Enhancement Sub-Protocol v2.0 · Section 8 & 15

Nesting
& Overhead

Enhancement sequences may contain transmission structures within themselves. A formally specified parser stack tracks every nesting boundary and restores outer context on exit.

Nesting allows one BitPads structure — a stream, a record, a session — to contain another complete structure, suspending the outer context while the inner structure executes, then restoring the outer exactly.

The sub-protocol defines six nesting scenarios, two of which require parser stack operations. A ParserStateFrame captures the complete decoder context at the moment nesting is entered. The stack is last-in-first-out. On inner structure close, the frame is popped and the outer context is restored bit-for-bit. Maximum nesting depth is negotiated at session open from three independent limits: hardware capacity, operator policy, and the negotiated minimum of both endpoints.

Overhead is a first-class design concern throughout the sub-protocol. The enhancement grammar is a microservice: it adds zero overhead to transmissions that do not use it. For transmissions that do, overhead is exactly proportional to what is expressed — one byte per signal slot declaration, one bit per session-level activation.

Section 8.5

Nesting Level Codes

The device declares its nesting capability in Layer 1. Two non-contiguous bits in the Layer 1 session defaults — bits 9 and 12 — form a 2-bit nesting level code. This code communicates the maximum nesting depth the sender supports and triggers the Nesting Declaration Extension byte when depth beyond four levels is required.

The Six Nesting Scenarios

The sub-protocol formally defines six scenarios in which one transmission structure contains or overlaps another. Scenarios 1 and 2 are structural nesting requiring stack push and pop. Scenarios 3–6 are partial or non-structural.

Section 8.2 — 8.3

Parser Stack

Scenarios 1 and 2 require a parser stack — a last-in-first-out structure of ParserStateFrame objects. Each frame is a complete snapshot of the decoder context at the moment nesting was entered: the outer mode, category, slot position, component flags remaining, all stream state, all session/batch context, and the enhancement grammar state. When the inner structure closes, the frame is popped and the outer context is restored exactly.

The effective stack depth is the minimum of three independent limits: the hardware capacity derived from available stack memory, the operator-configured policy maximum, and the agreed depth negotiated at session open via DC2 parameter exchange in the P1 slot.

ParserStateFrame Structure

Each frame is approximately 20 bytes on a 32-bit embedded device. The stack memory formula follows directly from this: allocate stack bytes, divide by 20, floor to integer for hardware_max.

Stack Push — Component Escalation

When enhancement field 101 is encountered in a stream signal slot, the current parser context is captured into a frame and pushed onto the stack. The inner record then reads its own Meta bytes, Layer 1, and components independently.

Stack Push — Sub-Session

Overflow Policy

When a push is attempted at the stack limit, the overflow policy from the operator's NestingPolicy dataclass determines the response.

Context Preservation on Push and Pop

Not all fields behave the same way at nesting boundaries. Some carry over into the inner structure; others reset. All are fully restored on pop.

Section 8.5

Nesting Declaration Extension

When the Layer 1 nesting code is 11 (bits 9 and 12 both set), a Nesting Declaration Extension byte follows Layer 1 immediately. This byte allows the sender to declare a precise depth up to 15 levels, select an overflow policy, and configure timeout behaviour — all in a single byte before the session begins handling records.

Depth Negotiation at Session Open

When sender and receiver declare different nesting limits, they negotiate at session open using the P1 signal slot. The agreed depth is the minimum of both declared values. Both sides must compute the same minimum — the protocol produces a single agreed depth with no ambiguity.

Section 10.2

Activation Scope Levels

The C0 Enhancement Module attaches at five scope levels. Each scope determines the activation mechanism, the overhead cost, and the lifetime of the enhancement context. Nesting depth interacts with scope: a session-scope activation permits nesting across all records in the session; a record-scope activation permits nesting only within the single record where the Signal Slot Presence byte is set.

Scope levels nest inside each other in strict containment order. A record-scope activation sits inside a batch-scope session. A batch-scope activation sits inside a session-scope session. The parser stack honours containment: a stack push from a record-level escalation is unwound before the record's batch context is closed.

Section 10.3

Industrial Strength Spectrum

The sub-protocol defines six levels of enhancement activation forming a progressive capability spectrum. Level 0 is standard BitPads with no C0 grammar active. Level 5 is full Telegraph Emulation with session-wide enhancement, deep nesting, and the complete Murray-Baudot lineage. Each level is a strict superset of the level below it.

Section 15.1 — 15.2

Overhead Reference

The overhead model is strictly pay-for-what-you-use. Every byte of overhead directly earns its cost by providing a signal, a confirmation, or a nesting boundary that the transmission explicitly needs. Transmissions that do not use a feature pay nothing for it.

Per-Element Costs

Overhead at Each Microservice Level

Section 15.3

Break-Even Analysis

At what transmission volume does enhancement grammar overhead pay for itself compared to application-layer alternatives?

The canonical comparison is between two approaches to conveying signal information alongside records: embedding a status byte at the application layer in every record (1 byte per record, always, in the payload), versus using the enhancement grammar at session scope (1 bit once at session open, then per-signal bytes only when a signal is actually sent).

Approach Comparison

Break-Even Table

Assume a typical record where one signal is sent per record on average (e.g., an STX at component start plus an ETX at end = 2 signal bytes per record when active). Enhancement grammar pays for itself when session overhead amortises below the per-record application-layer cost.

The formal break-even between session-scope (Level 4, 1 bit) and record-scope (Level 2, 1 byte per record) is at eight records. After eight records in a session, the session flag is always cheaper. Between batch-scope (Level 3, 1 bit per batch) and record-scope, break-even is also at eight records per batch. For any session with more than eight records requiring any enhancement, session-scope activation is the optimal choice.

Engineering Application

Deployment Guidance

Select the industrial strength level that matches the application's reliability requirements, hardware constraints, and signal density. The table below provides canonical starting points. Profiles may be tuned in the NestingPolicy dataclass and session profile JSON for production deployments.

NestingPolicy Configuration Reference

The NestingPolicy dataclass is stored in the session profile JSON. It is the single source of truth for nesting behaviour at runtime. All fields are operator-configurable.