Satellites - Single Point of Failure

One of the worst things you can have in combat is a single point of failure;  a linchpin upon which an entire operation depends.  If it fails, you lose everything.  For example, let’s imagine a ship has several types of sensors and all are operated off electrical power that ultimately passes through a single cable.  That cable would represent a single point of failure.  Sever the cable and all the sensor systems would be rendered inoperative by a single component’s failure.  We have seen examples in industry and the military where supposedly redundant, independent systems have failed because of an unrecognized single point of failure.  In our example, we might be tempted to proudly proclaim that we have set up multiple independent sensor systems.  After all, they’re physically separated.  They use different frequencies.  They each have their own cooling systems.  Those that need motors each have their own.  Completely independent;  nothing in common ...  except for the single power cable that no one thought about and did not recognize as a single point of failure.
 
Our military is currently constructing a single point of failure on a very large scale in the form of satellite communications.  Consider the vast number of systems that all depend on satellite communications:  drones, weapons, command and control, sensors, and on and on.  If we lose satellite communications, we lose … well … a huge chunk of our entire military capability.
 
You might be tempted to argue that satellite communications are not a single point of failure because we have dozens/hundreds/thousands of satellites and no enemy could possibly eliminate all of them.  Setting aside the highly debatable nature of that assertion, think, what do all satellites have in common?  That’s right, a very small number of control/communication stations … approaching a single point of failure.  It doesn’t matter how many satellites you have if they all depend on a very few control stations.
 
Think further, what do all control stations have in common?  An even smaller number of software programs that run them.  If an enemy can hack/virus the software (we’ve seen that it’s almost impossible to prevent hacking of military or commercial networks and programs), it instantly renders all the control stations and their associated satellites useless.
 
Well, none of that could ever happen, you say.  Except that already has and we have at least one public example.
 

A global outage across Elon Musk’s satellite network ​affecting millions of Starlink users had left two dozen unmanned surface vessels bobbing off the California coast, disrupting communications and halting operations for almost an hour.
 
The incident, which involved drones ‌intended to bolster U.S. military options in a conflict with China, was one of several Navy test disruptions linked to SpaceX's Starlink that left operators unable to connect with autonomous boats, according to internal Navy documents reviewed by Reuters and a person familiar with the matter.[1]

Think even further.  What if a key individual (a single point of failure, by definition!) were to shut down or sabotage the entire satellite system, perhaps motivated by bribes/payoffs or coerced by Chinese threats against him or his family?  Couldn’t happen, you laugh?  How about this:
 

Reuters reported last year that [Elon] Musk unexpectedly switched off Starlink access to Ukrainian troops as they sought to retake territory from Russia, denting allies' trust in the billionaire.[1]

There you have a single (non-military!) individual shutting down a satellite system because he disagreed with the military and geopolitical actions.
 
Think further still.  The single point of failure doesn’t have to involve destruction of assets, cyber attacks, or rogue villains.  It could be as simple as routine mechanical/software failure.  Here’s an example.
 

In April 2025, during a series ​of Navy tests in California involving unmanned boats and flying drones, officials reported that Starlink struggled to provide a solid network connection due to the high data usage needed to control multiple systems, according to a Navy safety report of the tests reviewed by Reuters.[1]

Given the complexity of modern systems, sometimes you don’t even know why a system failed.
 

In the weeks leading up to the global Starlink outage in August, another ​series of Navy tests was disrupted by intermittent connection issues with the Starlink network, Navy documents reviewed by Reuters show. ​The causes of the network losses were not immediately clear.[1]

Now let’s close with one of the dumbest statements I’ve heard in a while, courtesy of Mr. Bryan Clark at the Hudson Institute.
 

Despite the setbacks, the upside of Starlink – a cheap and commercially available service – outweighs the risk of a potential outage disrupting future military operations, said Bryan Clark, an autonomous warfare expert at the Hudson Institute. “You ​accept those vulnerabilities because of the benefits you get from the ubiquity it provides,” he said.[1]

In peacetime you might accept the occasional failure because the downside isn’t all that serious.  It’s not life or death … just inconvenience.  In war, it most certainly is life and death and you can’t afford systems with known single points of failure and a known, not insignificant, failure rate to begin with.
 
