The satellite industry has a habit of falling in love with whichever orbit happens to be generating the most launch footage. For much of the television age, that orbit was geostationary Earth orbit, the high, seemingly motionless ring 35,786 kilometers above the equator where a single spacecraft could stare at a continent and feed millions of rooftop dishes. Today the glamour has shifted downward. Low Earth orbit, or LEO, has become the orbit of reusable rockets, flat-panel antennas, high-speed internet promises, and spectacular night-sky trains of newly deployed spacecraft. Starlink, OneWeb, Kuiper, and national secure-connectivity projects have made LEO feel like the inevitable shape of the future. In that story, medium Earth orbit can appear awkwardly in-between: not low enough to match the raw latency of LEO, not high enough to dominate a hemisphere like GEO, and not flashy enough to own the public imagination.
That impression is misleading. MEO satellites still matter precisely because they occupy the middle ground, and in orbital engineering the middle ground is not a compromise so much as a design space. A satellite at several thousand kilometers altitude sees vastly more of Earth than a LEO spacecraft, remains close enough to support latency-sensitive communications, and can cover critical regions with far fewer satellites than a dense low-orbit mesh. A navigation satellite around 20,000 kilometers altitude can broadcast timing signals across immense footprints with orbital stability and geometry that a phone, aircraft, container ship, or power grid can exploit continuously. A communications satellite around 8,000 kilometers altitude can combine high-throughput spot beams, steerable capacity, and predictable enterprise-grade links without requiring thousands of spacecraft to remain commercially useful. MEO survives not because the industry forgot to replace it, but because physics still gives it a job.
The current LEO era has made that job easier to misunderstand. LEO’s great strength is proximity: shorter signal paths, smaller free-space loss, and the possibility of broadband latency that feels less like old satellite internet and more like terrestrial wireless. But proximity also creates motion, churn, and scale problems. A LEO spacecraft crosses the sky quickly, often remaining visible to a user terminal for only minutes. Networks must perform constant handovers, coordinate moving beams, manage gateway visibility, and keep enough satellites in enough orbital planes to prevent coverage gaps. That can be done, and done brilliantly, but it is not free. MEO’s higher altitude buys a longer dwell time, a wider field of view, simpler regional continuity, and different economics for customers that care less about shaving every last millisecond and more about assured capacity, sovereign control, service-level agreements, and resilient global reach.
The clearest proof is that MEO is not a relic. SES has continued expanding its O3b and O3b mPOWER medium Earth orbit systems for high-throughput, low-latency connectivity, with O3b mPOWER positioned around 8,000 kilometers altitude and aimed at mobility, government, cloud, enterprise, maritime, and remote-network applications. SES stated in September 2026 that the final three O3b mPOWER spacecraft had arrived at Cape Canaveral to join ten already operational MEO satellites, while Boeing described the delivery as completing a constellation designed to increase MEO network capacity and support demanding commercial and military users. At the same time, global navigation systems continue to rely heavily on MEO: Europe’s Galileo satellites operate in three circular MEO planes at 23,222 kilometers altitude, while GPS, GLONASS, and BeiDou also use medium-altitude constellations for positioning, navigation, and timing. The LEO boom is real, but it has not erased the engineering reasons that made MEO valuable in the first place.
The orbit between spectacle and infrastructure
Medium Earth orbit is usually described as the region between low Earth orbit and geostationary orbit, but that broad definition hides how varied the category is. A satellite at 2,000 kilometers behaves very differently from one at 23,000 kilometers, and both are technically in MEO. The orbit is less a single lane than a wide altitude band where designers tune coverage, latency, radiation exposure, revisit time, orbital period, launch energy, antenna size, power budget, and network architecture. Communications systems such as O3b occupy the lower portion of MEO, around 8,000 kilometers, where latency is far below GEO but coverage per satellite is much larger than LEO. Navigation systems sit higher, where stable geometry and global footprint matter more than broadband throughput.
