Bernard Tourancheau

Multiple Mobile Sinks Positioning in Wireless Sensor Networks for Buildings

Real deployment of wireless sensor networks inside buildings is a very challenging. In fact, in such networks, a large number of small sensor devices suffer from limited energy supply. These sensors have to observe and monitor their in-door environment, and then to report the data collected to a nearest information collector, referred to as the sink node. Sensor nodes which are far away from the sink relay their data via multiple hops to reach the sink. This way of communication makes the sensors near the sink deplete their energy much faster than distant nodes because they carry heavier traffic. So what is known as a hole appears around the sink and prevents distant nodes to send their data. Consequently the network lifetime ends prematurely. One efficient solution for this problem is to relocate sinks. In this work, we aim to find the best way to relocate sinks by determining their optimal locations and the duration of their sojourn time. So, we propose an Integer Linear Programming for multiple mobile sinks which directly maximizes the network lifetime instead of minimizing the energy consumption or maximizing the residual energy, which is what was done in previous solutions. Simulations results show that with our solution, the network lifetime is extended and the energy depletion is more balanced among the nodes. We also show that relocating mobile sinks inside a whole network is more efficient than relocating mobile sinks inside different clusters and we can achieve almost 52

Evaluating speedups on distributed memory architectures

%

Topology construction in RPL networks over beacon-enabled 802.15.4

network lifetime improvement in our experiments.

DTLS performance in duty-cycled networks

Abstract We discuss different ways of evaluating algorithmic speedup on a distributed memory machine. We use Gaussian elimination on a hypercube computer as a target example.

Energy cost of security in an energy-harvested IEEE 802.15.4 Wireless Sensor Network

In this paper, we propose a new scheme that allows coupling beacon-enabled IEEE 802.15.4 with the RPL routing protocol while keeping full compliance with both standards. We provide a means for RPL to pass the routing information to Layer 2 before the 802.15.4 topology is created by encapsulating RPL DIO messages in beacon frames. The scheme takes advantage of 802.15.4 command frames to solicit RPL DIO messages. The effect of the command frames is to reset the Trickle timer that governs sending DIO messages. We provide a detailed analysis of the overhead incurred by the proposed scheme to understand topology construction costs. We have evaluated the scheme using Contiki and the instruction-level Cooja simulator and compared our results against the most common scheme used for dissemination of the upper-layer information in beacon-enabled PANs. The results show energy savings during the topology construction phase and in the steady state.

Systolic Gauss-Jordan elimination for dense linear systems

The Datagram Transport Layer Security (DTLS) protocol is the IETF standard for securing the Internet of Things. The Constrained Application Protocol, ZigBee IP, and Lightweight Machine-to-Machine (LWM2M) mandate its use for securing application traffic. There has been much debate in both the standardization and research communities on the applicability of DTLS to constrained environments. The main concerns are the communication overhead and latency of the DTLS handshake, and the memory footprint of a DTLS implementation. This paper provides a thorough performance evaluation of DTLS in different duty-cycled networks through real-world experimentation, emulation and analysis. In particular, we measure the duration of the DTLS handshake when using three duty cycling link-layer protocols: preamble-sampling, the IEEE 802.15.4 beacon-enabled mode and the IEEE 802.15.4e Time Slotted Channel Hopping mode. The reported results demonstrate surprisingly poor performance of DTLS in radio duty-cycled networks. Because a DTLS client and a server exchange more than 10 signaling packets, the DTLS handshake takes between a handful of seconds and several tens of seconds, with similar results for different duty cycling protocols. Moreover, because of their limited memory, typical constrained nodes can only maintain 3-5 simultaneous DTLS sessions, which highlights the need for using DTLS parsimoniously.

Multiple redundancy constants with trickle

Wireless Sensor Networks (WSN) powered by energy harvesting from the environment represent a sustainable direction towards the future Internet of Things. In this paper, we evaluate the effect security features have on power consumption of such WSNs and revisit the previous conclusions regarding the energetic cost of security. We approach the problem of estimating the cost of security from a highly practical standpoint, and show that many real world, security agnostic factors affect the energy consumption of a device, but are often neglected by security researchers. Consequently, overpriced conclusions are often made that may affect decisions if security features should be by default enabled in real deployments. Our experimental results show that for practical applications and implementations security features introduce a negligible degradation that is often acceptable even for the most energy stringent systems, such as those based on energy harvesting.

Trickle-D: High Fairness and Low Transmission Load With Dynamic Redundancy

Abstract A systolic network for the Gauss-Jordan algorithms is presented for the solution of dense linear systems. This network compares favorably with Melhems network since its execution time (4 n − 1) is the same and the number of cells is decreased from n 2 2 to 3n 2 8 .

Improving trickle fairness: Locally-calculated redundancy constants

Wireless sensor network protocols very often use the Trickle algorithm to govern information dissemination. For example, the widely used IPv6 Routing Protocol for Low-Power and Lossy Networks (RPL) uses Trickle to emit control packets. We derive an analytical model of Trickle to take into account multiple redundancy constants and the common lack of synchronization among nodes. Moreover, we demonstrate message count unfairness when Trickle uses a unique global redundancy constant because nodes with less neighbors transmit more often. Consequently, we propose a heuristic algorithm that calculates a redundancy constant for each node as a function of its number of neighbors. Our calculated redundancy constants reduce unfairness among nodes by distributing more equally the number of transmitted messages in the network. Our analytical model is validated by emulations of constrained devices running the Contiki Operating System and its IPv6 networking stack. Furthermore, results very well corroborate the heuristic algorithm improvements.

OSCAR: Object security architecture for the Internet of Things

Embedded devices of the Internet of Things form the so-called low-power and lossy networks. In these networks, nodes are constrained in terms of energy, memory, and processing. Links are lossy and exhibit a transient behavior. From the point of view of energy expenditure, governing control overhead emission is crucial and is the role of the Trickle algorithm. We address Trickle’s fairness problem to evenly distribute the transmission load across the network, while keeping the total message count low. First, we analytically analyze two underlying causes of unfairness in Trickle networks: 1) desynchronization among nodes and 2) nonuniform topologies. Based on our analysis, we propose a first algorithm whose performance and parameters we study in an emulated environment. From this feedback, we design a second algorithm Trickle-D that adapts the redundancy parameter to achieve high fairness while keeping the transmission load low. We validate Trickle-D in real-life conditions using a large scale experimental testbed. Trickle-D requires minimal changes to Trickle, zero user input, emits 17.7% less messages than state-of-the-art and 37.2% less messages than state-of-practice, while guaranteeing high fairness across the network.