 
Discussion
 
Given our overwhelming dependence on satellites, do you think the Chinese are going to allow us to continue to use them, unhindered?  Of course not!  They’ll attack them physically, electronically, digitally (software), and via human operator vulnerabilities.  Anyone want to bet that there aren’t Chinese agents embedded in the military or Starlink?  One way or another, I’d be very surprised if we have many operating satellites two weeks after the war starts.
 
We see that satellite communications could be disrupted in a variety of ways:  physical destruction, cyber attacks, software viruses, sabotage, and routine failure, among other methods.  Does it make sense that so much of our military, current and future, has, as its very foundation, satellite communications?  Would any sane individual purposely build a system with that many known, anticipatable vulnerabilities?  And yet we’re doing exactly that.  Worse, we’re increasing our use of, and dependence on, satellites instead of decreasing it.
 
Consider that the military, some years ago, recognized GPS as a significant vulnerability.  All of our navigation and most of our weapons depend on GPS.  GPS was recognized as a single point of failure and the military, to their slow and belated credit, has set about mitigating that dependence.  We’re now designing systems that use alternates to GPS.  We knew we couldn’t accept a single point of failure, like GPS, and yet we’re intentionally and enthusiastically embracing satellites as a single point of failure.  That’s baffling.
 
 
 
 
______________________________

 
[1]Reuters website, “Exclusive: Starlink outage hit drone tests, exposing Pentagon’s growing reliance on SpaceX”, David Jeans, 16-Apr-2026,
https://www.reuters.com/business/media-telecom/starlink-outage-hit-drone-tests-exposing-pentagons-growing-reliance-spacex-2026-04-16/

Networks and Royal Navy Signal Flags

In several posts, we’ve examined historical naval battles and analyzed them from today’s perspective to learn lessons applicable to our current strategies, doctrine, and tactics.  The timelessness of the lessons has been striking.

 

Today, we’re going to look not at a battle but at a technology and see what lessons can be learned.  Specifically, we’re going to look at the English Royal Navy system of signal flags, implemented in the age of sail.  As we do, we’ll keep in mind the parallel between the English flag system and today’s networks and command and control (C2) systems.

 

Before the advent of radio, communication between ships was problematic, to say the least.  Ships that were out of sight of each other were, essentially, out of communication.  Even within sight of each other, communications were difficult.  Simplistic signals involving sails or flags or lanterns or some such were used but their effectiveness was subject to degradation by darkness, weather, distance, and the confusion of combat.

 

Eventually, the Royal Navy attempted to standardize the meaning of a limited set of signal flags, as described below.

 

Between 1337 and 1351 the British Navy lists two signals in their old “Black Book of the Admiralty.” The first was to hoist a flag of council high in the middle of the mast, to notify all captains to come aboard the admiral's flagship for a meeting. Hoisting another flag aloft reported the sighting of the enemy. (3)

 

By the late seventeenth century things still had not progressed much. A code book issued for the British Navy in 1673 defined 15 different flags, each with a single predefined meaning, which was probably not too much different from what had been used since antiquity. (3)

 

Following this, the English came up with a major step forward when they established an extensive system of standardized signal flags.  From Wikipedia (4),

 

Several wars with the Dutch in the 17th century prompted the English to issue instructions for the conduct of particular fleets, such as (in 1673) the Duke of York's "Instructions for the better Ordering of His Majesties Fleet in Sayling". Signals were primitive and rather ad hoc ("As soon as the Admiral shall loose his fore-top and fire a gun..."), and generally a one-way communication system, as only flagships carried a complete set of flags. In 1790 Admiral Lord Howe issued a new signal book for a numerary system using numeral flags to signal a number; the number, not the mast from which the flags flew, indicated the message. Other admirals tried various systems; it was not until 1799 that the Admiralty issued a standardized signal code system for the entire Royal Navy. This was limited to only the signals listed in the Signal-Book. In 1800 Captain Sir Home Popham devised a means of extending this: signals made with a special "Telegraph" flag referred to a separate dictionary of numbered words and phrases. (4)

 

The system was straightforward and simple.

