance of each waveguide is the same and the cross-section of the junction is small compared to the incident wavelength 48 , the splitting of the incident signal after passing the junction will be described by the following scattering matrices: where is the transmission coefficient of the junction, the number of connected waveguides, and are the identity and all-ones matrices, respectively.…”

Fault diagnosis in high-speed computing systems using big data analytics integrated evolutionary computing on the Internet of Everything platform

“…The advent of high-speed computing systems has become a fundamental aspect of recent technological progress, transforming diverse sectors and stimulating innovative developments [1]. The methods can handle big data, perform intricate calculations at remarkable velocities, and have a long history spanning numerous decades [2].…”

Complex Matrix Equation Solver Based on Computational Metasurface

“…Moreover, the circuit requires a large number of tunable components including amplifiers and phase shifters, which leads to significant increases in cost and complexity, and hinders physical implementation. In addition to the three categories mentioned above, several other innovative structures have been proposed [25,26] but they can only handle some simple digital operations at this stage.…”