Harmonic Participation Sieve for Electronic Network Design: Switch Fabric Reduction, Power Grid Routing, and Relay Network Optimization
Authors/Creators
Description
We apply the Harmonic Participation Sieve, a graph-theoretic edge ranking algorithm previously validated in 3D physics simulation and neural network pruning, to three electronic network design problems.
A breadth-first spanning tree guarantees connectivity, while a harmonic participation score ranks each connection by its structural importance.
Applied to a 16×16 switch fabric, the sieve removes 28.8% of switches while retaining 100% of available paths and maintaining an identical minimum hop count.
Applied to a 20×20 VLSI power grid, it identifies 56 critical copper traces for maximum-width routing, versus 397 redundant traces for minimum-width routing.
Applied to a 64-node relay network, it removes 30.6% of links while preserving maximum-throughput path capacity.
Algorithm licensed under PolyForm Noncommercial 1.0.0.
Notes
─────────────────────────────────────────
IMPLEMENTATION AND LICENSING NOTICE
─────────────────────────────────────────
Academic investigation, non-commercial
research, and citation are explicitly
welcomed under the terms of CC BY-NC 4.0
(paper text) and PolyForm Noncommercial
1.0.0 (algorithm and implementations).
Any commercial implementation of the
methods described herein — in any
programming language, hardware platform,
or system architecture — requires explicit
written authorization from the author.
This includes integration into commercial
products, revenue-generating services,
or industrial research pipelines.
The author reserves all intellectual
property rights herein, including the
right to seek patent protection.
A larger theoretical result underpinning
these methods is in preparation and not
disclosed in this publication.
Commercial licensing and partnership
inquiries: lctrnc1@gmail.com
© 2026 Andrés Sebastián Pirolo
─────────────────────────────────────
Files
pirolo2026_electronics_hte.pdf
Files
(366.4 kB)
| Name | Size | Download all |
|---|---|---|
|
md5:f2aac9a55ff3400be9f5628f23c5c4bc
|
163.7 kB | Preview Download |
|
md5:d9f67284da49674f748089468115c27b
|
202.7 kB | Preview Download |
Additional details
Software
- Repository URL
- https://zenodo.org/uploads/20573427
- Programming language
- Python