Systems and methods for converting an analog machine learning output signal into a digital signal
AssigneeEva Technology Corp
InventorsKevin Blaine Anderson, Donald Wood Loomis, III
Life of this patent
Estimating the term…
Loading patent…
AssigneeEva Technology Corp
InventorsKevin Blaine Anderson, Donald Wood Loomis, III
Estimating the term…
Systems and methods are disclosed for analog-to-digital-converter (ADC) solutions supporting analog AI deep learning systems. These systems can outperform their digital counterparts in speed and energy efficiency since computations are conducted directly in memory and analog processors inherently support parallel operations. ADCs play a crucial role in analog AI deep learning because they serve as the bridge between the analog and digital worlds. An analog AI deep learning system may comprise a DAC, a programming module, row/column switches, a crossbar array and an ADC. The ADC may comprise switches for time interleaving of an analog neural network output signal, received from the row/column switches associated with the crossbar array. The time interleaved signals are separately ported between two capacitive paths, and are subsequently coupled to separate inverters. A frequency signal is generated by a logic gate, which is coupled to a digital filter that generates an n-bit digital word.
A method for operation of an Analog-to-Digital Converter (ADC) comprising:
A system for an Analog-to-Digital Converter (ADC) comprising:
A system for an Analog-to-Digital Converter (ADC) for an analog deep learning system comprising:
Dependencies are inferred from the claim text.