Search

We present Elevator, a novel algorithm for hub samplingin peer-to-peer networks, enabling the construction of overlays with atopology between a random graph and a star network, and networksthat have both hubs and are resilient to failures. Our approach emergesfrom principles of preferential attachment, forming hubs spontaneously,offering an innovative solution for decentralized networks that can benefituse cases requiring a network with both low diameter and resilience tofailures.

Traditional machine learning systems were designed in a centralized manner. In such designs, the central entity maintains both the machine learning model and the data used to adjust the model's parameters. As data centralization yields privacy issues, Federated Learning was introduced to reduce data sharing and have a central server coordinate the learning of multiple devices. While Federated Learning is more decentralized, it still relies on a central entity that may fail or be subject to attacks, provoking the failure of the whole system. Then, Decentralized Federated Learning removes the need for a central server entirely, letting participating processes handle the coordination of the model construction. This distributed control urges studying the possibility of malicious attacks by the participants themselves. While poisoning attacks on Federated Learning have been extensively studied, their effects in Decentralized Federated Learning did not get the same level of attention. Our work is the first to propose a methodology to assess poisoning attacks in Decentralized Federated Learning in both churn free and churn prone scenarios. Furthermore, in order to evaluate our methodology on a case study representative for gossip learning we extended the gossipy simulator with an attack injector module.

We consider a collection of $k \geq 2$ robots that evolve in a ring-shaped network without common orientation, and address a variant of the crash-tolerant gathering problem called the \emph{Stand-Up Indulgent Gathering} (SUIG): given a collection of robots, if no robot crashes, robots have to meet at the same arbitrary location, not known beforehand, in finite time; if one robot or more robots crash on the same location, the remaining correct robots gather at the location of the crashed robots. We aim at characterizing the solvability of the SUIG problem without multiplicity detection capability.

We consider a strong variant of the crash fault-tolerant gathering problem called stand-up indulgent gathering (SUIG), by robots endowed with limited visibility sensors and lights on line-shaped networks. In this problem, a group of mobile robots must eventually gather at a single location, not known beforehand, regardless of the occurrence of crashes. Differently from previous work that considered unlimited visibility, we assume that robots can observe nodes only within a certain fixed distance (that is, they are myopic), and emit a visible color from a fixed set (that is, they are luminous), without multiplicity detection. We consider algorithms depending on two parameters related to the initial configuration: $M_{init}$, which denotes the number of nodes between two border nodes, and $O_{init}$, which denotes the number of nodes hosting robots. Then, a border node is a node hosting one or more robots that cannot see other robots on at least one side. Our main contribution is to prove that, if $M_{init}$ or $O_{init}$ is odd, SUIG can be solved in the fully synchronous model.

We investigate the solvability of the Byzantine Reliable Broadcast and Byzantine Broadcast Channel problems in distributed systems affected by Mobile Byzantine Faults. We show that both problems are not solvable even in one of the most constrained system models for mobile Byzantine faults defined so far. By endowing processes with an additional local failure oracle, we provide a solution to the Byzantine Broadcast Channel problem.

In this paper, we revisit the problem of classical \textit{meeting times} of random walks in graphs. In the process that two tokens (called agents) perform random walks on an undirected graph, the meeting times are defined as the expected times until they meet when the two agents are initially located at different vertices. A key feature of the problem is that, in each discrete time-clock (called \textit{round}) of the process, the scheduler selects only one of the two agents, and the agent performs one move of the random walk. In the adversarial setting, the scheduler utilizes the strategy that intends to \textit{maximizing} the expected time to meet. In the seminal papers \cite{collisions,israeli1990token,tetali1993simult}, for the random walks of two agents, the notion of \textit{atomicity} is implicitly considered. That is, each move of agents should complete while the other agent waits. In this paper, we consider and formalize the meeting time of \textit{non-atomic} random walks. In the non-atomic random walks, we assume that in each round, only one agent can move but the move does not necessarily complete in the next round. In other words, we assume that an agent can move at a round while the other agent is still moving on an edge. For the non-atomic random walks with the adversarial schedulers, we give a polynomial upper bound on the meeting times., Comment: 18 pages, 1 figure

We consider a variant of the crash-fault gathering problem called stand-up indulgent gathering (SUIG). In this problem, a group of mobile robots must eventually gather at a single location, which is not known in advance. If no robots crash, they must all meet at the same location. However, if one or more robots crash at a single location, all non-crashed robots must eventually gather at that location. The SUIG problem was first introduced for robots operating in a two-dimensional continuous Euclidean space, with most solutions relying on the ability of robots to move a prescribed (real) distance at each time instant. In this paper, we investigate the SUIG problem for robots operating in a discrete universe (i.e., a graph) where they can only move one unit of distance (i.e., to an adjacent node) at each time instant. Specifically, we focus on line-shaped networks and characterize the solvability of the SUIG problem for oblivious robots without multiplicity detection.

