Data that shows how many times an address is reused (and when) will be very helpful to figure out if most 1-2 transactions are actually spending the only input of an address or one of several inputs from it.
I'm not sure if u had a look at the paper or not, it does too analyze miner coinbase transactions differently (Fig10, which I didn't attach), but there r 2 comments:
.
1- u have to know the paper and even the official Bitcoin site does NOT recommend reusing the same address because it reduces anonymity. I just use the assumption to analyze the number of UTXOs and the tree structure ( considering replacement analogous to using the same address without the bad side effects; I'm just using the same node in the Merkle Tree with a different key address)
.
2-The most common 1:2 transactions u r talking about, even without Transaction Graph analysis, implies that the expected no of UTXOs in a given Bitcoin state is roughly the total number of transactions since the beginning of time.
It is like a recursive formula calling "no of UTXOs in a given block (i)", T(i), then:
T(i) = T(i-1)+Sum{(outputs-inputs)}for all TXs in block iIf we assume the av difference is one, then roughly speaking
T(i)=T(i-1)+no.of TXs
= Sum(no. of TXs)from"0"to"i"
.
I think this is straight forward although no one state it directly.
What I'm trying to do is minimize the tree operations to this limit by never delete if u r going to insert in the same next step for the same transaction. Can this transaction graph analysis help me predict the tree structure ( how balanced it is, or lend itself to degenerate?)
.
3-Can I find a programming partner in this group (software developer to do the coding steps & get the test results) obviously will be partner in the paper with name written and everything???