Starlink Under Siege: Iran's Jamming Crisis, GPS-Free Navigation, and the DTC Revolution

On January 8, 2026, Iran pulled the plug on 85 million people. The regime severed internet connections, cut telephone lines, and—for the first time ever—launched a coordinated, military-grade assault on Starlink. Urban connection rates plummeted by 80%. Packet loss hit 30% nationwide, with some areas experiencing 80% degradation.

This wasn’t a simple blackout. It was proof of concept: LEO constellations, long marketed as censorship-resistant, can be overwhelmed by nation-state electronic warfare.

For GNSS professionals, the implications extend far beyond connectivity. Starlink terminals depend on GPS for geolocation. Iran’s jamming strategy exploited exactly this vulnerability. And SpaceX’s response—GPS-independent positioning using LEO signals themselves—may represent the most significant development in alternative PNT since the advent of GNSS augmentation systems.

This article examines the technical realities behind the headlines: how Iran jammed Starlink, what countermeasures exist, and why SpaceX’s GPS-free positioning capabilities could fundamentally change the PNT landscape.

The Iran Jamming Campaign: What Actually Happened

The Trigger

Large-scale protests erupted across Iran in late December 2025 and intensified through January 2026. By January 8, the regime decided on total information blackout.

Unlike previous shutdowns targeting terrestrial infrastructure alone, this operation specifically targeted satellite communications. According to Amir Rashidi of the Miaan Group: “In previous incidents, none of the internet shutdowns were as severe as this one. We never saw Iran trying to jam Starlink. Now they’re doing it.”

The Attack Vector: GPS Dependency

Starlink user terminals require GPS signals to determine their position and orient their phased-array antennas toward passing satellites. Without accurate geolocation, terminals cannot establish or maintain links.

Iran exploited this dependency directly. Rather than attempting to overpower the Ku-band downlink from LEO satellites—which would require enormous resources given the signal strength from 550 km altitude—they jammed the GPS signals the terminals rely on for positioning.

The strategy mirrors Russian tactics observed in Ukraine: disrupt the weak GPS signal (arriving from 20,200 km altitude at approximately -130 dBm) rather than the stronger LEO signal.

The Hardware: Cobra V8 and Mobile Platforms

Iran reportedly deployed its Cobra V8 electronic warfare system, first unveiled in 2023. Designed as a “versatile multi-mission electronic warfare system capable of intercepting, analyzing, and jamming enemy radar systems,” the Cobra V8 was adapted for GPS interdiction.

Analysis suggests the interference pattern came from mobile jamming units moved between neighborhoods—a tactic that complicates countermeasures since the jamming source constantly relocates. This approach closely mirrors Russian “Kalinka” jamming operations.

Some observers speculate Iran received technical assistance from Russia, potentially including the Kalinka system itself, though this remains unconfirmed.

The Results

By January 20, Iranian authorities claimed they had disrupted 40,000 Starlink connections. Independent monitoring showed:

80% packet loss in heavily jammed urban areas
30% average packet loss nationwide
Near-total service disruption in protest hotspots
Intermittent connectivity in rural areas where jamming coverage was sparse

The human cost was severe. The U.S.-based Human Rights Activists News Agency reported at least 2,500 deaths during the crackdown, with the communications blackout enabling systematic suppression of documentation.

Starlink’s Anti-Jamming Arsenal

The Iran campaign exposed vulnerabilities, but SpaceX hasn’t been passive. The company has deployed—and continues to develop—multiple countermeasures.

1. Adaptive Null Steering

Starlink’s phased-array antennas can mathematically combine signals to create “nulls”—points of zero sensitivity—in the direction of interference sources.