We consider a swarm of mobile robots evolving in a bidimensional Euclidean space. We study a variant of the crash-tolerant gathering problem: if no robot crashes, robots have to meet at the same arbitrary location, not known beforehand, in finite time; if one or several robots crash at the same location, the remaining correct robots gather at the crash location to rescue them. Motivated by impossibility results in the semi-synchronous setting, we present the first solution to the problem for the fully synchronous setting that operates in the vanilla Look-Compute-Move model with no additional hypotheses: robots are oblivious, disoriented, have no multiplicity detection capacity, and may start from arbitrary positions (including those with multiplicity points). We furthermore show that robots gather in a time that is proportional to the initial maximum distance between robots., Comment: arXiv admin note: substantial text overlap with arXiv:2010.04400

We study a novel graph path planning problem for multiple agents that may crash at runtime, and block part of the workspace. In our setting, agents can detect neighboring crashed agents, and change followed paths at runtime. The objective is then to prepare a set of paths and switching rules for each agent, ensuring that all correct agents reach their destinations without collisions or deadlocks, despite unforeseen crashes of other agents. Such planning is attractive to build reliable multi-robot systems. We present problem formalization, theoretical analysis such as computational complexities, and how to solve this offline planning problem., Comment: to be presented at AAAI-23

We present QUANTAS: a simulator that enables quantitative performance analysis of distributed algorithms. It has a number of attractive features. QUANTAS is an abstract simulator, therefore, the obtained results are not affected by the specifics of a particular network or operating system architecture. QUANTAS allows distributed algorithms researchers to quickly investigate a potential solution and collect data about its performance. QUANTAS programming is relatively straightforward and is accessible to theoretical researchers. To demonstrate QUANTAS capabilities, we implement and compare the behavior of two representative examples from four major classes of distributed algorithms: blockchains, distributed hash tables, consensus, and reliable data link message transmission.

We consider transportation networks that are modeled by dynamic graphs, and introduce the possibility for traveling agents to use Backward Time-Travel (BTT) devices at any node to go back in time (to some extent, and with some appropriate fee) before resuming their trip. We focus on dynamic line graphs. In more detail, we propose exact algorithms to compute travel plans with constraints on the BTT cost or on how far back in time you can go, while minimizing travel delay (that is, the time difference between the arrival instant and the starting instant), in polynomial time. We study the impact of the BTT devices pricing policies on the computation process of those plans considering travel delay and cost, and provide necessary properties that pricing policies should satisfy to enable to compute such plans. Finally, we provide an optimal online algorithm for the unconstrained problem when the cost function is the identity.

The vast majority of existing Distributed Computing literature about mobile robotic swarms considers computability issues: characterizing the set of system hypotheses that enables problem solvability. By contrast, the focus of this work is to investigate complexity issues: obtaining quantitative results about a given problem that admits solutions. Our quantitative measurements rely on a newly developed simulation framework to benchmark pen and paper designs. First, we consider the maximum traveled distance when gathering robots at a given location, not known beforehand (both in the two robots and in the n robots settings) in the classical OBLOT model, for the FSYNC, SSYNC, and ASYNC schedulers. This particular metric appears relevant as it correlates closely to what would be real world fuel consumption. Then, we introduce the possibility of errors in the vision of robots, and assess the behavior of known rendezvous (aka two robots gathering) and leader election protocols when sensors are unreliable. We also introduce two new algorithms, one for fuel efficient convergence, and one for leader election, that operate reliably despite unreliable sensors.

In this paper, we consider the Byzantine reliable broadcast problem on authenticated and partially connected networks. The state-of-the-art method to solve this problem consists in combining two algorithms from the literature. Handling asynchrony and faulty senders is typically done thanks to Gabriel Bracha's authenticated double-echo broadcast protocol, which assumes an asynchronous fully connected network. Danny Dolev's algorithm can then be used to provide reliable communications between processes in the global fault model, where up to f processes among N can be faulty in a communication network that is at least 2f+1-connected. Following recent works that showed that Dolev's protocol can be made more practical thanks to several optimizations, we show that the state-of-the-art methods to solve our problem can be optimized thanks to layer-specific and cross-layer optimizations. Our simulations with the Omnet++ network simulator show that these optimizations can be efficiently combined to decrease the total amount of information transmitted or the protocol's latency (e.g., respectively, -25% and -50% with a 16B payload, N=31 and f=4) compared to the state-of-the-art combination of Bracha's and Dolev's protocols., Comment: This is an extended version of a paper that appeared at the IEEE ICDCS 2021 conference

