P25 Phase 2: TDMA

Key takeaways P25 Phase 2 doubles voice capacity by putting two-slot TDMA on the traffic channels: one 12.5 kHz frequency now carries two simultaneous calls — roughly 6.25 kHz equivalent each. Traffic uses a phase-modulated scheme (H-DQPSK inbound, H-CPM outbound), not C4FM, and the AMBE+2 half-rate vocoder instead of IMBE. Crucially, the control channel usually stays a Phase 1 C4FM channel — so the signaling is Phase 1 and only the traffic is TDMA. For a decoder, the new job is tracking two time slots per frequency, recovering slot timing from the ISCH.

P25 Phase 1 gave each call its own 12.5 kHz frequency. That is clean but spectrum-hungry, and busy metropolitan systems ran out of channels. Phase 2 is the answer: keep the same frequencies, but fit two calls on each traffic channel using TDMA.

The capacity idea: two calls, one frequency

In Phase 2 a traffic channel is divided in time into two slots that alternate in rapid turns. Two independent calls share the frequency — one in each slot — so a 12.5 kHz channel does the work of two. The standard often describes this as 6.25 kHz equivalent per call: not that the channel is physically narrower, but that each conversation consumes half the channel’s time, achieving the same spectral efficiency as a 6.25 kHz channel would.

Two independent calls (A and B) interleave in alternating time slots on a single Phase 2 frequency. The ISCH sync between bursts is what lets a decoder recover slot timing and demultiplex them.

The modulation changes — but not on the control channel

To fit a clean two-slot burst structure, Phase 2 traffic channels drop C4FM in favour of a phase-modulated scheme: H-DQPSK (Harmonized Differential QPSK) on the inbound link from subscriber to system, and H-CPM (Harmonized Continuous Phase Modulation) on the outbound link from system to subscriber. These are tuned for the precise timing a shared TDMA channel demands.

The important subtlety: in nearly all deployments the control channel stays a Phase 1 C4FM channel. A Phase 2 system is therefore Phase 1 signaling with Phase 2 traffic — the same TSBK grants you already know, but the channels they point to are TDMA. This is also why Phase 2 systems remain compatible with the wider P25 world and why a follower’s control-channel logic is essentially unchanged.

Half-rate voice: AMBE+2

Two calls in the bits of one means each call gets half the data rate, so Phase 2 swaps Phase 1’s IMBE for the AMBE+2 half-rate vocoder. The same family of math, but compressing voice into roughly half the bits — exactly what a single TDMA slot can carry. Unlike IMBE, whose core patents have expired, AMBE+2 is patent-encumbered; GopherTrunk ships a pure-Go AMBE+2 decoder by default but the patent risk falls on the deployer (see Vocoders). Phase 2 grants map to the ambe2 vocoder in GopherTrunk’s protocol table.

Aspect	Phase 1	Phase 2
Traffic access	FDMA	2-slot TDMA
Calls per 12.5 kHz	1	2
Spectral efficiency	12.5 kHz/call	≈6.25 kHz/call equiv.
Traffic modulation	C4FM	H-DQPSK / H-CPM
Control channel	C4FM	usually C4FM (Phase 1)
Vocoder	IMBE (full-rate)	AMBE+2 (half-rate)

Slot structure and ISCH

A Phase 2 frequency carries a repeating sequence of bursts, alternating between the two slots. To make sense of the stream a receiver must know which slot it is looking at and where each burst begins. That synchronization comes from the ISCH (Inter-Slot Signaling CHannel) embedded in the burst structure: it provides the sync and slot-identification markers a decoder uses to align to the frame, distinguish slot 1 from slot 2, and carry per-slot signaling. Lose ISCH sync and the two calls smear into noise; hold it and each slot becomes a clean, independent voice stream.

What changes for a decoder

For GopherTrunk the control-channel side barely changes — it is still reading Phase 1 TSBK grants. The traffic side is where Phase 2 demands more:

Demodulate a different scheme. The traffic channel is H-DQPSK/H-CPM, not C4FM, so the symbol recovery differs from Phase 1. You can still inspect recovered symbols on the Symbol scope.
Recover slot timing. Lock the ISCH, find the burst boundaries, and identify which slot is which.
Demultiplex two calls. Treat each slot as its own voice stream, decoding AMBE+2 separately per slot, because a single grant’s frequency may host a second, unrelated call in the other slot.
Map grants to slots. A Phase 2 grant references not just a frequency but a slot, so the follower tunes the frequency and then selects the correct time slot.

Quick check: how does P25 Phase 2 double voice capacity?

Recap

P25 Phase 2 adds two-slot TDMA on the traffic channels, doubling capacity to two calls per 12.5 kHz (≈6.25 kHz equivalent each).
Traffic uses H-DQPSK/H-CPM phase modulation; the control channel usually stays Phase 1 C4FM, so signaling is unchanged.
Voice uses the AMBE+2 half-rate vocoder instead of IMBE.
The ISCH provides the sync and slot identification a decoder needs to demultiplex the two slots.
For a decoder the new work is tracking two time slots per frequency and decoding each as its own stream.

Next we cross to the other dominant digital standard, with its own three-tier story: DMR Tier II & Tier III.

Frequently asked questions

What does P25 Phase 2 add over Phase 1?

Phase 2 puts two-slot TDMA on the traffic channels, so a single 12.5 kHz voice channel carries two simultaneous calls — about 6.25 kHz of spectrum per call. It doubles voice capacity without needing new frequencies. The control channel usually stays a Phase 1 C4FM channel, so a system is effectively Phase 1 signaling with Phase 2 traffic.

What modulation does P25 Phase 2 use?

Phase 2 traffic channels use a phase-modulated scheme — H-DQPSK on the inbound (subscriber-to-system) link and H-CPM on the outbound link — rather than the C4FM of Phase 1. These are engineered for the tight timing a two-slot TDMA channel needs. The control channel typically remains C4FM.

What vocoder does P25 Phase 2 use?

Phase 2 uses the AMBE+2 half-rate vocoder instead of Phase 1’s IMBE. Half-rate voice fits two conversations into the bits of one Phase 1 channel, which is what makes the two-slot TDMA capacity gain possible. AMBE+2 is patent-encumbered, unlike the now-expired IMBE patents.

Why must a Phase 2 decoder track two time slots?

Because a single Phase 2 traffic frequency interleaves two independent calls in alternating time slots. A decoder has to recover slot timing from the ISCH and demultiplex the bursts, treating each slot as its own voice stream. Locking the frequency is not enough — you must lock the slot.