The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
Many architects believe that major improvements in cost-energy-performance must now come from domain-specific hardware. This paper evaluates a custom ASIC-called a Tensor Processing Unit (TPU)-deployed in datacenters since 2015 that accelerates the inference phase of neural networks (NN). The heart of the TPU is a 65,536 8-bit MAC matrix multiply unit that offers a peak throughput of 92 TeraOps/second...
Most servo control algorithm implementations have been realized using digital signal processors (DSPs) due to large amount of digital signal processing. Nowadays, parallel programmable logic devices, such as the field programmable gate array (FPGA), have become powerful hardware options, offering low cost, high execution speed, reconfigurability and parallelism. This work intends to exploit the current...
The paper presents a reliable method to generate on line pulse width modulation (PWM) signals with selected harmonic elimination (SHE) by using field programmable gate array (FPGA). These signals is used to drive voltage source inverter. Due to the complexity of solving the nonlinear equations, and impossibility to achieve the solution in real time and on line, therefore the solution is obtained off...
Fixed-point VLSI architecture for 2-Dimensional Kurtotic FastICA with reduced and optimized arithmetic units, is proposed. This reduction is achieved through the removal of the dividers for eigenvector computation and replacing the dividers in the Whitening block of the architecture by multipliers. In addition, the number of multipliers required in the Whitening block is further reduced by exploiting...
Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.