The historical reason MEO became important was not glamour but geometry. Early satellite communications leaned heavily toward GEO because the value proposition was obvious: put three satellites in the right geostationary slots and you can approximate global coverage outside the polar regions. The ground antenna does not need to track; the satellite appears fixed; broadcast television, trunk telephony, and VSAT networks can be planned around stable beams. The cost is delay. A signal to GEO and back travels a long path even before routers, modems, interleavers, gateways, and terrestrial backhaul add their own delays. For broadcasting, that hardly matters. For interactive voice, cloud applications, financial systems, remote operations, and modern enterprise networking, it matters a lot.
LEO solved the delay problem in a more radical way: bring the satellites close. That idea is not new. Iridium, Globalstar, and Orbcomm all tried variants of low-orbit mobile communications decades before the current broadband boom. What changed was the economics of launch, phased-array antennas, digital payloads, mass production, user-terminal electronics, and vertically integrated network operations. Modern LEO constellations can place hundreds or thousands of satellites into coordinated shells and use software to manage moving coverage as a continuous service. Yet those earlier systems also taught a lesson that remains relevant: low altitude makes the space segment less forgiving. Coverage is built not by a few powerful spacecraft but by fleet density, replacement cadence, spectrum discipline, ground-network integration, and relentless operational control.
MEO emerged as a different answer to the same question. Rather than choosing GEO’s static reach or LEO’s dense swarm, MEO asks how much altitude is enough. At around 8,000 kilometers, a communications satellite still has far less propagation delay than GEO, but it sees a large enough portion of Earth that a modest constellation can serve broad regions. Its apparent motion across the sky is slower than LEO’s, reducing handover frequency and allowing terminals and gateways to maintain longer links. The spacecraft must still be tracked, and the antennas are not as simple as fixed GEO dishes, but the network architecture can be calmer. That calmness has value when the customer is an offshore energy platform, a cruise ship, a defense network, a remote island, or a mobile operator trying to backhaul traffic from difficult terrain.
The navigation story is even more fundamental. GPS did not choose MEO by accident. A useful global navigation satellite must be high enough that many users can see several satellites at once, but not so high that signal power, launch requirements, constellation maintenance, and orbital dynamics become unattractive. The receiver on Earth does not need a broadband pipe; it needs extremely precise timing and orbital information from multiple spacecraft. The geometry must be reliable across latitudes and hours, and the satellites must carry atomic clocks stable enough that nanoseconds become meters. MEO is the orbital altitude at which this architecture becomes elegant. A handful of orbital planes can provide global positioning coverage with satellites that move predictably through the sky, creating the changing geometry that receivers use to solve for position and time.
That is why the phrase “LEO era” can be deceptive. LEO is transforming broadband access and direct-to-device ambitions, but satellite infrastructure is not one market. It is a stack of markets with different physics. Broadcasting, trunking, resilient enterprise service, tactical networks, search and rescue, aviation safety, GNSS timing, maritime navigation, weather monitoring, Earth observation, IoT, and consumer broadband do not all optimize for the same orbit. Some want persistence; some want revisit; some want low latency; some want stable power budgets; some want sovereignty; some want cheap terminals; some want global standards and decades-long continuity. MEO remains relevant because several of those requirements line up unusually well with the middle altitudes.
Latency is important, but it is not the whole network
The case for LEO usually begins with latency, and for good reason. A shorter path through space can make satellite internet feel dramatically different from the GEO broadband systems many users remember: slow page loads, sluggish video calls, awkward gaming, and the subtle pause that made remote work feel remote. But latency is not a single number printed on a satellite brochure. It is a chain. Propagation delay through space is only the first link. Routing, queuing, modem processing, beam scheduling, gateway placement, inter-satellite links, congestion, encryption, acceleration, and terrestrial peering can all shape the user experience. A low satellite altitude does not automatically guarantee a low end-to-end application delay, just as a higher orbit does not automatically make a service unusable.