Technical implementation:

Antenna elements receive signals from all directions
Digital signal processing identifies jammer direction via angle-of-arrival estimation
Beamforming weights are adjusted to minimize gain toward the jammer
Null depth of 20–40 dB is achievable, effectively reducing jammer power by 100–10,000×

Limitations:

Requires sufficient signal-to-interference ratio to identify jammer location
Multiple distributed jammers can overwhelm null-steering capacity
Doesn’t help if the jammed signal is GPS rather than the Starlink link itself

2. Frequency Agility and Spread Spectrum

Starlink uses Orthogonal Frequency-Division Multiplexing (OFDM) with spread-spectrum techniques, distributing data across multiple subcarriers. This provides inherent jamming resistance:

Narrowband jammers affect only a subset of subcarriers
Wideband jammers require proportionally more power
Frequency hopping can avoid known interference bands

However, these techniques protect the Starlink link itself—not the GPS receiver that the terminal depends on.

3. Optical Inter-Satellite Links (ISLs)

Perhaps the most significant architectural advantage is the deployment of laser inter-satellite links on V2 and V3 satellites. These “space lasers” enable satellite-to-satellite communication in the vacuum of space at near-light speeds.

Why this matters for jamming resistance:

Ground-based jamming cannot reach optical links
Traffic can be routed around compromised ground segments
Reduces dependency on ground stations in contested regions

As of early 2026, optical ISLs are fully operational across the constellation, creating a mesh backbone that’s fundamentally immune to RF interference.

4. GPS-Independent Positioning: The Game Changer

This is where things get interesting for the PNT community.

SpaceX has deployed software enabling terminals to triangulate position using Starlink signals themselves, bypassing GPS entirely. The approach exploits several LEO-specific advantages:

Signal strength: LEO satellites at 550 km altitude deliver signals 30+ dB stronger than GPS satellites at 20,200 km. This makes them inherently more difficult to jam.

Doppler observables: LEO satellites move across the sky rapidly (completing an orbit in ~95 minutes), generating significant Doppler shift. This Doppler can be exploited for positioning.

Constellation geometry: With 10,000+ satellites, Starlink provides near-optimal geometric dilution of precision (GDOP) globally.

Deep Dive: LEO-Based Positioning Without GPS

The foundational research enabling GPS-free Starlink positioning comes from Dr. Zaher M. Kassas at Ohio State University’s ASPIN Laboratory. His team demonstrated the first successful Starlink-based positioning in 2021 and has since refined the approach through multiple experimental campaigns.

The methodology employs two receiver architectures:

R1: Cognitive Opportunistic Navigation

Utilizes minimal, publicly available knowledge about Starlink signal structure
Exploits known synchronization sequences and orbital ephemeris
Extracts carrier phase and Doppler observables
Achieves meter-level accuracy with multiple satellite observations

R2: Blind Approach

Assumes no prior knowledge of signal structure
Uses Generalized Likelihood Ratio (GLR) test methods adapted from radar signal processing
Detects and tracks Doppler without knowing transmission parameters
Demonstrated Hz-level Doppler tracking accuracy

Signal Acquisition and Tracking

The technical pipeline involves several stages:

1. Beacon Detection Starlink satellites transmit periodic synchronization signals. Using matched subspace detection, the receiver identifies these beacons without knowing their exact structure.

2. Doppler Estimation An adaptive Kalman filter-based phase-locked loop (PLL) extracts carrier phase and Doppler observables. The Doppler shift varies significantly as LEO satellites pass overhead—from approximately +40 kHz to -40 kHz over a ~10-minute pass at Ku-band.

3. Multilateration Multiple Doppler measurements from different satellites are fused through a nonlinear least-squares estimator. Given that each satellite’s orbit is known precisely (broadcast in ephemeris data), the receiver’s position can be computed.

Demonstrated Performance

Kassas’s team has validated LEO navigation across diverse platforms:

Platform	Satellites Used	3D Position RMSE	Final Error
Stationary receiver	6 Starlink	5.8 m	5.1 m (2D)
Ground vehicle (blind)	4 Starlink + 1 OneWeb + 3 others	9.5 m	4.4 m
High-altitude balloon (>25 km)	5 Starlink + 1 OneWeb	17.94 m	—
Arctic maritime	OneWeb + Starlink	27 m	—

Compare this to GPS-denied scenarios where inertial navigation alone accumulated errors of 472–525 m over similar trajectories.

