DOCSIS, High-Split & 1.2/1.8 GHz Modernization

The problem modernization is solving

Legacy HFC plant was frequency-planned for a world where almost all demand ran downstream. The upstream got a narrow slice at the bottom of the band, and for twenty years that was the correct trade. It is no longer: upstream demand has grown faster than downstream in relative terms, and the return path is now the binding constraint on a lot of plant.

There are two structurally different responses. Segmentation reduces how many homes share the existing spectrum. Modernization changes the spectrum itself — moving the upstream boundary up, extending the downstream ceiling, or both. Segmentation is narrower and less disruptive; modernization is more expensive and reaches further.

Most operators need both, and the sequencing between them is a real engineering decision with real money attached. Segmenting a market in a way that ignores an imminent spectrum move can mean touching the same actives twice.

Mid-split, high-split, and what each one costs

Expanding the upstream means moving the boundary between the return path and the forward path. Two positions dominate current programs.

Mid-split raises the upstream ceiling to roughly 85 MHz. It is the smaller move: less downstream spectrum is surrendered, and the plant impact is more contained.

High-split raises the upstream ceiling to roughly 204 MHz, with the forward path starting above the transition region. It buys substantially more upstream and costs substantially more to deploy — a larger share of the actives are affected, and the downstream spectrum given up has to be recovered elsewhere, usually by higher-order modulation or by extending the upper band.

Both are defined in the CableLabs physical layer specifications; the DOCSIS 3.1 PHY specification is the authoritative reference for the frequency plans available. Our mid-split vs high-split explainer covers the comparison in more depth.

The engineering consequence is the same in both cases: the actives and passives along the cascade have to support the new plan, which means a design pass to identify what changes and a construction program to execute it. That is the work, not the frequency number.

DOCSIS 4.0 and the 1.2 / 1.8 GHz question

DOCSIS 4.0 defines two routes to substantially higher capacity, and they make different demands of the plant. Extended spectrum extends the usable band upward — commonly discussed in terms of 1.2 GHz and 1.8 GHz targets — while full duplex reuses spectrum in both directions and imposes its own topology requirements. The authoritative definitions are in the DOCSIS 4.0 PHY specification and the DOCSIS 4.0 MULPI specification; our DOCSIS 4.0 explainer walks through the trade-offs.

From an outside-plant engineering perspective, the important point is that these are architecture choices with long shadows. Extending the upper band means the plant has to pass the higher frequencies, which is a device and passive question across the whole cascade. Approaches that constrain topology change what segmentation has to achieve first.

BOGO's role is not to tell an operator which generation to deploy — that is a strategy decision that belongs with the operator. It is to translate the chosen architecture into a plant program: what has to be touched, in what order, at what scale, and how it sequences against the segmentation and fiber work already underway.

What a modernization program looks like in practice

The scale is what makes these programs difficult. A spectrum change is not deep engineering per node — it is moderate engineering multiplied by every node in several markets, executed against permitting queues and finite crew capacity.

1. Scoping and architecture confirmation. What plan, what markets, what sequence, what is already committed elsewhere in the capital plan.
2. Field verification. The device inventory has to be right before the program can be sized. Guessing at active counts across a market is how modernization budgets go wrong. See field services.
3. Design. Per-node identification of what changes, issued in the operator's standards. Covered on our HFC design page.
4. Make-ready and permitting where the program touches shared infrastructure, at make-ready and permitting.
5. Construction execution, sequenced and coordinated through construction coordination.
6. Activation, QA/QC and closeout, including the sweep and as-built work at activation and closeout.

The failure mode is almost never a single stage. It is the hand-offs: design finished for a market whose permitting has not started, or construction crews idle because permits landed out of order. Holding all seven stages is what BOGO is for.

What BOGO takes on

BOGO supports modernization programs from strategy translation through closeout:

• Network modernization planning and program strategy
• Node segmentation planning where it precedes or accompanies the spectrum move
• HLD and LLD support for the new plan
• Mid-split and high-split upgrade design
• DOCSIS upgrade program engineering
• 1.2 GHz / 1.8 GHz modernization engineering
• Program scheduling and resource coordination across markets
• Construction, activation and closeout coordination

Operational and interface standards published through the SCTE standards program govern much of the field practice these programs rely on, and packages are built to be compatible with the operator's own standards on top of that.