Abstract
A method and system for transmitting real-time software behavior and control signals by mapping dynamic system states to the native compression channels of the H.265 (HEVC) video codec. The protocol, termed "SpikeNSL," treats the video frame as a raster plot of spiking neurons. The Luma (Y) channel encodes the "Spike Train" of system events (sparse binary activations), while the Chroma (U/V) channels encode contextual state transitions. This mapping exploits the temporal redundancy and sparsity optimizations of H.265 to achieve bandwidth efficiencies of 50-100x compared to traditional serialization (e.g., MQTT/gRPC) for sparse event streams. The system further enables "No-Compile" behavioral replication, where a receiving controller minimizes the Victor-Purpura Distance (DVP) between its local sensor spikes and the received SpikeNSL stream, bridging the Sim-to-Real gap without source code compilation.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Judson, Matthew D., "SpikeNSL (Spiking Neural Signal Language) for "No-Compile" Behavioral Transmission", Technical Disclosure Commons, (December 02, 2025)
https://www.tdcommons.org/dpubs_series/8949