The Physics: Why LEO Works for PNT

Proximity advantage: LEO satellites are ~36× closer than GPS satellites. Signal strength scales with the inverse square of distance, meaning LEO signals arrive approximately 1,300× stronger (31 dB) at the receiver.

Doppler information richness: A GPS satellite, moving slowly relative to a ground observer, produces only ~5 Hz/s Doppler rate. A LEO satellite produces ~50 Hz/s—an order of magnitude more information per unit time.

Rapid geometry change: GPS satellites take 12 hours to orbit; LEO satellites take ~95 minutes. The fast-changing geometry means positioning solutions converge quickly as satellites move.

Constellation density: GPS has 31 satellites. Starlink has 10,000+. This density ensures multiple simultaneous observations from diverse angles.

SpaceX’s FCC Filing: Official PNT Capabilities

In response to the FCC’s Notice of Inquiry on PNT resilience (WT Docket No. 25-110), SpaceX formally disclosed Starlink’s positioning capabilities:

Timing accuracy: Nanosecond-level using time-of-arrival measurements
Position accuracy: Meter-level using broadband signals
Independence: Starlink satellites already operate autonomously from GPS
Spectrum: PNT can be delivered within existing Ku- and Ka-band allocations

SpaceX emphasized its Direct-to-Cell L-band signals as particularly suitable for PNT. Todd Humphreys at UT Austin cautioned that current Ku-band timing is “so irregular that accurate pseudorange-based PNT is not possible,” but noted that L-band D2C—with proper implementation—could potentially rival GPS if multiple satellite observables are available.

The Direct-to-Cell Landscape: Status Quo

The Iran jamming crisis intersects with another major Starlink development: Direct-to-Cell (DTC) service. Understanding DTC’s current state is essential because the same L-band signals enabling smartphone connectivity may become the foundation for next-generation PNT.

Starlink DTC: Where We Are

Constellation status: SpaceX has launched 650+ DTC-capable satellites, sufficient for global text coverage.

Current capabilities:

Text messaging (SMS)
Location sharing
Voice/video calls via WhatsApp
Navigation apps (Google Maps)
30+ DTC-optimized applications

Commercial launch: T-Satellite service launched July 2025 in partnership with T-Mobile.

User base: 12 million customers in 2025, averaging 6 million monthly active users across 22 countries.

Pricing: $10/month add-on (included free for T-Mobile Experience Beyond plans).

The V3 Horizon

SpaceX plans to deploy Starlink V3 satellites via Starship starting in 2026. The performance jump is substantial:

Metric	V2 Mini	V3
Downlink capacity	Baseline	10×
Uplink capacity	Baseline	24×
Cellular capability	4G/LTE	Full 5G

V3 DTC satellites are expected to deliver “a comparable experience to terrestrial 5G”—sufficient bandwidth for streaming video, not just text messages.

The eleventh Starship test (October 2025) was fully successful, clearing the path for operational V3 deployment in the first half of 2026. However, constellation-wide V3 coverage likely won’t arrive until 2027–2028.

Competitive Landscape

AST SpaceMobile:

Launched BlueBird 6 (January 2026)—the largest commercial communications satellite ever deployed
BlueBird 6 is 3× the size of previous satellites
Plans 45–60 additional satellites by end of 2026
Partners: AT&T, Verizon
Targeting 5G data service in the U.S. once constellation reaches ~25 satellites (mid-2026)

Lynk Global:

8 satellites operational
Pending merger with Omnispace (backed by SES)
Combining S-band spectrum with Lynk’s platform
Best suited for intermittent SMS; lacks bandwidth for sustained broadband

Globalstar (Apple partnership):

Emergency SOS, roadside assistance, Find My
$1.7 billion Apple investment for 17 new satellites
Apple holds 20% stake, securing 85% of network capacity
Expanding beyond emergency use with iOS 18

Amazon Leo (Project Kuiper):

Preparing for 2026 launch
Primary focus: broadband internet
D2C capabilities possible with future infrastructure

Why DTC Matters for PNT

The L-band frequencies used for Direct-to-Cell are inherently better suited for PNT than Ku/Ka-band broadband:

Propagation: L-band penetrates foliage and buildings more effectively
Receiver simplicity: Smartphone-compatible signals mean simpler, cheaper receivers
Standardization: Mobile industry standards (3GPP) provide interoperability
Dual-use: Same signal serves both communication and positioning

If SpaceX implements proper timing structures in its L-band D2C signals, every smartphone becomes a potential LEO PNT receiver—creating a resilient backup to GPS without requiring dedicated hardware.