Previous works on formally studying mobile robotic swarms consider necessary and sufficient system hypotheses enabling to solve theoretical benchmark problems (geometric pattern formation, gathering, scattering, etc.). We argue that formal methods can also help in the early stage of mobile robotic swarms protocol design, to obtain protocols that are correct-by-design, even for problems arising from real-world use cases, not previously studied theoretically. Our position is supported by a concrete case study. Starting from a real-world case scenario, we jointly design the formal problem specification, a family of protocols that are able to solve the problem, and their corresponding proof of correctness, all expressed with the same formal framework. The concrete framework we use for our development is the PACTOLE library based on the COQ proof assistant.

We consider the gathering problem for asynchronous and oblivious robots that cannot communicate explicitly with each other, but are endowed with visibility sensors that allow them to see the positions of the other robots. Most of the investigations on the gathering problem on the discrete universe are done on ring shaped networks due to the number of symmetric configuration. We extend in this paper the study of the gathering problem on torus shaped networks assuming robots endowed with local weak multiplicity detection. That is, robots cannot make the difference between nodes occupied by only one robot from those occupied by more than one robots unless it is their current node. As a consequence, solutions based on creating a single multiplicity node as a landmark for the gathering cannot be used. We present in this paper a deterministic algorithm that solves the gathering problem starting from any rigid configuration on an asymmetric unoriented torus shaped network., Comment: 41 pages

We consider the problem of constructing a maximum independent set with mobile myopic luminous robots on a grid network whose size is finite but unknown to the robots. In this setting, the robots enter the grid network one-by-one from a corner of the grid, and they eventually have to be disseminated on the grid nodes so that the occupied positions form a maximum independent set of the network. We assume that robots are asynchronous, anonymous, silent, and they execute the same distributed algorithm. In this paper, we propose two algorithms: The first one assumes the number of light colors of each robot is three and the visible range is two, but uses additional strong assumptions of port-numbering for each node. To delete this assumption, the second one assumes the number of light colors of each robot is seven and the visible range is three. In both algorithms, the number of movements is $O(n(L+l))$ steps where $n$ is the number of nodes and $L$ and $l$ are the grid dimensions.

In this paper, we focus on the uniform bipartition problem in the population protocol model. This problem aims to divide a population into two groups of equal size. In particular, we consider the problem in the context of \emph{arbitrary} communication graphs. As a result, we clarify the solvability of the uniform bipartition problem with arbitrary communication graphs when agents in the population have designated initial states, under various assumptions such as the existence of a base station, symmetry of the protocol, and fairness of the execution. When the problem is solvable, we present protocols for uniform bipartition. When global fairness is assumed, the space complexity of our solutions is tight.

We consider two mobile oblivious robots that evolve in a continuous Euclidean space. We require the two robots to solve the rendezvous problem (meeting in finite time at the same location, not known beforehand) despite the possibility that one of those robots crashes unpredictably. The rendezvous is stand up indulgent in the sense that when a crash occurs, the correct robot must still meet the crashed robot on its last position. We characterize the system assumptions that enable problem solvability, and present a series of algorithms that solve the problem for the possible cases.

Quixo is a two-player game played on a 5$\times$5 grid where the players try to align five identical symbols. Specifics of the game require the usage of novel techniques. Using a combination of value iteration and backward induction, we propose the first complete analysis of the game. We describe memory-efficient data structures and algorithmic optimizations that make the game solvable within reasonable time and space constraints. Our main conclusion is that Quixo is a Draw game. The paper also contains the analysis of smaller boards and presents some interesting states extracted from our computations., Comment: 19 pages

Self-stabilizing protocols enable distributed systems to recover correct behavior starting from any arbitrary configuration. In particular, when processors communicate by message passing, fake messages may be placed in communication links by an adversary. When the number of such fake messages is unknown, self-stabilization may require huge resources: (a) generic solutions (a.k.a. data link protocols) require unbounded resources, which makes them unrealistic to deploy and (b) specific solutions (e.g., census or tree construction) require $O(n\log n)$ or $O(\Delta\log n)$ bits of memory per node, where $n$ denotes the network size and $\Delta$ its maximum degree, which may prevent scalability. We investigate the possibility of resource efficient self-stabilizing protocols in this context. Specifically, we present a self-stabilizing protocol for $(\Delta+1)$-coloring in any n-node graph, under the asynchronous message-passing model. It is deterministic, it converges in $O(k\Delta n^2\log n)$ message exchanges, where $k$ is the bound of the link capacity in terms of number of messages, and it uses messages on $O(\log\log n+\log\Delta)$ bits with a memory of $O(\Delta\log\Delta+\log\log n)$ bits at each node. The resource consumption of our protocol is thus almost oblivious to the number of nodes, enabling scalability. Moreover, a striking property of our protocol is that the nodes do not need to know the number, or any bound on the number of messages initially present in each communication link of the initial (potentially corrupted) network configuration. This permits our protocol to handle any future network with unknown message capacity communication links. A key building block of our coloring scheme is a spanning directed acyclic graph construction, that is of independent interest, and can serve as a useful tool for solving other tasks in this challenging setting.

