Carlos Luna

Adrián Silveira, Gustavo Betarte, Maximiliano Cristiá et al.

MimbleWimble (MW) is a privacy-oriented cryptocurrency technology which provides security and scalability properties that distinguish it from other protocols of its kind. We present and discuss those properties and outline the basis of a model-driven verification approach to address the certification of the correctness of the protocol implementations. In particular, we propose an idealized model that is key in the described verification process, and identify and precisely state sufficient conditions for our model to ensure the verification of relevant security properties of MW. Since MW is built on top of a consensus protocol, we develop a Z specification of one such protocol and present an excerpt of the $\{log\}$ prototype generated from the Z specification. This $\{log\}$ prototype can be used as an executable model where simulations can be run. This allows us to analyze the behavior of the protocol without having to implement it in a low level programming language. Finally, we analyze the Grin and Beam implementations of MW in their current state of development.

Guido De Luca, Carlos Luna

Android is a platform for mobile devices that captures more than 85% of the total market-share. Currently, mobile devices allow people to develop multiple tasks in different areas. Regrettably, the benefits of using mobile devices are counteracted by increasing security risks. The important and critical role of these systems makes them a prime target for formal verification. In our previous work (LNCS 10855, https://doi.org/10.1007/978-3-319-94460-9_16), we exhibited a formal specification of an idealized formulation of the permission model of version \texttt{6} of Android. In this paper we present an enhanced version of the model in the proof-assistant Coq, including the most relevant changes concerning the permission system introduced on versions Nougat, Oreo, Pie and 10. The properties that we had proved earlier for the security model has been either revalidated or refuted, and new ones have been formulated and proved. Additionally, we make observations on the security of the most recent versions of Android. Using the programming language of Coq we have developed a functional implementation of a reference validation mechanism and certified its correctness. The formal development is about 23k LOC of Coq, including proofs.

Gustavo Betarte, Maximiliano Cristiá, Carlos Luna et al.

Emin Gün Sirer once said: It's clear that writing a robust, secure smart contract requires extreme amounts of diligence. It's more similar to writing code for a nuclear power reactor, than to writing loose web code [...] Yet the current Solidity language and underlying EVM seems designed more for the latter. Formal methods (FM) are mathematics-based software development methods aimed at producing "code for a nuclear power reactor". That is, due application of FM can produce bug-free, zero-defect, correct-by-construction, guaranteed, certified software. However, the software industry seldom use FM. One of the main reasons for such a situation is that there exists the perception (which might well be a fact) that FM increase software costs. On the other hand, FM can be partially applied thus producing high-quality software, although not necessarily bug-free. In this paper we outline some FM related techniques whose application the cryptocurrency community should take into consideration because they could bridge the gap between "loose web code" and "code for a nuclear power reactor".