MEO’s position in that chain is more nuanced than the LEO-versus-GEO comparison suggests. At 8,000 kilometers, propagation delay is clearly higher than LEO. A signal must travel farther, and physics cannot be negotiated. But compared with GEO, MEO reduces the path dramatically, bringing round-trip latency into a range that supports many interactive enterprise and cloud workloads. SES describes O3b mPOWER as a MEO system intended for predictable low latency, high throughput, and flexible capacity, and industry analyses often place its practical service latency far below traditional GEO satellite internet even if not as low as the best LEO paths. For many commercial customers, that middle range is not a defect. It is enough for voice, corporate networking, cloud access, maritime connectivity, remote operations, cellular backhaul, and government traffic, while the wider satellite footprint reduces some of the complexity that comes with LEO.
The engineering trade is especially visible in handovers. A LEO broadband satellite may cross a user’s sky rapidly, forcing the terminal and network to shift sessions among satellites many times during a long connection. Sophisticated phased-array terminals and network software can make this invisible, but invisible does not mean simple. Each handover is an opportunity for transient degradation, scheduling conflict, or coordination burden, particularly in congested regions or at the edge of coverage. MEO satellites move too, but more slowly. Their larger footprints and longer visibility windows can make it easier to deliver predictable service to a ship, aircraft, island gateway, or fixed enterprise terminal. In networks where uptime and service guarantees matter more than peak speed-test glory, fewer handovers can be a meaningful advantage.
There is also the question of gateways. LEO systems need access to ground infrastructure or optical inter-satellite links to route traffic efficiently. If a satellite is over the ocean and lacks a working path to a gateway or neighboring satellites, the network must manage that constraint. Optical inter-satellite links can reduce dependence on local gateways, and newer LEO architectures increasingly rely on them, but they add hardware, pointing, acquisition, tracking, thermal, and operational complexity. MEO systems, because of their broader view of Earth, can often connect user beams and gateway beams across larger regions in a single satellite hop. That does not eliminate the need for ground infrastructure, but it changes the planning problem. SES has highlighted a network of O3b mPOWER gateways distributed across countries including South Africa, Peru, Brazil, Portugal, Australia, Greece, the United States, Chile, the United Arab Emirates, and Senegal, illustrating how MEO still depends on terrestrial integration even as it benefits from wide-area reach.
Capacity is equally subtle. LEO advocates often point to enormous aggregate constellation capacity, and at fleet scale they are right: thousands of satellites can reuse spectrum across many small cells and produce large total throughput. But a customer does not buy “aggregate constellation capacity” in the abstract. A customer buys capacity at a location, on a route, during a season, under weather, regulatory, and congestion constraints. MEO high-throughput satellites can concentrate powerful steerable beams where demand exists, serving trunk routes, mobility corridors, enterprise sites, and government missions with more controlled contention. O3b mPOWER’s software-defined payload approach is designed around that flexibility: digital beamforming, dynamic resource allocation, and the ability to shape capacity not merely by where a fixed beam was drawn years earlier, but by where traffic is actually needed.
This is one reason MEO remains attractive for maritime and aviation networks. Ships and aircraft are not evenly distributed across the planet. Demand clusters along shipping lanes, cruise routes, offshore fields, flight corridors, polar approaches, military theaters, and disaster zones. A constellation that can steer capacity and maintain longer satellite visibility can be easier to engineer for these patterns than a pure broadcast-era GEO system, while requiring fewer moving pieces than a massive LEO swarm. The result is not that MEO “beats” LEO, but that it solves a different version of the broadband problem: fewer satellites, wider footprints, lower latency than GEO, and capacity that can be engineered around predictable high-value traffic.
