Skip to main content

Documentation Index

Fetch the complete documentation index at: https://openntl.org/llms.txt

Use this file to discover all available pages before exploring further.

Why the fabric layer requires IPv6 and why this aligns with Africa’s infrastructure trajectory.
Status: RESEARCH — Blocker identified, solution documented

The Problem

NTL’s signal model requires any node to receive signals from any other node at any time, without pre-established connections. IPv4 NAT breaks this — devices behind NAT are not directly addressable.

How NAT Breaks Signal Propagation

WorkaroundProblem
Device initiates firstContradicts activation model
Relay serverCentralised, adds latency, single point of failure
STUN/hole punching60-80% success, seconds of latency, coordination server needed
None are compatible with neural signal propagation at scale.

The Solution: IPv6

IPv6 gives every device a globally routable address. No NAT, no relay, no hole punching. Any node can signal any other node directly.

Layered Requirements

LayerScopeIP Requirement
ChannelSame processNone
LocalSame machineNone
NetworkSame LANIPv4 or IPv6
FabricGlobalIPv6 required
GSPA (API transport) works on IPv4 through traditional client-server model.

Africa’s IPv6 Trajectory

African mobile networks are deploying IPv6 faster than legacy Western networks — no legacy IPv4 to protect, mobile-first infrastructure, AFRINIC address pressure, Android IPv6 support since 2014.

Fallbacks

  1. GSPA on IPv4 (lose GSPN advantages, retain sync)
  2. IPv6 tunnel (adds latency)
  3. Relay node with IPv6 (intermediary)
  4. Dual-stack (most common as deployment progresses)

Implementation Notes

  • Device-level IPv6 detection at startup
  • IPv6 privacy extensions (RFC 4941) to prevent tracking
  • Firewall keepalive via NTL heartbeat signals

April 2026 — The Bundu Foundation
Last modified on April 23, 2026