Photon AI Chip Digital Pulse-Width Modulation Without DACs
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing photon neural network chips face issues with low accuracy, poor anti-interference performance, high dynamic range requirements, and quantization errors due to analog signal modulation and the need for digital-to-analog conversion, which increases system cost and restricts speed.
Innovation Solution
A digital signal modulation method that converts digital electrical signals into optical signals using timing signals with fixed periods and base clocks, eliminating the need for digital-to-analog converters and improving accuracy and stability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If analog signals are used for optical modulation and computation, then the photonic circuit can perform linear operations with ultra-high bandwidth, but the accuracy is low and anti-interference performance is poor
Solution Approach 1:
The patent changes the signal representation parameter from continuous analog values to discrete digital pulse widths. By encoding information in the duration of optical pulses rather than their amplitude, the system maintains high bandwidth while achieving high precision through the discrete nature of digital timing signals.
2Adaptability or versatility
If digital-to-analog conversion is used to convert digital signals to analog optical signals, then the system can interface with digital chips, but quantization error is introduced and system cost increases
Solution Approach 1:
The patent replaces the electronic digital-to-analog conversion mechanism with a direct digital optical modulation mechanism. Instead of converting digital signals to analog electrical signals and then to optical signals, the system directly modulates optical signals using digital pulse width modulation, eliminating the DA converter and its associated quantization errors.
3Speed
If high-speed and high-precision digital-to-analog converters are used, then the system speed can be improved, but the cost increases and there is risk of embargo
Solution Approach 1:
The patent extracts and removes the digital-to-analog converter from the signal path entirely. By using direct digital optical modulation through pulse width control, the system eliminates the need for high-speed DA converters, thereby reducing cost, complexity, and supply chain risks while maintaining high system speed.
4Ease of operation
If analog signals are used for modulation, then the dynamic range requirements for electro-optical modulators and optical detectors are high, but this increases system complexity
Solution Approach 1:
The patent changes the modulation parameter from analog amplitude to digital pulse width. This transformation reduces the dynamic range requirements because the information is encoded in discrete time intervals rather than continuous amplitude levels, simplifying the requirements for electro-optical modulators and optical detectors.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The method enhances accuracy and precision, reduces power consumption, and softens input requirements, thereby improving the anti-interference performance and stability of photon neural network chips.
Implementation Method 1
modulating one or more groups of digital electrical signals into optical signals
Data Source
AI summary
A digital signal modulation method for a photon artificial intelligence computing chip, including: modulating one or more groups of digital electrical signals into optical signals; where the group of digital electrical signals comprises several timing signals being outputted in sequence in a channel within a fixed period; where each timing signal has the same base clock and signal time length; where each timing signal conveying N-bit digital information has 2N−1 base clocks, the number of the base clocks of a high-level signal or the number of a digital signal “1” in the timing signal is a signal value of the timing signal, and the signal value is equal to a value of the N-bit digital information being transmitted; and where the timing signal is a modulating signal for converting the electrical signal to the optical signal. Compared with the existing calculation scheme based on digital-to-analog conversion and analog signal modulation, the calculation scheme in the present invention avoids using a digital-to-analog converter with a high cost and power consumption, can be directly connected with the digital signals of the electronic chip, and avoids quantization error during the digital-to-analog conversion of digital signals.