Global Title translation formats and prefix handling in SS7 networks
In SS7 and SIGTRAN networks, Global Title Translation (GTT) is the mechanism that lets a signaling point resolve a destination based on a logical address rather than a hard-coded point code. Every SCCP-routed message — from a mobile location update to an SMS delivery — depends on translation tables somewhere in the path to map digits, prefixes, and nature-of-address indicators into a usable destination.
The format of those addresses varies between regions and operators. In Australia, where Telstra and Optus maintain separate but interconnected core networks, engineers often work with mixed E.164 and E.214 formats at the same STPs. That mixed reality makes prefix handling one of the most error-prone parts of any GTT configuration.
A solid understanding of how prefixes are stripped, preserved, or replaced is essential for anyone managing signaling in modern hybrid TDM-IP environments. This piece walks through the formats, the translation rules, and the common traps that catch even experienced teams off guard.
The role of global title translation
GTT exists because point codes alone cannot describe every destination an SCCP message might need to reach. A point code identifies a node; a Global Title identifies a subscriber, a service centre, or a logical entity. The STP must convert that logical address into a point code plus subsystem number so the message can be forwarded.
When an SMSC in Sydney receives a mobile-originated short message destined for a number in Melbourne, it cannot simply forward the digits to the destination MSC. The SCCP layer must look up the digits against translation tables, decide whether to strip the leading "0", and determine which point code owns that number range. The result of that lookup is what gets passed to the MTP3 label routing logic.
Without correctly configured GTT, messages loop, get dropped, or arrive at the wrong node. In a country as geographically spread as Australia, where traffic regularly crosses state borders and sometimes hops through international gateways, robust translation is a basic operational requirement.
Common GT formats and numbering plans
| GT Format | Typical Use | Nature of Address | Prefix Behaviour |
|---|---|---|---|
| E.164 | Mobile and fixed subscribers | Subscriber number | Leading "0" stripped or preserved per carrier rule |
| E.214 | Mobile roaming (MGT) | IMSI-derived | MCC stripped, converted to E.164 |
| E.212 | IMSI-based routing | MCC + MNC + MSIN | Routed to HLR based on MNC match |
| E.116 | Network signalling point addressing | Point code digits | No prefix manipulation |
| National-specific | Local carrier variants | Mixed | Configured per operator |
Each format has its own way of expressing the address. E.214, for instance, is derived from the IMSI by replacing the MCC with a country code and converting the digits into a routable E.164 number. E.212 stays close to the IMSI structure and is mostly used for HLR lookups during attach procedures.
When a Telstra engineer configures an STP for inbound roamers from a Singapore carrier, the translation table has to recognise the E.214-derived digits, map them to the correct E.164 prefix range, and then forward to the appropriate HLR. A mismatch at any of those steps will surface as a failed attach or a missing SMS.
Prefix stripping, insertion, and preservation
Prefix handling sits at the heart of every translation. Three operations dominate the rule sets: stripping the leading digits, inserting new digits before forwarding, and preserving the original prefix unchanged.
Stripping is the most common. Australian fixed-line numbers typically arrive at the STP with a leading "0" that must be removed before the message is routed to the destination exchange. Insertion is the opposite — a translated number may need a new prefix appended, such as when traffic is re-routed into a test environment or a partner network. Preservation is rarer and usually applies to inter-carrier handoffs where the receiving party expects the full dialled string.
A misconfigured prefix rule can quietly double-charge billing systems or send calls to entirely wrong regions. Local training providers in Melbourne often run scenario exercises around these failure modes because they are simple to demonstrate and disproportionately expensive in production.
Translation tables and match logic
STP translation tables are organised hierarchically. A first match on the Global Title Indicator selects the family of rules; a second match on the translation type narrows further; and a final match on the digit string determines the action. Each match can specify whether to strip a number of leading digits, what to insert in their place, which point code to forward to, and which subsystem number to use.
Operators in Australia often split translation tables between fixed-line, mobile, and international gateways. That separation makes troubleshooting easier but also means that a change in one table does not automatically propagate to others. ACMA's numbering plan documents define which number ranges belong to which carrier, and engineering teams use those documents as the authoritative source when building prefix rules.
When label routing replaces GT routing
Not every SCCP message needs Global Title translation. Some traffic uses point-code-and-subsystem routing directly, also called label routing. In those cases, the MTP3 header carries enough information for the STP to forward without consulting translation tables at all.
This is faster but less flexible. A discussion of how the STP decides between the two paths is covered well in material that explains STP message routing, including the criteria for choosing label routing over GT-based routing. The choice has consequences for resilience, because label-routed traffic cannot benefit from the same failover logic that translation tables provide.
Common configuration pitfalls
Several mistakes show up repeatedly in field escalations. Translating on the wrong translation type is a classic — many operators have both GTT and STP-internal variants, and mixing them up produces subtle routing errors. Forgetting to update prefix rules after a number-range reallocation by ACMA is another, particularly after the rollout of the NBN reshuffled fixed-line ranges.
Engineers sometimes also forget that nature-of-address indicators must align between sender and translation rule. A "national" indicator arriving at a translation rule configured for "international" will not match, even if the digits themselves are correct.
Reading the output correctly
When GTT is misbehaving, the first place to look is the translation output: what digits did the STP actually forward, and to which point code. Comparing that output against the original Global Title reveals whether stripping, insertion, or routing is at fault. Local signalling teams often run a quick capture on a quiet Saturday arvo when traffic is low, before pushing rule changes into production.
The thing to remember is that GTT is deterministic. Given the same input digits and the same table, the same output will occur every time. That determinism is what makes prefix handling fixable — once the table is correct, every downstream node behaves predictably.
