Bitcoin Forum

As said the title;
Is it possible to know if two or more addresses are related to one wallet? and if from transactions i can figure possible connection (same wallet) between addresses.

I post this here because i think it's a beginner question, no?

If two addresses both sign a transaction then, with near certainty, both addresses are controlled by the same entity. The addresses may be deposit addresses to a business so a customer that provides the address to send bitcoin to may not control both addresses.

There are some instances in which two people/entities will sign one transaction but they are currently rare.

Yes, it is possible. The most simple example I can think of is this. Suppose you have 1 BTC and you send 0.6 BTC to your friend. In this case, you have a change of 0.4 BTC that goes to your change address, the other 0.6 BTC goes to your friend. Now you have 2 addressess: the first one with 0 BTC and the second one with 0.4 BTC that anyone knows are belong to the same person.

There is another example. Suppose you have many addressess with small amounts of BTC in your wallet. Let's say you have 10 addresses with 0.1 BTC each. But you need to send  0.91 BTC to your friend. In this case all your 0.1 amounts are gathered in 1 transaction and sent to your freind's address. Everyone will see that you used 10 addresses to complete that transaction. You will also receive 0.09 BTC to your change address which will also be seen in the blockchain. Now everyone will know that those 11 addresses belong to the same person.

And there's also a growing number of analysis tools and services that could provide particular data which saves time and effort. I think the most used right now is walletexplorer.com which does address grouping.

I think you are referring to addresses generated by a deterministic wallet from the same seed. If so, then the answer is no (please correct me if I am wrong), there is no way to tell if different addresses are generated from the same seed unless the master public key is revealed.

I have seen in wallet explorers that a lot of the transactions has the “change” address. Like the left overs of a certain wallet going to the part where you also own it. I’m not sure if every “change” address in a certain transaction is correct but I think it could be possible but not 100% certainty.

I second this, it's more likely the Unspent Transaction Outputs or UTXOs.
https://i.imgur.com/PlHWtNy.jpg

You can read it here, it's actually a good article from coinsutra.
What Are Unspent Transaction Outputs (UTXOs)? (https://www.google.com/amp/s/coinsutra.com/unspent-transaction-outputs-utxos/amp/)

I think it's safe to assume that it comes in the same wallet (same phrases etc), right?

No one should be able to see if you are the owner of the addresses in the same wallet, as long as you don't mix output from those addresses with each other. And as long as the bitcoin didn't come from the same addresses before they ended in the addresses you have them in now.

I would not be surprised to see that some of those addresses with 10,000 bitcoins are to be found in the same wallet. But as long as the owner doesn't mix the coins no one but the owner of the wallet will ever know.

https://bitinfocharts.com/top-100-richest-bitcoin-addresses.html

May be this is what am really looking to understand, after know that it's not possible to link addresses, so what's it's all about those analysis tools and what help for?

Yes it possible to know if two bitcoin address belongs to the same wallet there are tool that I design for that. And some members of this forum with good skill make use of that tool from time to time to trace the ownership of addresses when the need arise.

Those tools (like walletexplorer.com) simply analyse transactions inputs. If a given transaction has many addresses as inputs then it will assume they belong to the same wallet, which is not always true (coinjoin transactions made by some wallets like Wasabi can be a good counterexample).

From the context of what you just posted, with just looking at the transaction details, observing the change addresses help. However if you want to analyze 1 address from 1 random-picked address and determine whether they came from the same wallet? I don't think that will be pretty easy to do, or perhaps even make sense at all.


Analysis tools are usually for following the 'trail' to which the coins are sent to, and are used mostly for nabbing criminals and people who use bitcoins for bad intentions. It can also help in linking a connection in between different addresses, though knowing whether the addresses came from a single wallet? I don't think so.

Ofcourse it is. Every bitcoin is coming from somewhere.

Unless you use some sort of mixer, there is a really big chance certain tools, (especially from FVEY Five Eyes (https://en.wikipedia.org/wiki/Five_Eyes)) can trace your Bitcoin from where you currently stored it, to where you first bought it, which date, which exchange, which payment method you used.

*Almost* every exchange nowadays requires KYC, and has your banking info. it's really a myth that bitcoin offers any privacy at all when used against the government without proper measures as i described above.

As i said, a mixer breaks this chain. But yeah, if the mixer is controlled by the group described above, it's not of much help. (Which could be the case for any mixer really, we simply don't know who's behind them.)
Which is quite freaking dangerous when you think about it. I have no idea if Chipmixer isn't controlled by the CIA, serving as a huge honeypot.

Anyway.

That would be an interesting story to tell. Imagine one of the best mixers are actually your own government.

If users want complete privacy, then they should never touch exchanges that require KYC and maybe just use privacy oriented coins instead of pseudo-anonymous coins like BTC.

It is hard to find a respected exchange that doesn't require any sort of KYC. Traders usually want low fees and enough liquidity and no bot trading as well.
Binance for example allows withdrawals up to 5 BTC in a 24 hour period and for most traders that can be enough. But it is kind of difficult to get listed there and expensive.
You are left with DEX solutions like IDEX where new erc-20 projects list as their first exchange.

Just from the transaction, depending on how you execute the transactions and if you use any interpolating xfers, maybe.

If you created a wallet from a seed, and use different addresses from that seed, and someone, lets say law enforcement gets your seed,
then yes, they can prove the different addys are connected.

/KX

I've read through this thread and I still have one question: Isn't it that when you spend from an address, that a public key gets exposed? couldn't that public key be used to generate addresses? Or maybe this public key is different from what they called the "Master key'?

It's different from a wallet's "master public key" (xpub).
The exposed public key from a transaction input is just a child of the xpub key (if it's from an HD wallet)
A child key can't be used to retrieve its parent key, because deriving the child from the parent involves a cryptographic hash function which is irreversible.

Great answer. Thank you

Further to nc50lc's answer, the public key which is revealed when you spend from an address also cannot be used to generate new child keys, as this process requires something known as the chain code. The chain code is something which is generated from the parent keys, but it is not stored anywhere on the blockchain and is not revealed when you make a transaction; it is only generated and used locally by your wallet software.

Simple answer to this question is - -

No...

Though there are exceptions, but before I discus the exceptions, let me say that, if it was possible to know that two or more addresses belong to one wallet, then there wouldn't be any need to implement the "automatic generate new address" feature each time transaction is received, which is found in some wallets like coinomi.
And secondly, the generate new address feature also serves as a way to keep a user's privacy since it keeps preying eyes away from knowing the amount of bitcoins that's contained in the wallet, if it was possible to know if two or more addresses belong to one wallet, then again, the generate new address feature will be useless here.

Now, talking about exceptions, I am not a technical person, but I am just imagining that there might be a way, or better put, loophole through which those who have good  technical knowledge of wallets (and how they work on the development side of it) might know how to figure out if two or more addresses belong to a single wallet, this I don't know and have not heard of, I am just speculating here, and I hope I am wrong.

Technically it is impossible to know if two addresses are linked (whether from the same wallet or have any connection whatsoever). All we can do and all the blockchain analysis companies do is report the likelihood of it being linked.

For example one thing they do is to speculate which output is the change address (eg. round number is payment, the other is change) and claim that it belongs to the same wallet. But this is speculation and we still can't say for sure. The round output can be the change or it may not even have a change.
Another thing is claiming that inputs of a transaction coming from 2 different addresses links those 2 addresses but again this is speculation. The two people could have joined their inputs to "mix" them.

You're missing the point by a lot. Anyway, this thread was created 3 years ago. I just pumped it up a few days ago to ask a related question, but there is no point in answering in this thread anymore.