A system-theoretic framework for privacy preservation in continuous-time multiagent dynamics

Altafini, Claudio

doi:10.1016/j.automatica.2020.109253

Cited by 68 publications

(38 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In our next statement, we investigate whether a malicious agent can derive the local reference value of an agent if it knows condition (13) and all the transmitted messages to and from the agent.…”

Section: Theorem 34 (A Necessary Condition On Admissible Perturbation...mentioning

confidence: 99%

“…Let the interaction topology G of the agents implementing algorithm (9) with ∆ ∈ (0, min{2, ∆}), initialized at x i (0) ∈ R and v i (0) = 0, i ∈ V, be strongly connected and weight-balanced digraph. Suppose agents are implementing locally chosen admissible perturbation signals that satisfy (13), and let the knowledge set of malicious agent 1 be as in Definition 1. If agent 1 is the in-neighbor of agent i ∈ V\{1} and all its out-neighbors, then agent 1 can employ the observer…”

Section: Theorem 34 (A Necessary Condition On Admissible Perturbation...mentioning

confidence: 99%

“…Lastly, the guarantees established in all the methods discussed above are only for connected undirected graphs. Interested reader can also find privacy preservation methods for the continuous-time implementation of algorithm (1) in [13] and [14].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Deterministic Privacy Preservation in Static Average Consensus Problem

Esteki

Kia

2021

2021 American Control Conference (ACC)

View full text Add to dashboard Cite

In this paper we consider the problem of privacy preservation in the static average consensus problem. This problem normally is solved by proposing privacy preservation augmentations for the popular first order Laplacian-based algorithm. These mechanisms however come with computational overhead, may need coordination among the agents to choose their parameters and also alter the transient response of the algorithm. In this paper we show that an alternative iterative algorithm that is proposed in the literature in the context of dynamic average consensus problem has intrinsic privacy preservation and can be used as a privacy preserving algorithm that yields the same performance behavior as the well-known Laplacian consensus algorithm but without the overheads that come with the existing privacy preservation methods.

show abstract

Section: Theorem 34 (A Necessary Condition On Admissible Perturbation...mentioning

confidence: 99%

Section: Theorem 34 (A Necessary Condition On Admissible Perturbation...mentioning

confidence: 99%

See 1 more Smart Citation

Deterministic Privacy Preservation in Static Average Consensus Problem

Esteki

Kia

2021

2021 American Control Conference (ACC)

View full text Add to dashboard Cite

show abstract

“…In [15], the authors proposed an interesting privacy-preserving consensus algorithm, which adds edge-based perturbations. There are also some other methods to preserve privacy, such as the state decomposition [19], output masking [20], hotpluggable methods [21] and observability based methods [22], [23], [24].…”

Section: Introductionmentioning

confidence: 99%

Privacy-Preserved Average Consensus Algorithms with Edge-based Additive Perturbations

Xiong¹,

Li²

2021

Preprint

View full text Add to dashboard Cite

In this paper, we consider the privacy preservation problem in both discrete-and continuous-time average consensus algorithms with strongly connected and balanced graphs, against either internal honest-but-curious agents or external eavesdroppers. A novel algorithm is proposed, which adds edge-based perturbation signals to the process of consensus computation. Our algorithm can be divided into two phases: a coordinated scrambling phase, which is for privacy preservation, and a convergence phase. In the scrambling phase, each agent is required to generate some perturbation signals and add them to the edges leading out of it. In the convergence phase, the agents update their states following a normal updating rule. It is shown that an internal honest-but-curious agent can obtain the privacy of a target agent if and only if no other agents can communicate with the target agent. As for external eavesdroppers, it is proved that this kind of attackers can never obtain any agent's privacy.

show abstract

“…Frequently used methods include isomorphic (linear or affine) transformation, homomorphic encryption schemes (e.g., RSA, ElGamal, and Paillier), and time-varying transformation [5]. Although some encryption schemes like Paillier and ElGamal can transform (encrypt) one plaintext into different ciphertexts by selecting a different random parameter, the inverse transformation (decryption) maps those ciphertexts back on to the same plaintext again.…”

mentioning

confidence: 99%