 

The signaling system in use by the Royal Navy at the time of the Battle of Trafalgar had been introduced in 1799 by Admiral Lord Howe and was further developed by Captain Sir Home Popham. Howe's original system was based on a set of flags numbered 0 to 9 that when hoisted in various combinations could transmit words, numbers, or messages listed in the Signal Book. Each letter of the alphabet was allotted a number, enabling words to be spelled out, The numeral flags could also be used to send numbers as such. But most signals were two-or three-flag combinations which referenced messages in the Signal Book, e.g. to discontinue the engagement, to pursue the enemy, to anchor, etc. The Signal Book was so arranged that the sender and receiver could quickly compose and interpret flag hoists. (2)

 

Howe's system was impractical for sending long messages due to the large number of flags required to spell out individual words. In 1800, therefore, Captain Popham developed a "vocabulary" system by which three- or four-flag hoists referenced words or phrases in the Signal Book. This was the system in use at the time of Trafalgar and Nelson's famous signal is a fine example of how it worked. This type of hoist was preceded by the so-called telegraph flag, indicating that a vocabulary message was to be sent. The signal was terminated by a finishing flag, functioning like the period at the end of a sentence. Receiving ships acknowledged the signal by hoisting the Affirmative flag, indicating that it had been seen and understood. If the signal was seen but not understood, receiving ships hoisted the Affirmative flag over numeral flag no. 8. For repeated numbers a substitute flag was provided. In Nelson's signal this flag, yellow with black stripes along the top and bottom, was employed in the hoist for the word do (No. 221), substituting for a second no. 2 flag.

 

The second signal that Nelson ordered to be hoisted was No. 16 in the Signal Book: "Engage the enemy more closely." This remained aloft until it was shot away in the heat of battle. (2)

 

 

Standardized signal flag systems, while a vast improvement over ad hoc signals, were still subject to distance limitations.  The sender and receiver had to be within visual range.  This was partially remedied by using signal chains whereby the sender could disseminate a signal to ships out of visual range by using a repeater chain of ships, each ship relaying the message to the next ship in line.  Of course, this required time for the message to slowly repeat its way along the chain and required a string of ships.  Fortunately, naval battles in the age of sail were generally fairly slow affairs, often requiring many hours to lead up to the actual battle.  The system also suffered from the confusion of battle where signals might be missed, misread, shot away, or otherwise fail since both the sender and receiver had, understandably, higher priorities.

 

Here is an example of the signal chain from the Battle of Trafalgar:

 

Accordingly the French admiral Villeneuve hoisted the signal to weigh anchor, and at six in the morning of 19 October the British frigate Sirius, waiting outside Cadiz, signalled to the fleet below the horizon ``Enemy have topsails hoisted.'' An hour later it hoisted signal no. 370, ``Enemy ships are coming out of port.'' The hoists were made to the next frigate in the signalling chain, Euryalus, which in turn signalled no. 370 to Phoebe with the accompanying admonition--superfluous in a service schooled to such discipline--``Repeat signals to lookout ships west.'' And so no. 370 travelled down the chain, from Phoebe to Naiad, Naiad to Defence (a line-of-battleship), Defence to Colossus and Colossus to Mars, standing in Nelson's line of battle itself, 48 miles from the mouth of Cadiz harbour. The news reached Nelson at 9:30. He immediately ordered ``General chase southeast'' and steered to place the fleet between Cadiz and the Straits of Gibraltar. The opening move of the battle of Trafalgar had begun. (3)

 

 

Regarding Nelson and Trafalgar, it is interesting that Nelson opted not to depend on signals for his C2, choosing, instead, to emphasize doctrine over command and control.  In fact, he conducted the entire battle while sending only two signals, neither of which was of a tactical nature.  Nelson understood that even a simple system like signal flags could, and would, be misapplied and misunderstood in the heat of battle and that the preferred method of exercising C2 was to not attempt to exercise C2.  He chose, instead, to depend on clearly defined doctrine which allowed local commanders (the individual ship Captains) to act without higher command direction and yet still unerringly act with total adherence to his intent.

