The graph of human evolution is far from linear, however the graph of technological evolution is definitely exponential. A couple of millennia ago, humans used pigeons to send messages to their loved ones. Eventually, we invented phone lines for better connectivity, and then we developed telecommunication satellites for ubiquitous connectivity. A couple of centuries ago, humans used good Samaritans to find their way through a city. Then came in detailed maps and signboards, and eventually we moved on to satellite-based GPS to guide us anywhere, anytime. A couple of decades ago, humans only used bulky books and papers to exchange ideas. Then we invented a huge, complicated network of cables and PCs to connect people around the world. We have pushed the envelope of innovation to eventually use satellites and beam Internet from Space.
Now, there are at least a dozen different organisations trying to develop novel technologies to provide affordable satellite internet. But before we get to know how they’re planning to fulfil the world’s internet needs, let us understand what is Satellite Internet.
To gain a simple, vague understanding, imagine a human nervous system. There are a lot of neurons in the spine and the brain, which constitute the majority of the signal flow. These get distributed to smaller neurons, until each and every organ and cell is connected to the system (directly or indirectly). Some organs are in constant need of signals, like when your limbs or your eyes. Others seldom get any signals, like your adrenaline glands.
In this analogy, the cables connecting major countries and continents are the spine of the system. Smaller networks and cables are like individual neurons, while the user is like an individual cell. Finally, your data is the neural signal that is carried to some major server.
Now imagine if we came up with a Neuralink-like device, which can receive and send signals from and to any part of the body. This device can tap into the potential that the person has no conscious control over. That is Satellite Internet. A connectivity system around the Earth, which will complement the core internet, while also tapping into its hidden potential.
But how do major companies differ in their ability to provide such connectivity? And how does the design of their system affect their service? Given the limited information we have about these companies, it is not easy to have an unbiased comparison. So, for the sake of accuracy, we’ll just compare certain major characteristics of the four most-popular front-runners in this race: Starlink (SpaceX), TeleSat, OneWeb and SpaceNet (Yours truly, Astrome). Here is how their satellite internet constellations (read ‘group of multiple satellites working together’) compare:
Altitude is one of the most basic yet important parameters of a satellite constellation. The higher your satellite is, the more area it can cover. But the farther away the satellite is, the higher the latency (read ‘lag’) of the connection. Also, the lower the orbit, the shorter the life span of a satellite. So it is a clear trade-off between latency, life and coverage.
Starlink is planning to deploy its satellites in orbits ranging from 550km to over 1325 km in altitude. TeleSat will utilise the orbits between 1000km and 1250km, while OneWeb uses the 1200km orbit. SpaceNet will have its satellites in a slightly higher, 1530 km orbit.
Number of Satellites
The number of satellites in the constellation are a good indicator of the financial model of a constellation. The higher the number of satellite, the more capital is required to launch the constellation, and higher the cost of maintenance. However, the lower number of satellites, the smaller your service region. So the number of satellites is a trade-off between the cost of setting up & running the system, and the expected demand in the service region.
At a whopping 4409, Starlink proposes the largest number of satellites in a constellation achieving global coverage. In comparison, OneWeb needs only 720 satellites for the same coverage. TeleSat requires 292 while SpaceNet needs just 198 satellites to provide services to most people in the world. SpaceNet promises coverage between the 40th north and south parallels, while the others claim global coverage.
The signal frequency is a characteristic feature of any communication satellite. Frequency bands (read ‘certain usable range of the Electro-Magnetic Spectrum’) are a precious commodity, and sell like gold dust. Each band has its pros and cons, in terms of the data rates it supports and attenuation (read ‘disturbances in the signal while transmitting’) it suffers. Generally, the higher the frequency, the higher the data rates as well as attenuation.
Also, popular bands have a lot of commercial users, only leaving certain channels for one company. So, bands like the Ku- Band are congested due to such usage. In summary, the frequency band is a trade-off between data rates, congestion and signal attenuation.
Currently, only SpaceNet plans to utilise the E band, in the 71–76 GHz and 81–86 GHz range. The other three of the Fantastic Four plan to use the Ku/Ka Bands, between 10 GHz to 30 GHz.
Capacity per Satellite
While the mass is an important parameter in satellite design, the capacity per satellite is one of the most significant metric for a space-internet constellation. The higher the capacity per satellite, the more users it can service at higher data rates. The capacity is an amalgam of various other parameters, including the frequency band, as well as the underlying technology. So, it is a trade-off between data rates, number of users, size and underlying technology of the satellite.
At 9 Gbps per satellite, OneWeb has the lowest capacity per satellite. Starlink raises the bar to about 20 Gbps, followed by TeleSat at 36 Gbps. Satellites in these constellations belong to a class called High Throughput Satellites (HTS). Each SpaceNet satellite, on the other hand, has a capacity of 180 Gbps — several times more than the highest capacity achieved by the best among the Fantastic Four. With this technology, SpaceNet creates a new class of satellites, called the Ultra-HTS.
Weight of Satellite
The weight of an individual satellite does not directly relate to the network performance, but it gives an indirect metric of the complexity and technical proficiency of the satellite design. The lighter the satellite, the less expensive it is to launch it. Smaller satellites also indirectly reflect on a superior design and integration process, ceteris paribus.
OneWeb is the lightest of the lot, at just 145 kg, followed by TeleSat’s test satellite at 168 Kg. Starlink follows suite with about 227 kg per satellite. Additionally SpaceNet is expected to have the weight at an even 300 kg. However, if we look at the metric of ‘Capacity per kg’, SpaceNet takes the cake, over TeleSat, SpaceX, and then OneWeb.
Total Network Capacity
The world’s total Internet Capacity was ~400 Tbps in 2018. To explain it in layman terms, it is roughly the total capacity to transfer data over the entire internet infrastructure, every second. Now, as the constellations take their final form, their network capacity add up to the total internet capacity. The higher the number, the better the ability of the network to manage data flow.
OneWeb adds up to 2 Tbps to the total internet capacity, while TeleSat adds up a modest 8 Tbps. Both Starlink and SpaceNet will add up 24 Tbps each to the total capacity. Once the constellations are up, only Starlink and SpaceNet will account for almost 10% of our total global internet capacity.
So, who wins The Amazing Race? In a competition like this, there is no one winner. Globally, the demand for internet is ineffably large, ensuring a business opportunity for everyone. Each company is trying to trade-off on different metrics, and giving priority to others. Looking at the number of satellites alone, Starlink seems to give higher priority to coverage area than the data rates, while SpaceNet is focusing more on capacity per satellite than the coverage area. OneWeb seems focused on reducing their operating cost at the expense of a lower network capacity, while TeleSat is trying to play it safe and use it’s experience over reinventing the wheel. Overall, OneWeb looks like the easiest to set up, financially. TeleSat sounds more feasible. SpaceNet is more innovative and financially efficient. And Starlink, like any Musk venture, is the most ambitious due to its shear scale.
There are a lot of other parameters these constellations can be compared on. And all these comparisons do not even include other players like Kuiper Systems (Amazon) and LeoSat. Irrespective of who is most successful, it is abundantly clear that the real winners of this race, are the consumers. The people and the organisations who will use these services. I, for one, cannot wait to see that day.