Why Network Reliability Matters in Large Parking Garages and Underground EV Charging

---

How structural interference, not just electrical capacity, determines whether garage EV charging actually works

Property owners planning EV charging infrastructure spend most of their diligence on electrical capacity: panel headroom, transformer sizing, demand charges. Network connectivity rarely makes the checklist, yet it is one of the most common reasons a fully powered, correctly installed charger simply does not work. Large parking structures and underground levels are, from a radio-frequency standpoint, some of the most hostile environments a connected device can be placed in. Reinforced concrete, structural steel, and multiple floors of overhead mass block cellular and Wi-Fi signal so thoroughly that a charger with perfect electrical service can still show as offline, refuse to authorize a driver, or silently fail to report a completed session. This article explains why that happens, what it costs a property when it does, and how to design connectivity into a charging deployment rather than discovering the problem after the ribbon-cutting.

Every networked EV charger is, functionally, a small computer that depends on a live connection to a cloud-based charging station management system, or CSMS, to do most of what makes it 'smart.' Authorization, live pricing, session billing, remote diagnostics, firmware updates, and coordinated load management all travel over that connection. When a charger sits in an open surface lot, a nearby cell tower or the building's own Wi-Fi network usually keeps that link stable without anyone thinking about it. Structured and underground parking removes that assumption entirely.

Why Concrete and Steel Defeat Cellular Signal

Reinforced concrete is dense and electrically conductive because of the steel rebar embedded throughout it, and that combination absorbs and reflects radio waves rather than letting them pass through. Industry testing on cellular propagation through building materials has found that roughly every six inches of concrete strips about 20 decibels from a cellular signal, and parking structures rarely stop at one layer. Add steel beams, precast concrete decking, block walls, and the metal ductwork and conduit that run through every level, and the cumulative signal loss can push a perfectly strong outdoor signal down to nothing measurable indoors. The structure is, in effect, acting as a Faraday cage around anything installed inside it.

Underground Levels Compound the Problem

Below-grade levels add the mass of the earth and the floor slabs above them on top of every other source of attenuation, which is why second-level basements in dense urban buildings routinely show zero bars on every carrier. Wi-Fi extended from a building's main network faces a related but distinct problem: standard access points have limited range to begin with, and the same concrete and steel that block cellular signal also degrade Wi-Fi, so a charger even one level below the nearest access point may fall outside usable range even though the building's internet service itself is fine. Both problems point to the same underlying reality: connectivity in structured parking has to be engineered deliberately, not assumed to arrive along with the building's existing internet or cell service.

A charger that loses its connection to the CSMS does not necessarily stop delivering power, but it does lose most of the functionality that separates a 'smart' charger from a dumb outlet, and the exact behavior depends on how the station and the network operator have configured OCPP, the open protocol nearly all commercial charging equipment uses to talk to its management platform.

The OCPP Offline Authorization Gap

Under OCPP, a charger can be configured to hold a local authorization list or authorization cache so it can still recognize known RFID cards or app credentials while disconnected, and a configuration key called LocalAuthorizeOffline determines whether it is even allowed to start a session on that basis. Many commercial and multifamily deployments run without local authorization enabled by default, because the network operator wants every session verified centrally for billing accuracy. In that configuration, a connectivity outage does not just interrupt reporting; it prevents new sessions from starting at all. Where offline authorization is enabled, the charger can begin a session locally, but it must queue the transaction data and transmit it once the connection is restored, and the central system has no way to distinguish that queued session from real-time activity other than by inference from timestamps. Either way, an extended outage leaves a property with either stranded drivers or a growing backlog of unverified transactions.

Lost Revenue, Lost Data, Lost Trust

For a property monetizing its charging infrastructure, a disconnected charger is a station that cannot bill, cannot report utilization to a dashboard, and cannot participate in coordinated load management, which matters enormously on sites using smart load management to avoid expensive electrical upgrades. Beyond the immediate revenue and data gaps, connectivity failures are also disproportionately responsible for driver-facing failures. Independent analysis of charging network failure modes attributes roughly a quarter of downtime incidents to network and software communication problems, a category that includes the frustrating case where an app shows a charger as available but the station refuses to actually start a session. For a tenant, employee, or guest standing at a dead charger, the distinction between an electrical fault and a connectivity fault is invisible; both simply read as 'the charger doesn't work.'

Charging reliability has become a measurable, publicly tracked metric rather than a marketing claim, and the gap between what networks report and what drivers actually experience is well documented.

Reported Uptime vs. Real-World Charging Success

Charging networks commonly publish uptime figures in the high 90s, but independent research analyzing more than one hundred thousand charging sessions found the real first-time charging success rate across a large sample of public chargers was closer to 71 percent, meaning close to one in three charging attempts failed outright. That study is vendor-sourced and worth flagging for independent verification, but it is consistent with third-party monitoring data: national average reliability scores across U.S. DC fast charging infrastructure sat at roughly 93.5 out of 100 in the first quarter of 2026, with meaningful state-by-state variation and some states still well below that average. Academic case-study data covering a single site over several years found a similar pattern, with connectivity-related 'unreachable time' overtaking hardware faults as the leading driver of measured downtime by 2023. None of these figures isolate structured parking specifically, but the physical connectivity challenges of garages make it a reasonable inference that sites with poor signal skew toward the lower end of these ranges.