We investigate gathering algorithms for asynchronous autonomous mobile robots moving in uniform ring-shaped networks. Different from most work using the Look-Compute-Move (LCM) model, we assume that robots have limited visibility and lights. That is, robots can observe nodes only within a certain fixed distance, and emit a color from a set of constant number of colors. We consider gathering algorithms depending on two parameters related to the initial configuration: $M_{init}$, which denotes the number of nodes between two border nodes, and $O_{init}$, which denotes the number of nodes hosting robots between two border nodes. In both cases, a border node is a node hosting one or more robots that cannot see other robots on at least one side. Our main contribution is to prove that, if $M_{init}$ or $O_{init}$ is odd, gathering is always feasible with three or four colors. The proposed algorithms do not require additional assumptions, such as knowledge of the number of robots, multiplicity detection capabilities, or the assumption of towerless initial configurations. These results demonstrate the power of lights to achieve gathering of robots with limited visibility., Comment: This is a full version of the conference paper in OPODIS2019

The paper details the first successful attempt at using model-checking techniques to verify the correctness of distributed algorithms for robots evolving in a \emph{continuous} environment. The study focuses on the problem of rendezvous of two robots with lights. There exist many different rendezvous algorithms that aim at finding the minimal number of colors needed to solve rendezvous in various synchrony models (e.g., FSYNC, SSYNC, ASYNC). While these rendezvous algorithms are typically very simple, their analysis and proof of correctness tend to be extremely complex, tedious, and error-prone as impossibility results are based on subtle interactions between robots activation schedules. The paper presents a generic verification model written for the SPIN model-checker. In particular, we explain the subtle design decisions that allow to keep the search space finite and tractable, as well as prove several important theorems that support them. As a sanity check, we use the model to verify several known rendezvous algorithms in six different models of synchrony. In each case, we find that the results obtained from the model-checker are consistent with the results known in the literature. The model-checker outputs a counter-example execution in every case that is known to fail. In the course of developing and proving the validity of the model, we identified several fundamental theorems, including the ability for a well chosen algorithm and ASYNC scheduler to produce an emerging property of memory in a system of oblivious mobile robots, and why it is not a problem for luminous rendezvous algorithms.

In this paper, we consider the problem of scattering a swarm of mobile oblivious robots in a continuous space. We consider the fully asynchronous setting where robots may base their computation on past observations, or may be observed by other robots while moving. It turns out that asynchronous scattering is solvable in the most general case when both vision (the ability to see others robots positions) and weak local multiplicity detection are available. In the case of a bidimensional Euclidean space, ASYNC scattering is also solvable with blind robots if moves are rigid. Our approach is constructive and modular, as we present a proof technique for probabilistic robot protocols that is of independent interest and can be reused for other purposes. On the negative side, we show that when robots are both blind and have no multiplicity detection, the problem is unsolvable, and when only one of those is available, the problem remains unsolvable on the line.

We revisit Byzantine tolerant reliable broadcast with honest dealer algorithms in multi-hop networks. To tolerate Byzantine faulty nodes arbitrarily spread over the network, previous solutions require a factorial number of messages to be sent over the network if the messages are not authenticated (e.g. digital signatures are not available). We propose modifications that preserve the safety and liveness properties of the original unauthenticated protocols, while highly decreasing their observed message complexity when simulated on several classes of graph topologies, potentially opening to their employment.

Ensuring reliable communication despite possibly malicious participants is a primary objective in any distributed system or network. In this paper, we investigate the possibility of reliable broadcast in a dynamic network whose topology may evolve while the broadcast is in progress. In particular, we adapt the Certified Propagation Algorithm (CPA) to make it work on dynamic networks and we present conditions (on the underlying dynamic graph) to enable safety and liveness properties of the reliable broadcast. We furthermore explore the complexity of assessing these conditions for various classes of dynamic networks.

