Consider the case of Tintagel on the north Cornish coast. The post-Roman kingdom of the Dumnovians (modern-day Devon and Cornwall) was in a unique position to trade across the Atlantic coast of Europe, and with its valuable reserves of tin (and also gold and copper, to a lesser degree), it became a powerful place for a century or more.

Tintagel is often held up as a possible Arthurian site, and certainly regional folklore has made that association. There was much celebration when archaeologist found an early etching on stone with a name that seems to recall something close to Arthur.

Some work in stone (largely because it was such a readily available building material) as well as turf, wood, and thatch, would have been the choices of the day. See this site as well - but there are many others.

There are many threads that fed into Arthurian folklore and literature, and unraveling that tangle - which reaches sometimes into misty semi-literate times - is a challenge. The following is what I wrote long ago for this subreddit, and I incorporated it into my Introduction to Folklore, which I used when teaching folklore at university; the answer deals with other motifs, and then Arthur:

When I see the posts [i.e., about whether something in legend is real - recognizing that's not the question asked here!] like this asking about whether there were real people or events behind legends, myths, and/or the ancient gods, I respond with several observations. First, the idea that the gods and heroes of legend are based on real people had an early proponent in the Greek, late-fourth-century BCE writer, Euhemerus, giving his name to this approach to myth and legend: Euhemerism. Folklorists generally regard the idea that there was an actual basis for most oral tradition as barking up the wrong tree, because the original “real” event behind a story is usually elusive and searching for that core is a futile exercise. In addition, research into how stories began usually concludes that they emerge in a rather spontaneous way, typically without an actual incident to inspire them.

A few examples: the Classical Greek story of Perseus is an early manifestation of a widespread folktale, catalogued by the twentieth-century folklorists Antti Aarne and Stith Thompson as AT 300, “The Dragon Slayer.” Was there a proto-Perseus who rescued a maiden from some sort of extraordinary threat or perhaps from some sort of human sacrifice? It is hard to answer that question, but it is not hard to imagine how far back in time that proto-incident would have had to occur: AT 300 is spread throughout Eurasia. It was collected from cultures that could have no conceivable literary connection with ancient Greece, and yet the shared assortment of motifs in the numerous variants clearly show some sort of genetic, that is, historically connected relationship. Would we need, therefore, to go back thousands of years before the first recordation of the Perseus story to find this proto-Perseus? It is much easier to understand that the folktale simply diffused and that one of its manifestations was in ancient Greece.

Now, let’s consider another example that has inspired a lot of spilled ink. A simple Google search for the “origins of King Arthur” provides more websites than one could easily read in a week. Was there a proto-Arthur? Perhaps. Maybe there were several. But what does that prove? Every society has remarkable characters, and it may be a natural process for these sorts of individuals to attract all manner of traditional stories that have nothing to do with the original inspiration of the cycle of legends.

So, what do we have with Arthur? Was there a core source (or sources) for this legendary character? Let’s concede for the sake of argument that the answer is yes. Now, did this individual have a great warrior at his side who became ensnared by the leader’s wife in the fashion of Lancelot and Guinevere? That is more problematic since this type of story is also associated with Diarmuid and Grainne in the Irish court of King Finn and with the Cornish stories of Tristan and Isolde in the court of King Mark. One could even argue that it is the story behind Helen of Troy. In fact, it appears that this was a widespread type of story that became associated with various courts of historical legend. We cannot conclude that every great king had a queen who was attracted to one of his warriors and coerced him to take her away. This is simply a story that was attached to cycles involving great courts. In short, the further one goes back to find the “real Arthur,” the less the candidate (or candidates) look like the King Arthur who has been beloved for centuries. The proto Arthurs are not really King Arthur. They may be seeds but they look nothing like the tree that would grow over the centuries. We do not hold an acorn and say “Ah, I have in my hand a mighty oak tree.” It is not yet a tree. It is a seed. And the two look very different even if they are genetically linked.

One more example: there is a widespread legend told by countless families of the ghostly appearance of a loved one in anticipation of news that the individual died. This became a popular tradition in post-Famine Ireland because so many relatives lived in North America or elsewhere. But it is frequently told by all sorts of people internationally. So, we can ask, are there real-life, actual inspirations for this legend? That is, do the spirits of the dead actually come to visit loved ones? Well, how the hell should I know? To paraphrase a famous line from the television show “Star Trek,” “Damn it Jim, I’m a folklorist, not a ghost hunter.” And I have no intention of becoming a ghost hunter. It doesn’t matter what is behind stories so much as it does that people tell these stories. I’m in it for that part of the game; I consider stories as they are told over time, to gain from that material some insight into the past, into culture, and into the human condition. I am a folklorist. And with that, my plate is full.

When thinking about Post-Roman Britain you have to be aware that a certain amount of Roman buildings remained visible and at least partly usable. There is quite an amount of archaeological evidence from different sites that were once built and maintained by the roman army that were used in a similar way in Post-Roman Britain.

Especially the impressive 3rd century stoneworks of the socalled "Saxon shore forts" were massive and survived the Roman occupation for centuries. To give on example you can take a look at Portchester castle. This is an impressive fortification, likely built between 285 - 290 under the reign of Diocletian and was one of the aforementioned Saxon Shore forts and thought to have been a base of the Roman fleet. Its strategic importance is evident, as it is guarding Portsmouth harbour.

Excavations were carried out and published by Barry Cunliffe, thoug I believe that further work is ongoing. Todays ruins still feature the rectangular Roman wall, whereas the medieval castle is set in one corner, partly reusing the Roman walls. The archaeological evidence is suggesting a hall and a tower for the tenth century. This does leave a gap were the archaeological evidence of construction work within the site gives nothing - but it is thought that the place was never completely abandoned.

In some places on Hadrians' wall there was also some archaeological evidence that suggested someone took over at least some control while reusing Roman features. The picture is not very clear though, at least as far as I am aware.

-Barry Cunliffe, Excavations at Portchester Castle. 1: Roman, London 1976.

-Barry Cunliffe, Excavations at Portchester Castle 4: Medieval, the inner bailey, London 1985.

-Andrew Pearson, The Roman Shore Forts: Coastal defences of Southern Britain, Tempus Publishing 2002.

For and overview and more specific literature on Roman military features in Britain, there is:

-David Breeze, Roman Forts in Bitain, Shire publications 1994. If I remember correctly, he collected literature on every installation and also remarks when there are traces of Late Antiquity and early medieval use.

-Bari Jones and David Mattingly, An Atlas of Roman Britain, Oxbow 1990. They discuss the period after the Romans at the end of the book.

-Nicholas J. Higham and Matin J. Ryan, The Anglo-Saxon World, Yale University Press 2013 might give you an impression of the Anglo-Saxon times.