Regulatory and Liability Exposure for Property Owners

Reliability is no longer a purely commercial concern. Federally funded charging stations under the National Electric Vehicle Infrastructure program must meet a minimum average annual uptime standard, and the underlying federal rule specifically requires that a charger remain functional and able to complete a charging session even when its connection to the charging network is temporarily disrupted. While that requirement applies directly only to NEVI-funded highway corridor stations, it reflects the direction regulators are heading, and it is a useful benchmark for any property owner weighing connectivity investment against a get-lucky approach to signal strength. For multifamily and commercial properties, the more immediate exposure is reputational and contractual: a charging amenity marketed to tenants or guests that fails at a materially higher rate than the property's other systems erodes the trust that justified adding it in the first place.

There is no single correct connectivity solution for every garage; the right choice depends on signal testing at the actual charger locations, the size of the structure, and how much redundancy the property is willing to invest in. Four approaches account for most real-world deployments.

Matching the Solution to the Site

A cellular signal booster or distributed antenna system captures the strongest available carrier signal from outside the structure and rebroadcasts it to indoor antennas placed near the charging equipment, which is often the fastest fix for an existing garage with otherwise decent outdoor coverage. A dedicated Wi-Fi network, run on access points installed specifically for the charging equipment rather than extended from the building's general-purpose network, avoids cellular limitations entirely but requires either new cabling back to a telecommunications room or mesh hardware capable of bridging multiple garage levels. A hardwired Ethernet backhaul, pulled directly from the building's network closet to each charger location or to a local switch serving several chargers, is the most reliable option where conduit runs are feasible during construction or major retrofit, since it removes wireless propagation from the equation altogether. Multi-carrier cellular connectivity using chargers with dual-SIM or multi-network modules adds a layer of redundancy by automatically switching to whichever carrier has the strongest signal at that location, which matters because coverage from different carriers can vary significantly within the same structure.

Redundancy as the Real Standard

The properties with the fewest connectivity-driven outages typically do not rely on a single method. A common pattern pairs a hardwired or boosted primary connection with cellular failover, so that a fiber cut, a booster power failure, or a carrier outage does not take every charger on a level offline simultaneously. For sites where even brief connectivity gaps are unacceptable, an offline-first charging architecture — one designed so authorization and session logic run locally at the charger and simply sync to the cloud opportunistically — removes the dependency on a live connection entirely, at the cost of a more specialized and typically more expensive charging platform.

Cost ranges are illustrative planning estimates based on vendor-published figures for signal amplification and network hardware; actual pricing depends on structure size, floor count, and existing telecom access, and should be confirmed with a site-specific quote.

Connectivity is far cheaper to solve during planning than to retrofit after chargers are installed and complaints start arriving, which is why it belongs in the same due-diligence phase as electrical capacity and charger sizing rather than treated as an afterthought.

Site Assessment Before Charger Selection

A proper connectivity assessment tests actual signal strength at the specific locations where chargers will be installed, not just at the garage entrance or a random spot on each level, since attenuation varies significantly by proximity to structural steel, distance from the nearest exterior wall, and depth below grade. Best practice is to conduct that walkthrough with the same measurement tools an installer will use to verify a booster or access point afterward, so the pre- and post-installation readings are directly comparable. Where a site tests poorly across every carrier and there is no existing building Wi-Fi within reach, that finding should shape which connectivity approach gets budgeted, not get discovered after chargers are already mounted.

Working Connectivity Into Capital Planning

Because connectivity infrastructure like conduit runs, telecom cabling, and antenna placement is far less disruptive to install before finishes go in, properties planning new construction or a major garage renovation should scope connectivity alongside the broader infrastructure design and investment decision, not as a change order after the electrical work is complete. For existing structures already carrying chargers with a connectivity problem, a phased retrofit, starting with a booster or dedicated access point at the worst-performing level, is usually more practical than a single large capital project, and it gives a property a chance to verify improvement in signal readings before committing to build out every level.

Can an EV charger lose its network connection but still deliver power?

Sometimes. It depends on how the charger's OCPP configuration handles offline authorization. If local authorization or an authorization cache is enabled, a charger may still start a session for a recognized user and queue the transaction data until connectivity returns. If offline authorization is disabled, which is common on networks prioritizing centralized billing accuracy, the charger typically cannot start a new session at all while disconnected.

Why does a garage with strong building Wi-Fi still have charger connectivity problems?

General-purpose building Wi-Fi is usually designed for coverage in occupied spaces like lobbies and units, not for the specific locations where chargers are mounted, which are often against exterior walls, in corners, or on lower levels furthest from existing access points. The same concrete and steel that block cellular signal also degrade Wi-Fi range, so a charger can sit just outside a network's effective coverage even though the building's internet service works fine everywhere else.

Is a cellular signal booster or a dedicated Wi-Fi network the better investment?

It depends on what a pre-installation signal assessment finds. A booster or distributed antenna system makes sense where usable outdoor cellular signal exists nearby and just needs to be captured and redistributed indoors. A dedicated Wi-Fi network tends to make more sense where cellular coverage is weak across all carriers but the building has accessible telecom infrastructure to extend from. Many properties end up combining both for redundancy.

Does poor connectivity affect smart load management, not just billing?

Yes. Coordinated load management that shares available capacity across multiple chargers typically relies on the same network connection used for authorization and billing. A charger that has dropped off the network cannot receive updated charging profiles from the load management system, which can affect how power is allocated across the rest of the site until connectivity is restored.

Should connectivity testing happen before or after chargers are selected?

Before, ideally at the same time as the broader infrastructure planning process. Signal testing at the exact proposed charger locations can reveal whether a site needs a booster, dedicated Wi-Fi, hardwired backhaul, or some combination, and that finding can affect both charger selection and the capital budget. Testing after installation still has value for verification, but it turns connectivity into a retrofit rather than a planned line item.