The Maryland Plant, the Digital Voice Switch, and the Forty-Year-Old Infrastructure Keeping You Talking to ATC

A new manufacturing facility in Maryland is producing digital voice switches for the FAA’s air traffic control modernization program, according to AVweb. The hardware will replace analog Voice Communications Switching Systems that have routed ATC communications across the U.S. for decades - in many facilities since original installations made in the 1980s. This is a significant operational milestone: a program moving from contract negotiation into active hardware production.

What a Voice Switch Actually Does

When a pilot transmits to a controller, the audio doesn’t travel directly from aircraft to headset. It arrives at a ground receiver, routes through landline or microwave links to the ATC facility, and passes through a Voice Communications Switching System (VCSS) before reaching the correct controller position. The controller’s response runs the same chain in reverse.

In a large en route center with dozens of active positions, the switching system is simultaneously handling hundreds of audio paths - sector handoffs, military coordination lines, supervisory override functions, and inter-facility communications. It is the router at the center of every exchange between pilot and controller.

Why the Current Infrastructure Is a Problem

The baseline VCSS architecture was specified and installed beginning in the 1980s. Many facilities have been running variants of the same core hardware - patched and kept alive through maintenance contracts - for 30 to 40 years.

That alone isn’t unusual in aviation infrastructure. What has changed is the environment those systems operate in. Over the past two decades, the FAA has fundamentally modernized the data side of ATC: the Standard Terminal Automation Replacement System (STARS) updated TRACON workstations, En Route Automation Modernization (ERAM) replaced the HOST computer system that ran en route centers since the 1970s, and ADS-B and Data Communications (CPDLC) have been deployed across the national airspace. The voice switching infrastructure largely did not come along for that transformation.

The result is a national airspace where the data infrastructure is modern and the voice infrastructure is not.

How This Asymmetry Affects Controllers and Pilots

At a facility running digital STARS workstations alongside a legacy analog voice switch, a controller sees a rich, modern data picture on screen - track data, flight plan information, conflict alerting - but the voice system handling their radio doesn’t communicate with any of it. Coordination that could happen automatically in a fully integrated system requires the controller to do it manually. That is additional cognitive load in an environment where managing cognitive load is a safety-critical concern.

For pilots, the limitations show up as audio quality. Analog switching systems introduce noise over long-haul routing chains, and analog compression used to fit audio through narrow-band circuits makes some transmissions noticeably muddier. If certain en route center airspace consistently sounds cleaner on the frequency than others, part of that difference is the age and design of the switching infrastructure behind it.

There is also a redundancy gap. Modern digital switching systems are designed with multiple independent failure paths - if one routing module fails, traffic shifts to another in milliseconds. Analog systems of a certain vintage can have single points of failure that require sector consolidation while maintenance responds. Multiple fallback layers exist, but digital architecture is more resilient by design from the ground up.

What Digital Voice Switching Actually Means for ATC

The replacement hardware uses Voice over Internet Protocol (VoIP) switching - related to commercial telephony protocols, but purpose-built for aviation. Latency requirements are strict, with audio delay measured in tens of milliseconds. Reliability standards require continuous operation through power anomalies and equipment failures that would take a commercial telecommunications system offline. Certification is entirely distinct from the commercial world.

The security architecture is also baked in from the specification stage. Moving ATC voice to an IP architecture introduces potential exposure that physical analog hardware does not have. The FAA, working alongside defense and intelligence stakeholders, has spent years defining what secure VoIP for ATC means in practice - network segmentation, encryption standards, air gapping where appropriate. This is not voice running over consumer broadband.

The Scale of What This Program Requires

The FAA operates more than 300 ATC facilities across the national airspace system, including 21 en route centers, more than 100 TRACONs, and hundreds of airport traffic control towers. Not every facility requires the same scale of switching infrastructure, but the combined deployment is an enormous manufacturing and integration effort.

The Maryland plant represents a response to that supply chain requirement. Equipment handling ATC communications must meet FAA technical specifications distinct from commercial electronics standards - environmental hardening for temperature, vibration, and humidity; electromagnetic compatibility for facilities coexisting with active RF environments; power supply redundancy exceeding commercial settings. This is not a general-purpose contract electronics shop. It is a specialized industrial capability that the program specifically requires.

Why This Has Taken So Long

The FAA has been working toward ATC modernization in one form or another for more than four decades. The Advanced Automation System program, begun in the late 1970s, aimed to replace essentially everything in the en route ATC environment simultaneously. By the early 1990s, it had consumed billions of dollars and missed nearly every major milestone. The Government Accountability Office described it as one of the most expensive automated system development failures in government history. The program was restructured, pieces were salvaged, and the current baseline en route infrastructure emerged from what remained.

That history produced lasting caution at the FAA and in Congress. Large-scale ATC technology programs now proceed in more contained phases, with independent oversight, and with a skepticism toward ambitious timelines that the track record warrants.

NextGen, authorized by Congress in 2007, has delivered some of its headline capabilities - ADS-B out equipage is essentially universal in the IFR fleet, performance-based navigation procedures have expanded at airports across the country, and Data Comm is operational at most major en route centers. Voice switching modernization has been the longer road within that program, stretched significantly past original projections by multiple acquisition cycles, program restructures, and funding fluctuations.

A manufacturing facility actively producing hardware for field deployment is one of the more concrete signs that a program has crossed from planning into production.

What This Means for Pilots Now - and Later

Near-term, the impact will not be perceptible in the cockpit. These transitions happen facility by facility, sector by sector, over years, and are designed to be operationally transparent. You will not pick up a Center frequency one morning and hear a different sounding controller because a new switch came online overnight.

The medium and long-term implications are more significant. Communications reliability at cruise altitude - particularly in high-density sectors and areas with challenging terrain and marginal transmitter coverage - is directly connected to the quality and architecture of voice switching infrastructure. Audio clarity, redundancy, and the integration of voice with data systems all improve as legacy hardware is replaced.

There are capabilities ahead that depend on this transition as a foundation. Voice recording and replay quality improves dramatically with digital infrastructure, benefiting both accident investigation and controller training. Voice recognition tools - automatically transcribing controller-pilot communications and cross-referencing them with traffic data - require digital switching as a prerequisite. That capability is not widely deployed today, but the architecture decisions being made now will determine whether it becomes achievable in the next decade.

Digital voice infrastructure also enables remote tower operations. Countries including Sweden and Norway have operational remote tower facilities where controllers manage airport traffic from a location miles from the actual runway, using camera systems and digital communications infrastructure in place of a physical cab. The FAA has been conducting remote tower evaluations in the United States. None of that scales operationally with legacy analog voice switching in the loop.

The staffing picture connects here as well. The FAA has been operating below target controller staffing levels at a significant number of facilities, and the training pipeline to address that shortage takes years to work through. Infrastructure that reduces workload on low-complexity coordination creates space for human judgment on the decisions that actually require it.

Key Takeaways

• A Maryland manufacturing facility is now producing digital voice switches for the FAA, marking the transition from program planning into active hardware production.
• ATC voice switching infrastructure at many facilities dates to the 1980s - analog systems running alongside modern digital data displays, creating an integration gap with real operational consequences.
• The FAA operates 300+ ATC facilities, making this a massive deployment requiring a dedicated, FAA-certified domestic manufacturing capability.
• The underlying technology is aviation-grade VoIP - purpose-built for strict latency, redundancy, and security requirements that commercial telephony does not meet.
• Digital voice infrastructure is the prerequisite for future capabilities including voice recognition integration, remote tower operations, and tighter coordination between voice and data systems.