Benoit Darties

Assignment of Roles and Channels for a Multichannel MAC in Wireless Mesh Networks

A multichannel MAC improves throughput in wireless mesh networks by multiplexing transmissions over orthogonal channels. In this paper, we propose an efficient way for constructing the wireless mesh structure associated with Molecular MAC, a multichannel MAC layer designed for efficient packet forwarding. Molecular MAC outperforms other classical approaches, but requires a specific structure for efficient operation. First, we propose a centralized protocol that provides an upper bound for constructing such a molecular structure through a MILP (Mixed Integer Linear Programming) formulation that maximizes network capacity. Then, we present two distributed self-stabilizing heuristic protocols derived from the protocols for constructing respectively a Maximum Independent Set and a Spanning Tree. We compare the performance of the proposed protocols in terms of network capacity and route length.

Priority levels based multi-hop broadcasting method for vehicular ad hoc networks

This paper deals with broadcasting problem in vehicular ad hoc networks (VANETs). This communication mode is commonly used for sending safety messages and traffic information. However, designing an efficient broadcasting protocol is hard to achieve since it has to take into account some parameters related to the network environment, for example, the network density, in order to avoid causing radio interferences. In this paper, we propose a novel Autonomic Dissemination Method (ADM) which delivers messages in accordance with given priority and density levels. The proposed approach is based on two steps: an offline optimization process and an adaptation to the network characteristics. The first step uses a genetic algorithm to find solutions that fit the network context. The second one relies on the Autonomic Computing paradigm. ADM allows each vehicle to dynamically adapt its broadcasting strategy not only with respect to the network density, but also in accordance to the priority level of the message to send. The experimental results show that ADM effectively uses the radio resources even when there are globally many messages to send simultaneously. Moreover, ADM allows to increase the message delivery ratio and to reduce the latency and radio interferences.

Capacity and Energy-Consumption Optimization for the Cluster-Tree Topology in IEEE 802.15.4

IEEE 802.15.4 proposes to use a cluster-tree hierarchy to organize the transmissions in Wireless Sensor Networks. In this letter, we propose a framework to analyze formally the capacity and the energy consumption of this structure. We derive a Mixed Integer Linear Programming (MILP) formulation to obtain a topology compliant with the standard. This formulation provides the optimal solution for the network capacity: this constitutes an upper bound for any distributed algorithm permitting to construct a cluster-tree. This framework can also be used to evaluate the capacity and to compare quantitatively different cluster-tree algorithms.

Coupled-Tasks in Presence of Bipartite Compatibilities Graphs

We tackle the makespan minimization coupled-tasks problem in presence of incompatibility constraints. In particular, we focus on stretched coupled-tasks, i.e. coupled-tasks having the same sub-tasks execution time and idle time duration. We study several problems in the framework of classic complexity and approximation for which the compatibility graph is bipartite (star, chain, \(\ldots \)). In such context, we design efficient polynomial-time approximation algorithms according to different parameters of the scheduling problem.

Improved Complexity for Power Edge Set Problem.

We study the complexity of Power Edge Set (PES), a problem dedicated to the monitoring of an electric network. In such context we propose some new complexity results. We show that PES remains \(\mathcal {NP}\)-hard in planar graphs with degree at most five. This result is extended to bipartite planar graphs with degree at most six. We also show that PES is hard to approximate within a factor lower than Open image in new window in the bipartite case (resp. \(17/15-\epsilon \)), unless \(\mathcal {P}=\mathcal {NP}\), (resp. under \(\mathcal {UGC}\)). We also show that, assuming \(\mathcal {ETH}\), there is no \(2^{o(\sqrt{n})}\)-time algorithm and no \(2^{o(k)}n^{O(1)}\)-time parameterized algorithm, where n is the number of vertices and k the number of PMUs placed. These results improve the current best known bounds.

Completely independent spanning trees in some regular graphs

Let

An autonomic message dissemination protocol for Vehicular Ad hoc Networks: a density and priority levels aware approach

k\ge 2

A novel multi-hop broadcasting method for VANETs based on autonomic computing paradigm

be an integer and

Almost disjoint spanning trees: Relaxing the conditions for completely independent spanning trees

T_1,\ldots, T_k

Completely independent spanning trees for enhancing the robustness in ad-hoc Networks

be spanning trees of a graph