Merge pull request #39 from AAU-Dat/background

Background

report/src/bib/main.bib

@@ -243,6 +243,17 @@ @article{consensus-spec-phase-0
     url = {https://github.com/ethereum/consensus-specs/blob/dev/specs/phase0/beacon-chain.md}
 }
 
+@misc{heimbach2024deanonymizingethereumvalidatorsp2p,
+    title={Deanonymizing Ethereum Validators: The P2P Network Has a Privacy Issue},
+    author={Lioba Heimbach and Yann Vonlanthen and Juan Villacis and Lucianna Kiffer and Roger Wattenhofer},
+    year={2024},
+    eprint={2409.04366},
+    archivePrefix={arXiv},
+    primaryClass={cs.CR},
+    url={https://arxiv.org/abs/2409.04366},
+}
+https://github.com/ethereum/consensus-specs/blob/dev/specs/phase0/beacon-chain.md
+
 @article{random-selection,
     title        = {Random Selection},
     author       = {@corwintines and @pettinarip and @nhsz and @nalepae},
@@ -258,4 +269,3 @@ @inproceedings{upgrading-ethereum-randomness
     booktitle = {Upgrading Ethereum},
     pages = {134–149}
 }
-

report/src/sections/02-background.tex

@@ -1,6 +1,6 @@
 
 \section{Background}\label{sec:background}
-In this section, we will go through some of the concepts that will be used in the rest of the paper as well as some surrounding context like attacks performed.
+In this section, we will go through some of the concepts that will be used in the rest of the paper as well as some surrounding context for the attack.
 
 \subsection{Ethereum and Proof of Stake}\label{subsec:ethereum-and-proof-of-stake}
 Ethereum is a blockchain platform that allows developers to create decentralized applications using smart contracts.
@@ -17,6 +17,17 @@ \subsection{Ethereum and Proof of Stake}\label{subsec:ethereum-and-proof-of-stak
 If a fork happens the validators have to choose which fork to follow.
 This is done by using the \gls{lmd-ghost} algorithm which chooses the fork with the greatest weight of attestations in its history~\cite{EthereumProof-of-stakePoS}.
 
+\subsection{subnets}\label{subsec:subnets}
+The Ethereum network is split up into smaller networks called subnets.
+Being subscribed to a subnet is also be referred to as being backbone of a subnet.
+These subnets are used to help with the scalability of the network.
+The nodes in the network are split into total of 64 subnets and an additional subnet for attestation aggregates with each node being part of at least two subnets.
+Within a subnet, nodes choose a subset of peers in the same subnet to share its messages with.
+Choosing which notes are a part of this subset is done based on the peers performances.
+Nodes send all messages they hear about within a subnet to these best-performing peers.
+The peers a node can reach within the same subnet is called its fanout~\cite{heimbach2024deanonymizingethereumvalidatorsp2p}.
+
+
 
 %\subsection{Zero-Knowledge Proofs}\label{subsec:zero-knowledge-proofs}
 %A~\gls{zkp} is a cryptographic method that allows one party to prove to another party that something is true without revealing any information.
@@ -160,6 +171,24 @@ \subsection{ENR}\label{subsec:enr}
 you would be able
 to see the corresponding IP addresses and public keys of all the nodes that have been discovered by the node.
 
+
+\subsection{other paper}\label{subsec:other-papers} \todo{new titel}
+\todo{make acronym for the paper}
+In the paper "Deanonymizing Ethereum Validators: The P2P Network Has a Privacy Issue" the authors show that it is possible to deanonymize validators on the Ethereum network by observing attestations and subscribing to subnets~\cite{heimbach2024deanonymizingethereumvalidatorsp2p}.
+This paper is relevant to our work as it shows that it is possible to get information about the validators on the network.
+This paper is also the main inspiration for our attack.
+The paper takes advantage of the attestations, including information such as the IP of the sender node, and subnet setup to get information about the validators.
+For their setup they use a custom version of a Prysm node called RAINBOW that subscribe to all subnets, and they use to log and color the information gathered.
+This information consists of all attestations, their origin, and their origin subnet, all advertised static subscriptions of our peers and precise connection data for all nodes we interact with.
+To help speed up the discovery of the peers they also used a crawler to more quickly find the peers using the discovery protocol and the peer tables.
+In their execution of their experiment they set up four nodes spread out across four different geographical locations.
+They then let the nodes run for three days and managed to deanonymize 235,719 validators and reached out to 11,219 peers.
+These peers were also divided into 4 categories based on their heuristic.
+Those being deanonymized where they located validators on the machine with the heuristic conditions being upheld, No validators where they did not receive a single non-backbone attestation received from the peer, and so they assume that there are no validators on the peer, 64 subnets where they never receive a non-backbone attestation from the peer which makes it impossible to deanonymize the validators and the rest where they got at least one non-backbone attestation but where not able to locate any validators on the peer.
+
+
+
+
 \subsection{Proposer DoS Attack}\label{subsec:proposer-dos-attack}
 In this subsection, we will be describing the attack that we will be using as a basis for our experiment in ~\autoref{sec:experimental-protocol}.
 The attack is a~\gls{dos} attack that aims at hitting the proposers selected for creating blocks in the chain.