The user terminal also matters. LEO broadband has pushed phased-array terminal development forward at impressive speed, but electronically steered antennas remain cost-sensitive, thermally constrained, and power-hungry compared with simple fixed antennas. For moving platforms, flat panels are often necessary; for fixed installations, mechanically steered or hybrid systems may remain acceptable. MEO can use tracking antennas that follow fewer, slower-moving satellites, sometimes with dual antennas to maintain make-before-break handovers. In a consumer market, that might be too expensive or bulky. In an enterprise, defense, maritime, or cellular-backhaul market, the economics look different. A terminal serving a remote community, mine, vessel, base, or mobile operator can justify hardware that would never make sense for a mass-market home broadband subscriber.
Rain fade and spectrum behavior further complicate the orbit debate. High-throughput satellite systems often use Ku-band or Ka-band frequencies, where atmospheric absorption and rain attenuation can become serious, especially in tropical climates. A shorter LEO path through the atmosphere does not make weather irrelevant; a higher MEO link budget does not make it doomed. Designers respond with adaptive coding and modulation, uplink power control, site diversity, gateway placement, fade margins, and traffic management. What matters is not simply orbit altitude, but how the entire radio-frequency chain behaves under real weather, interference, blockage, and congestion. MEO operators serving equatorial and maritime markets have had to engineer around these realities for years, making the orbit less experimental than outsiders sometimes assume.
The quiet dominance of MEO in navigation and timing
Broadband gets the headlines, but the most important MEO satellites may be the ones most people never think about: navigation spacecraft. GPS, Galileo, GLONASS, and much of BeiDou form the invisible timing layer under modern civilization. Their signals guide aircraft approaches, synchronize mobile networks, timestamp financial transactions, help ships traverse oceans, steer tractors through fields, coordinate emergency response, support power-grid phasor measurements, and let smartphones merge satellite ranging with inertial sensors, cellular signals, Wi-Fi maps, and barometric pressure. When people say LEO is the future of satellites, they usually do not mean they want to move global navigation entirely to low orbit tomorrow. The installed base, receiver ecosystem, orbital geometry, spectrum allocation, and trust model of MEO GNSS are too deeply embedded.
MEO is well suited to GNSS because positioning is a geometry problem before it is a communications problem. A receiver estimates its distance from multiple satellites by comparing timing codes and navigation messages. To solve for latitude, longitude, altitude, and receiver clock error, it needs several satellites visible at once, preferably spread across the sky. Too few satellites, too narrow a geometry, or too much signal blockage can degrade accuracy. MEO satellites provide large footprints, predictable orbital motion, and a balance between signal strength and coverage that lets a constellation of dozens serve the entire planet. Galileo’s architecture, for example, uses three MEO orbital planes at 23,222 kilometers altitude, while GPS uses MEO spacecraft at roughly 20,200 kilometers altitude in multiple planes; these are not arbitrary choices, but engineering compromises tuned for global availability and timing performance.
The signal from a GNSS satellite is extraordinarily weak by the time it reaches the ground, often below the thermal noise floor before correlation processing recovers it. That weakness is part of the design challenge. The satellite must broadcast with enough power and spectral discipline to be useful worldwide, but receivers must remain cheap, compact, battery-powered, and capable of operating with tiny antennas. The system depends on precise clocks, accurate ephemerides, ionospheric correction models, relativistic adjustments, ground monitoring, and continuous control. In this environment, MEO offers a kind of architectural stability that LEO does not naturally provide. The satellites do not whip overhead in minutes; their orbits are stable enough for carefully modeled predictions; their coverage geometry supports global service without needing thousands of spacecraft.
That does not mean LEO has no role in positioning. Quite the opposite: LEO positioning, navigation, and timing is one of the most interesting emerging ideas in the satellite industry. Low-orbit signals are stronger at the receiver, satellites move quickly enough to create useful Doppler observables, and proliferated constellations could add resilience against jamming or spoofing. Commercial LEO broadband networks already transmit structured radio signals that researchers and companies have explored for opportunistic navigation. Dedicated LEO PNT constellations could complement GNSS by adding regional robustness, faster geometry changes, and stronger signals for difficult environments. But complement is the key word.
Top comments (0)