RF Interconnections for Oscillatory Neural Networks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Implementing neural networks, particularly oscillatory neural networks (ONNs), on integrated circuits is challenging due to high connectivity requirements, leading to routing difficulties and signal latency issues, especially when operating at higher speeds like radio frequency signals, which affect phase accuracy.
Innovation Solution
The implementation of ONN gates using radio frequency (RF) interconnections via capacitive coupling between oscillator circuits and a transmission line, where the transmission line acts as an averaging circuit, reducing signal latency and improving phase synchronization by synchronizing oscillator frequencies within a determined locking range.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If wired connections are used to connect oscillator circuits in ONN gates, then signal transmission is achieved, but signal latency increases and phase accuracy deteriorates
Solution Approach 1:
The patent replaces traditional wired mechanical/electrical connections with radio frequency electromagnetic field-based communication. Oscillator circuits communicate through RF signals in the electromagnetic spectrum rather than through physical wire connections, eliminating the propagation delays and signal latency inherent in wired systems while maintaining connectivity between neural network components.
Solution Approach 2:
The patent introduces RF interconnection circuits as intermediary components that facilitate communication between oscillator circuits. These RF intermediaries convert electrical signals to electromagnetic signals for transmission and back to electrical signals for processing, enabling faster communication while reducing the direct dependency on physical wire length and routing complexity.
2Productivity
If high connectivity is implemented in neural networks, then computational capability improves, but routing complexity increases
Solution Approach 1:
The patent implements universal RF interconnection circuits that can handle multiple communication functions simultaneously. The same RF infrastructure supports data transmission, synchronization, and coordination across all oscillator circuits in the neural network, eliminating the need for separate dedicated wiring for each connection and significantly reducing routing complexity while maintaining high connectivity.
Solution Approach 2:
By replacing the physical wiring infrastructure with RF electromagnetic field communication, the patent eliminates the need for complex physical routing of numerous connections. The RF signals can propagate through space without requiring physical wire paths, thereby decoupling computational connectivity from physical routing complexity.
3Reliability
If RF interconnections are used in ONN gates, then phase synchronization improves, but energy consumption increases
Solution Approach 1:
The patent employs periodic RF signaling at specific frequencies that correspond to the oscillation frequencies of the neural network oscillators. By using periodic rather than continuous RF transmission, the system achieves phase synchronization only when needed for computational operations, reducing energy consumption compared to continuous RF operation while maintaining reliable phase synchronization during active computation.
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
This approach enables efficient communication between synapses and neurons in ONN gates with reduced latency and energy dissipation compared to wired connections, addressing routing complications and phase instability issues, and can be applied across various platforms including laptops and smart devices.
Implementation Method 1
a first capacitive coupler to couple the first oscillator circuit to the transmission line to generate an oscillating signal in the transmission line
Data Source
AI summary
Techniques are provided for radio frequency interconnections between oscillators and transmission lines for oscillatory neural networks (ONNs). An ONN gate implementing the techniques according to an embodiment includes a transmission line, a first oscillator circuit tuned to a first frequency based on a first tuning voltage associated with a first synapse weight, and a first capacitive coupler to couple the first oscillator circuit to the transmission line to generate an oscillating signal in the transmission line. The ONN gate further includes a second oscillator circuit tuned to a second frequency based on a second tuning voltage associated with a second synapse weight, and a second capacitive coupler to couple the second oscillator circuit to the transmission line to adjust the oscillating signal in the transmission line such that the amplitude of the adjusted oscillating signal is associated with a degree of match between the first frequency and the second frequency.