射频光子深度学习处理器,具有Shannon限制数据移动
Ronald Davis1, Zaijun Chen1,2, Ryan Hamerly1,3
1Research Laboratory of Electronics, MIT, Cambridge, MA 02139, USA.
Science advances
|June 11, 2025
概括
研究人员开发了一种新的光学神经网络 (ONN),用于更快的AI. 这种乘法模拟频率转换光学神经网络 (MAFT-ONN) 加快了对射频信号的深度学习,为先进的6G通信提供了一条道路.
科学领域:
- 光子学和人工智能的人工智能
- 先进的通信系统先进的通信系统
- 半导体技术 半导体技术
背景情况:
- 埃德霍姆定律预测通信数据速率的指数增长,需要超越摩尔定律的新计算范式.
- 深度神经网络 (DNN) 面临着日益增长的计算需求,挑战当前的硬件加速器.
- 光学神经网络 (ONN) 为高速AI提供了潜力,但面临着可扩展性和系统开销问题.
研究的目的:
- 为先进的通信系统引入一种新型的人工智能硬件加速器.
- 展示一种完全模拟的深度学习方法来处理原始射频 (RF) 信号.
- 在可扩展性和系统开销方面解决当前ONN的局限性.
主要方法:
- 开发和实验验证一个乘法模拟频率转换光学神经网络 (MAFT-ONN).
- 实现完全模拟的深度学习计算直接在射频信号上.
- 测试MAFT-ONN用于调制分类和MNIST数字分类任务.
主要成果:
- MAFT-ONN在调制分类任务中实现了95%的准确性,并实现了快速的融合.
- 经过近400万次完全模拟操作的证明可扩展性,用于MNIST数字分类.
- 由于模拟数据的移动,速度比传统的射频接收器快数百倍.
结论:
- 在未来的通信系统中,MAFT-ONN为人工智能硬件加速提供了一个有前途的解决方案,例如6G.
- 模拟处理方法克服了数字射频接收器和当前ONN架构的局限性.
- 这项技术使原始射频信号的高效,高速深度学习成为可能,为下一代无线技术铺平了道路.
相关概念视频
Ampere-Maxwell's Law: Problem-Solving
567
A parallel-plate capacitor with capacitance C, whose plates have area A and separation distance d, is connected to a resistor R and a battery of voltage V. The current starts to flow at t = 0. What is the displacement current between the capacitor plates at time t? From the properties of the capacitor, what is the corresponding real current?
To solve the problem, we can use the equations from the analysis of an RC circuit and Maxwell's version of Ampère's law.
For the first part of...
To solve the problem, we can use the equations from the analysis of an RC circuit and Maxwell's version of Ampère's law.
For the first part of...
567
Fast Decoupled and DC Powerflow
178
The fast decoupled power flow method addresses contingencies in power system operations, such as generator outages or transmission line failures. This method provides quick power flow solutions, essential for real-time system adjustments. Fast decoupled power flow algorithms simplify the Jacobian matrix by neglecting certain elements, leading to two sets of decoupled equations:
178