Implications for the PNT Community

The Layered Approach Reality

The FCC’s inquiry envisions “layered” PNT resilience: multiple independent systems providing redundancy. Starlink fits this model well because:

It operates on different frequencies than GPS
It uses different orbital regimes (LEO vs. MEO)
It’s commercially operated (diversifying from government-only systems)
It provides orders-of-magnitude stronger signals

But Starlink is not a GPS replacement. It’s a complement—one layer in what should be a multi-source navigation architecture.

What Iran Taught Us

The Iran campaign demonstrated both vulnerabilities and resilience:

Vulnerabilities:

GPS dependency remains a single point of failure for standard terminals
Coordinated, mobile jamming can achieve 80% degradation
Nation-states with dedicated resources can impact service significantly

Resilience factors:

Service wasn’t completely eliminated even under heavy jamming
Optical ISLs maintained backbone connectivity
GPS-independent positioning (where deployed) provided fallback
Constellation redundancy meant no single point of failure

Consider these takeaways:

1. Plan for LEO integration. Commercial LEO signals will increasingly supplement GNSS. Receivers capable of opportunistic LEO positioning will outperform GPS-only solutions in contested environments.

2. Doppler-based positioning is mature. The Kassas framework and similar approaches have demonstrated sub-10m accuracy without any cooperation from satellite operators. This capability exists today.

3. D2C signals deserve attention. The L-band DTC ecosystem may provide the most practical path to consumer-grade LEO PNT. Monitor 3GPP NTN standards development closely.

4. Jamming is an operational reality. The Iran campaign is a preview, not an anomaly. Design systems assuming GPS denial.

What Comes Next

Near-term (2026)

SpaceX V3 deployment begins (dependent on Starship cadence)
Starlink DTC expands to UK and additional markets
AST SpaceMobile targets beta service with ~25 satellites
Continued FCC rulemaking on PNT resilience
Likely more nation-state jamming incidents as capabilities proliferate

Medium-term (2027–2028)

V3 constellation reaches critical mass for “5G from space”
Potential SpaceX PNT service offering (if FCC framework permits)
Commodity LEO PNT receivers as academic designs commercialize
Integration of LEO positioning into automotive and aviation systems

Long-term implications

The Starlink constellation—with its 10,000+ satellites, strong signals, and global coverage—represents a potential paradigm shift in PNT. Not as a GPS replacement, but as the foundation of a truly resilient multi-source navigation ecosystem.

Iran’s jamming campaign was a stress test. Starlink passed, if imperfectly. The next phase is building systems that exploit LEO’s inherent advantages while acknowledging its limitations.

For GNSS professionals, the message is clear: LEO isn’t coming. It’s here. The question is how we integrate it.

TL;DR

Iran jammed Starlink during January 2026 protests by targeting GPS signals that terminals depend on—achieving 80% packet loss in urban areas
Countermeasures exist: adaptive null steering, spread spectrum, optical ISLs, and crucially, GPS-independent positioning using Starlink signals themselves
LEO PNT is real: Research demonstrates sub-10m positioning using Doppler multilateration from Starlink/OneWeb without GPS—signals are 30+ dB stronger than GNSS
DTC status: 650+ satellites operational, 12M users, voice/data working via 30+ apps; V3 satellites (10× capacity, full 5G) deploying in 2026
SpaceX told the FCC Starlink can provide nanosecond timing and meter-level positioning independently of GPS—potentially revolutionizing PNT resilience