Hotel POTS Replacement: Keeping Room Phones When Copper Ends

The fastest way to modernise a hotel's phone system is usually to leave the in-room handsets where they are and replace everything behind them. Room phones sit inside finished walls, carry PMS signalling that nobody wants to re-certify, and are among the cheapest devices in the building to keep forever. New Rock's official solution library already treats Hotel POTS Replacement as a named scenario, and the supporting hardware is a combination of FXS-capable IP-PBX platforms and FXS/FXO gateways rather than a special-purpose appliance.

Definition: Hotel POTS replacement means keeping analog room phones (POTS handsets) in service while the switching, call routing, recording and PSTN connectivity behind them move to IP. The endpoint stays; the system behind it becomes SIP-based. In New Rock's architecture this is handled by the built-in FXS ports on the OM series IP-PBX, by MX series VoIP gateways where analog endpoints sit on floors the PBX does not reach, and by MA50-LS local survivability where in-room calling must survive a WAN outage.

Why Room Phones Resist Full IP Replacement

There are three separate obstacles, and they are worth separating because each one has a different answer. Only one of them is technical.

Why the analogue layer tends to stay:

• Cabling and floor boxes. Analog handsets run on a single twisted pair from a riser or IDF to each room. Re-cabling a guest room to run Ethernet means opening walls, replacing jacks and faceplates, and working in occupied rooms.
• PMS signalling dependencies. Room status, do-not-disturb and message-waiting often reach the handset over the same pair as voice. That integration is usually the reason a room phone is treated as fixed rather than as a replaceable endpoint.
• The handset itself. A POTS handset needs no power, no PoE switch port, no VLAN, no firmware policy. From a maintenance point of view it is an asset that costs almost nothing to leave alone.

There is also the commercial reality. A full IP refresh touches every guest room, and each room touched is a room that may be taken out of inventory while the work happens. A gateway-led refresh touches the riser, the IDF and the comms room for most of the building, and the guest room only where a handset must be swapped.

The Analogue Layer: FXS Port Planning

FXS is the port type that presents a line to an analog phone. FXO is the port type that presents itself as a phone to something else, such as an analogue PSTN line or the line side of a legacy PBX. Getting these two the wrong way round is the most common mistake in POTS migration planning, and it is worth settling before any hardware is quoted.

Count every analog endpoint, not just guest rooms

The guest room is the visible population, but it is rarely the whole one. Walk the building and count every device that terminates an analog line: room handsets, front desk stations, housekeeping and maintenance handsets, the night audit position, fax points, parking and gate intercoms, elevator intercoms, back-of-house telephones and any emergency or lift-entrance intercom that has to survive a network fault.

New Rock's solution library groups the POTS Replacement work into three scenario types, and the difference between them is largely about how many analog endpoints are in play and where they sit relative to the IP network. A property with a handful of analog devices in a single comms room is a different job from a tower where every floor has a bank of handsets behind a riser.

Gateway port tiers:

The MX8G gateway carries 8 FXS/FXO ports behind one WAN port and one PC port, both rated at 10/100/1000Mbps, which makes it a practical unit for a floor landing where there is a single uplink and no room for a stack of small boxes. The MX60E-G gateway scales the same idea to a building: up to 48 FXS/FXO ports, SIP registration failover, load balancing across registrations and PSTN failover, with optional 1+1 redundancy on the network port and the power supply.

For the switch side, the built-in analogue ports on the OM series are the deciding factor, and the published capacities are set out below.

The OM80E carries 48 built-in FXS/FXO ports and 90 concurrent calls, which covers a substantial number of analog endpoints from a single unit. The OM20G is described on the official site as an all-in-one IP telephony system with up to 4 built-in FXS/FXO ports, supporting PSTN, analog phones, SIP trunks, SIP phones and IMS access, and it is the smallest box that can terminate analogue handsets without a separate gateway.

Room Phone Features PMS Integration Depends On

Keeping analog handsets only works if the interface the property management system expects still works. Before committing to the approach, read the room phone as an interface, not as a telephone.

What the PMS actually needs from the room phone

• Hook switch status. The most common PMS signal: on-hook, off-hook and on-hook flash tell the PMS whether the guest is in the room, has picked up, or has put the handset down mid-call. A hybrid interface that drops this signal breaks room status and call accounting at the same time.
• Message waiting. Voicemail and switchboard messages are announced by lamp and tone on the handset. If the new system cannot signal message waiting on the analog port, the guest never knows a message exists.
• Caller ID presentation. Guests see the number of the incoming call and the name the PMS has associated with the room. This is delivered by the analogue line, so it depends on the port handing the caller ID through rather than stripping it.
• Do-not-disturb and privacy. Set from the room phone on many properties, so the DND state has to survive a port reset and a gateway power cycle.

Where the property runs a third-party PMS, the integration is usually handled on the IP side rather than in the room. PMSI is the middleware for this job: it connects the OM system to third-party hotel management systems and handles the information exchange between the two. Its presence is what makes an analog-heavy property a configuration task rather than an integration project.

Recording and audit. Call recording on an analog estate is anchored on the system, because the voice path terminates in an FXS port rather than on the handset. Recordings can be stored on the IP-PBX itself, and RMS recording management software collects, stores, queries, plays back and manages the recordings produced by OM series IP-PBX systems and MX series gateways centrally. Recording Agent, which runs on Windows, collects recordings from the OM system and saves them as MP3 files.

New Rock's POTS Replacement Building Blocks

Three components cover most hotel deployments, and they solve different failure modes.

