I've also looked at the beacon bases and reached a similar conclusion. There only appear to be 4 triangles per row, however, which does make for a nice hexadecimal digit. There is also a correlation with the obelisk glyphs, as pointed out by another poster who developed an application which simulates the glyphs (see <ixalon.github.io/elitedangerous/> for more details). I am not sure what the large triangle is for, unless it is redundancy to show the bit ordering or perhaps a grouping separator. With four bits per group, the values I derived were 0xd and 0x7 for the five triangles on the beacon sides. There are some arrangement of digits that might be implied by the large triangles, so that possible binary groups might be 0xdd, 0x7d, and 0xd7. These may also be aggregated into larger groups, like 0xdd00 and possibly 0xdd00dd00 or 0xdddd for the side with binary groups either above or below the large triangle.
If you go to Ixalon's program, stop the glyphs from cycling through the values, and superimpose the large triangles within the glyphs, you can see various numeric values around the outer edge of the large triangles. Stepping the glyphs through the cycle will show different numbers. There is also a time element to the sequence as it appears each triangle which illuminates in a particular glyph is turned off for about one second out of four, and this appears to be consistent for all illuminated triangles.
Note that with the arrangement of illuminated triangles in the glyphs, you can have a binary digit above and below the large inner triangle, two binary digits above a large bottom triangle, or two binary digits below a large top triangle. There are three illuminated triangles in the glyphs which would not participate in forming a hexadecimal digit because there are in the interior of the glyph. These may indicate whether the large triangle is centered, above, or below the hexadecimal digits. There is also likely to be some meaning in the illuminated small triangles which make up the large triangle, but I am uncertain what this might be at this time. There appear to be 96 unique states distributed through the set of glyphs as shown on the simulator. Each state represents either the illumination or extinguishing of a small triangle, which probably changes the numeric quantity being displayed.
I will try and work on this again tomorrow and see what possible numeric quantities can be derived from this interpretation of the glyphs. Of course, all the obelisks have the same glyph display as far as I have seen, so it's likely questionable as to whether this helps in decoding the appropriate materials for getting data from the obelisks.