We investigate exploration algorithms for autonomous mobile robots evolving in uniform ring-shaped networks. Different from the usual Look-Compute-Move (LCM) model, we consider two characteristics: myopia and luminosity. Myopia means each robot has a limited visibility. We consider the weakest assumption for myopia: each robot can only observe its neighboring nodes. Luminosity means each robot maintains a non-volatile visible light. We consider the weakest assumption for luminosity: each robot can use only two colors for its light. The main interest of this paper is to clarify the impact of luminosity on exploration with myopic robots. As a main contribution, we prove that 1) in the fully synchronous model, two and three robots are necessary and sufficient to achieve perpetual and terminating exploration, respectively, and 2) in the semi-synchronous and asynchronous models, three and four robots are necessary and sufficient to achieve perpetual and terminating exploration, respectively. These results clarify the power of lights for myopic robots since, without lights, five robots are necessary and sufficient to achieve terminating exploration in the fully synchronous model, and no terminating exploration algorithm exists in the semi-synchronous and asynchronous models. We also show that, in the fully synchronous model (resp., the semi-synchronous and asynchronous models), the proposed perpetual exploration algorithm is universal, that is, the algorithm solves perpetual exploration from any solvable initial configuration with two (resp., three) robots and two colors. On the other hand, we show that, in the fully synchronous model (resp., the semi-synchronous and asynchronous models), no universal algorithm exists for terminating exploration, that is, no algorithm may solve terminating exploration from any solvable initial configuration with three (resp., four) robots and two colors., Comment: 35 pages

We present an algorithm that ensures in finite time the gathering of two robots in the non-rigid ASYNC model. To circumvent established impossibility results, we assume robots are equipped with 2-colors lights and are able to measure distances between one another. Aside from its light, a robot has no memory of its past actions, and its protocol is deterministic. Since, in the same model, gathering is impossible when lights have a single color, our solution is optimal with respect to the number of used colors.

In this paper we prove lower and matching upper bounds for the number of servers required to implement a regular shared register that tolerates unsynchronized Mobile Byzantine failures. We consider the strongest model of Mobile Byzantine failures to date: agents are moved arbitrarily by an omniscient adversary from a server to another in order to deviate their computation in an unforeseen manner. When a server is infected by an Byzantine agent, it behaves arbitrarily until the adversary decides to move the agent to another server. Previous approaches considered asynchronous servers with synchronous mobile Byzantine agents (yielding impossibility results), and synchronous servers with synchronous mobile Byzantine agents (yielding optimal solutions for regular register implementation, even in the case where servers and agents periods are decoupled). We consider the remaining open case of synchronous servers with unsynchronized agents, that can move at their own pace, and change their pace during the execution of the protocol. Most of our findings relate to lower bounds, and characterizing the model parameters that make the problem solvable. It turns out that unsynchronized mobile Byzantine agent movements requires completely new proof arguments, that can be of independent interest when studying other problems in this model. Additionally, we propose a generic server-based algorithm that emulates a regular register in this model, that is tight with respect to the number of mobile Byzantine agents that can be tolerated. Our emulation spans two awareness models: servers with and without self-diagnose mechanisms. In the first case servers are aware that the mobile Byzantine agent has left and hence they can stop running the protocol until they recover a correct state while in the second case, servers are not aware of their faulty state and continue to run the protocol using an incorrect local state.

We present MCFR, a multicasting concurrent face routing algorithm that uses geometric routing to deliver a message from source to multiple targets. We describe the algorithm's operation, prove it correct, estimate its performance bounds and evaluate its performance using simulation. Our estimate shows that MCFR is the first geometric multicast routing algorithm whose message delivery latency is independent of network size and only proportional to the distance between the source and the targets. Our simulation indicates that MCFR has significantly better reliability than existing algorithms.

We study verification problems for autonomous swarms of mobile robots that self-organize and cooperate to solve global objectives. In particular, we focus in this paper on the model proposed by Suzuki and Yamashita of anonymous robots evolving in a discrete space with a finite number of locations (here, a ring). A large number of algorithms have been proposed working for rings whose size is not a priori fixed and can be hence considered as a parameter. Handmade correctness proofs of these algorithms have been shown to be error-prone, and recent attention had been given to the application of formal methods to automatically prove those. Our work is the first to study the verification problem of such algorithms in the parameter-ized case. We show that safety and reachability problems are undecidable for robots evolving asynchronously. On the positive side, we show that safety properties are decidable in the synchronous case, as well as in the asynchronous case for a particular class of algorithms. Several properties on the protocol can be decided as well. Decision procedures rely on an encoding in Presburger arithmetics formulae that can be verified by an SMT-solver. Feasibility of our approach is demonstrated by the encoding of several case studies.