Local survivability for hotel and apartment intercom

The MA50-LS local survivability branch exchange exists for the case where the link to the central server is not a controlled thing. It supports FXO 2-8 ports and SIP user registration of 50-200, and when the headquarters or cloud server connection drops it keeps calls running locally, switching back once the connection is restored. That is the relevant behaviour for a hotel where a WAN cut would otherwise take the in-room phones and the intercom dead. The official scenario list for the MA50-LS local survivability branch exchange includes Hotel POTS Replacement, and the same local-survival approach appears in New Rock's case material for emergency intercom in a Singapore apartment, where the OM50G is used for emergency intercom service.

For a hotel, the practical split is usually this: MA50-LS sits on the site to hold the analog endpoints and the emergency calling path, while the OM platform handles the features. Add a gateway with PSTN failover and the property keeps a working phone system through a WAN fault, a local server fault or an uplink failure.

Legacy PBX and carrier interconnect. Most refreshes are not greenfield. The existing PBX often still carries PSTN trunks or ISDN PRI that nobody wants to replace in the same project. The OM1000-TE media gateway converts between IMS/SIP and ISDN PRI (30B+D), with a maximum of 120 concurrent calls, up to 40K busy-hour call attempts, 500 routing and number-transformation rules, 512 registered SIP trunk accounts, dual network port and dual power redundancy, and 6000MIPS of DSP processing capacity. It is the clean way to keep the existing carrier contract while the internal estate moves to IP.

Phased Replacement Without Closing the Hotel

The order of operations matters more than the hardware choice, because most of the risk in a hotel project is operational rather than technical.

The phased cutover sequence

1. Inventory the analogue estate. Produce a list of every analog endpoint with its location, its PMS function and the floor it sits on. Record the devices as well as the lines: Finder is a PC tool that discovers the IP addresses of New Rock devices on the LAN, which is a fast way to find the FXO endpoints already deployed on the network before anything is decommissioned.
2. Segment the building. Group endpoints by riser or floor and by PMS function, so a phase is a coherent block rather than a scatter of rooms.
3. Land the IP core first. Install the OM platform and the gateways, register the gateways to it, and confirm that a call between two analog handsets completes before any existing service is touched.
4. Add survivability. Deploy MA50-LS where the property needs calling to survive a WAN or server outage, then test it by pulling the uplink rather than by trusting a configuration screen.
5. Migrate one block at a time. Cut the legacy ports over group by group, with the legacy system still live behind the new one. Because MX60E-G supports SIP registration failover, load balancing and PSTN failover, a partial cutover does not have to be a partial outage.
6. Verify the PMS interface before the old system is removed. Room status, message waiting, caller ID and call accounting are the four things to check on a sample of rooms per floor, and again after a gateway power cycle to confirm the states are not held in volatile memory.
7. Decommission last. Retire the legacy PBX and its cabling only after a full trading period with the new arrangement in place, and only after the PMS integration is signed off.

Cost Model: Gateway Approach vs Full IP

The comparison is rarely a single line, because the two approaches spend money on different things.

Two cost drivers are regularly missed. First, room downtime: a refresh that takes occupied rooms out of service has a revenue cost that does not appear on any equipment list. Second, the cost of leaving a legacy system alive alongside the new one for longer than planned. A phased gateway migration deliberately runs both in parallel for a period, and that period should be written down and owned rather than left open-ended.

There is also a hybrid that suits many mid-size properties: IP handsets in the lobby, restaurant, conference and back-office areas where new endpoints are cheap to deploy and features matter, with FXS ports and gateways carrying the guest rooms. The OM20G all-in-one description, with up to 4 built-in FXS/FXO ports, supports PSTN, analog phones, SIP trunks, SIP phones and IMS access in one unit, which is what makes the mixed estate practical rather than theoretical.

Hotel POTS Replacement FAQ

Can we keep our existing room phones completely unchanged?

Usually yes, which is the point of the approach. The handsets keep working when they terminate on an FXS port instead of the old PBX port. What has to be checked on a sample of rooms is the signal set the PMS depends on: hook switch status, message waiting, caller ID and DND.

Do we need to touch the room cabling?

Normally not beyond the riser and the floor landing. If the existing pair runs from the riser to the room jack, the analogue port stays where it is and only the switch at the other end changes. A site survey should still verify that the pair is continuous and correctly terminated, because aged copper is a common reason a port fails after the changeover.

What happens to our PSTN lines during the migration?

They can be carried over rather than re-provisioned. The OM1000-TE media gateway handles IMS/SIP and ISDN PRI (30B+D) conversion for existing carrier arrangements, and the MX gateways provide PSTN failover so calls still route if the IP side has a problem.

Can we keep calling if the internet connection goes down?

Yes, with the right component in the design. The MA50-LS local survivability branch exchange keeps calls running locally when the headquarters or cloud server connection drops, and switches back when it is restored. That capability should be tested by pulling the link during commissioning.

How long does the whole project take?

The honest answer is that it depends on the size of the analogue estate and how many floors can be cut over in one phase. The part that sets the pace is usually the PMS verification, not the telephony. Plan the phases against blocks of rooms and keep the legacy system live until verification passes.

What does an operator need to run this day to day?

OM series platforms and MX series gateways are managed together through UMS unified management software, which provides unified management and remote operations support for both ranges. That matters in a multi-property group, where the alternative is walking each comms room.

Conclusion

Hotel POTS replacement works when the room phone is treated as an interface to be preserved rather than as a device to be replaced. The sequence that gets a hotel through it is straightforward: count every analog endpoint including the intercoms, size FXS capacity against the published port tables, put a survivability device on the WAN-dependent path, and migrate one coherent block at a time with the legacy system still live behind it.

Before committing to a design, agree three things on paper: the full analogue endpoint count, the PMS signal list that must keep working, and the failover behaviour the property expects during a WAN or server fault. Those three answers determine the hardware, and everything else is configuration.