Indoor cell coverage problems and fixes: from free settings to signal boosters

Limitations & caveats

• The building-materials argument is a statistical model, not a measurement of your house. ITU-R P.2109-2 finds that “buildings fall into two distinct populations: where modern, thermally-efficient building methods are used (metallised glass, foil-backed panels) building entry loss is generally significantly higher than for ‘traditional’ buildings without such materials.” It predicts a distribution across many buildings — the loss through your own walls is whatever step 1 measures.
• That model also splits buildings by construction more sharply than by band across the low-to-mid range; the steep frequency penalty appears at millimeter wave. Read the low-band / mid-band / high-band ordering above as directional, not as a per-band budget for your address.
• The dBm figures in step 1 are a drive-test finding, not a standard. The −80 dBm and −105 dBm numbers come from the FCC’s Mobility Fund Phase II Coverage Maps Investigation staff report, whose success criterion was a 4G LTE speed test reaching at least 5 Mbps down; “observed cell edge” is that report’s own term for the RSRP at which an 80% success rate is achieved, and it ranged from about −115 dBm (T-Mobile) to about −100 dBm (U.S. Cellular) across the three carriers tested, so no single number is a threshold for your phone. What the federal rules fix is only the unit and the reported range: 47 CFR 1.7004(c)(3)(v) has carriers file RSRP maps in dBm “as would be measured at the industry standard of 1.5 meters above ground level,” in 10 dB increments “beginning with a maximum value of −50 dBm and continuing to −120 dBm.” No source we could open sets −110 or −120 dBm as a break point, and the level that matters varies by band, technology and which metric your phone is showing.
• The booster band list is the 2013 list, and it is a legal limit rather than an engineering one. 47 CFR 20.21(a)(4) and (e)(3) confine consumer boosters to frequencies used under parts 22 (Cellular), 24 (Broadband PCS), 27 (AWS-1, 700 MHz Lower A-E Blocks and 700 MHz Upper C Block) and 90 (Specialized Mobile Radio, still awaiting a Commission public notice). The FCC’s own 2018 fact sheet confirms the “original Consumer Signal Booster rules, adopted in 2013, limit operation to specific spectrum bands that were in operation at the time.”
• Consent and registration are the rule as written: 20.21(a)(1) and (a)(2) require the subscriber to obtain the licensee’s consent and register the booster before operating it, and 20.21(h) ends “Licensee consent is voluntary and may be withdrawn at the licensee’s discretion.” That is what the rule requires — it is not a judgment about whether any particular installation of yours is compliant. Ask your carrier.
• The oscillation and auto-shutdown behavior in step 3 is built into the certification standard: 20.21(e)(5) requires consumer boosters to “detect and mitigate any unintended oscillations … (such as may result from insufficient isolation between the antennas),” and (e)(4) requires them to self-correct or “shut down automatically.”
• On femtocells, we can only show one carrier narrowing its lineup: Verizon’s support pages list the 4G LTE Network Extender and Network Extender 2 as “no longer available for purchase” and as no-longer-supported devices. We did not find a source for the claim that Wi-Fi calling is the reason, and t-mobile.com and att.com could not be read server-side on September 13, 2026, so treat availability as a question for your own carrier.
• Wi-Fi calling is verified here on Verizon only, whose FAQ describes it as “a built-in feature on most of our current smartphones” for use “in a place with weak or no cellular network coverage.” The other two nationwide carriers’ pages were unreadable on the date above, so “supported by all major U.S. carriers” is not something this panel proves.
• The step-4 MVNO point rests on the FCC’s own definition: the 2024 Communications Marketplace Report states that MVNOs “have not owned any network facilities, but instead purchase mobile wireless services on a wholesale basis from facilities-based service providers.” The same radio footprint follows from that; price and congestion priority do not, and vary by brand.
• Nothing here is a product recommendation: no models, no brands, no prices, no rankings. The diagnosis in step 1 decides whether a booster is even the right category of purchase, which comes before choosing a unit.
• The “bars” point is anchored to one vendor’s own admission, and an old one. Apple’s July 2, 2010 letter to iPhone 4 owners says the formula Apple used to turn signal strength into bars was “totally wrong” and “in many instances, mistakenly displays 2 more bars than it should for a given signal strength,” and that a software update would adopt AT&T’s recommended formula. Apple framed that as a long-standing error rather than a marketing decision, and wrote that “the real signal strength remains the same” — so read it as evidence that a bar mapping is a vendor’s own choice and can be wrong by whole bars, not that bars are deliberately inflated. We did not open an Apple-owned page confirming that the corrected formula shipped, and a 2010 statement says nothing about how any current handset draws its bars. Separately, the FAQ’s verdict on passive “signal-boosting” stickers and cases is not sourced by anything on this panel.