 

 

Lessons

 

As with the historical battles we’ve examined, nothing has changed from age of sail to today in terms of the lessons about communications and networks.  The same problems that plagued naval communications then still plague us today and will continue to do so on the future naval battlefield.  Weather will still degrade communications.  Confusion will still rear its head.  Signals are still susceptible to detection and interception.  And so on …

 

The technology has changed but the lessons remain constant.

 

Inevitable Confusion – Anyone who has ever played the child’s party game of passing a whispered message around a circle of people and seeing how distorted it comes out at the end can testify to the inevitable confusion and garbling of any communication.  A message that is crystal clear to the sender may be confusing gibberish to the receiver – and that’s without considering the effects of weather, distance, light, jamming, false signals, combat, or whatever other form of disruption occurs.  The sources of confusion may have changed (jamming, false signal induction, cyber attack, etc.) but the effects and the lessons remain the same.  Communications will be misunderstood and the advent of radio and electronic signals has done nothing to mitigate the confusion.  A famous example of this is the Admiral Nimitz’s radio message to Admiral Halsey during the battle of Leyte, in WWII, where the meaningless padding phrase, “the world wonders”, was accidentally left in the message and interpreted by Halsey as a sharp rebuke of his actions and wound up changing his plans.

 

Distance – Distance has been the nemesis of communications since time immemorial.  Even today, electronic signals degrade with distance as interfering phenomena are encountered.  Line of sight communications are impacted by distance and impose constraints and altered behavior by the sending platform. 

 

Interception – The mere sending of a signal, and frequency of signaling, even if the meaning is not understood, can provide the enemy with valuable intelligence (see, “The Battle of Heligoland Bight”). 

 

Here is an anecdotal story from a reader,

 

I spent a while working with a blokey from EW element of the Royal Signals, they would spend an entire exercise doing SIGINT traffic analysis of our own side during exercises in Germany.

It is really scary what can be discerned by traffic analysis alone (all the more prevalent when crypto is so effective - in the context of the battlefield timescale). (5)

 

 

The only truly secure signal, the only signal that cannot be detected, intercepted, and analyzed is the one that is not sent.

 

 

Doctrine and Commander’s Intent – The only sure form of communication is no communication and the best communication is no communication.  This is where doctrine and Commander’s Intent comes in.

 

During the Cold War, the US Navy carrier groups would routinely launch entire strike forces without a single radio transmission and aircraft could conduct air-to-air refueling with no transmissions.  Today, our forces have acquired a severe case of verbal diarrhea and are utterly lost without constant communications. 

 

We need to re-establish doctrine and relearn comms-out operations.  The corollary to this is that we need to drop our obsession with ever more intrusive C2 and return to allowing local commanders to command.  If we allow local commanders to command then we don’t need to communicate with them and that is, after all, the best form of communication.

 

 

Conclusion

 

Despite all the historical and current evidence to the contrary, we still cling to the fantasy of region-wide networks of sensors and shooters all tied together in a mammoth, flawless, seamless, real time network of instantaneously shared data and intel.  Just as the Royal Navy learned that the best form of communication was no communication, we, too, must learn to operate without communication and that means emphasizing doctrine and commander’s intent instead of the ceaseless – and useless – drone of higher authority clogging the airwaves.

 

 

 

________________________________

 

(1)Richard Howe, Signal and Instruction Book, ca. 1776

https://quod.lib.umich.edu/c/clementsead/umich-wcl-M-899how?view=text

 

(2)https://tmg110.tripod.com/british21.htm

 

(3)https://www.royal-signals.org.uk/Datasheets/SignallingFlags.php

 

(4)Wikipedia, “Naval Flag Signaling”, retrieved 25-Jan-2021

https://en.wikipedia.org/wiki/Naval_flag_signalling

 

(5)Navy Matters, “The Battle of Heligoland Bight”, Dave Wolfy, January 6, 2020 at 12:10 PM

https://navy-matters.blogspot.com/2020/01/the-battle-of-heligoland-bight.html?showComment=1578341445590#c1103685532335291121