Digital twins have recently emerged as a trans-formative paradigm in wireless networking, offering promising avenues to enhance network efficiency and adaptability. By digitally replicating the physical network, they enable advanced monitoring, analysis, and optimization—key enablers for next-generation wireless systems. Although extensive efforts have been made in modeling and simulating wireless networks, their accuracy often falls short in practical environments. Real-World conditions, such as physical obstructions and complex radio propagation phenomena, are particularly difficult to replicate without resource-intensive local measurement campaigns. To address this challenge, we propose a lightweight data-driven approach to modeling wireless networks. Specifically, we utilize a simple yet informative Key Performance Indicator—the Packet Delivery Ratio (PDR)—to train link-specific quality prediction models. By training the models individually for each link, we effectively capture network heterogeneity and achieve high prediction accuracy. Moreover, we introduce a dynamic and adaptive approach for continuously selecting the most relevant model for the prediction. Experimental results from a real-world deployment show that this approach achieves a high prediction accuracy over both short and mid-term horizons, highlighting the effectiveness of individualized, adaptive modeling in dynamic wireless environments.

In industrial environments, cyber threats are escalating at an unprecedented rate, yet many existing security solutions fail to account for both contextual factors and the criticality of different network segments. This challenge is especially pronounced in diverse, large-scale, and highly dynamic Industrial Internet of Things (IIoT) environments. This paper presents a dynamic Zero Trust Access Control (ZTA) model that adapts to real-time device status, network conditions, and user behavior to enforce context-aware, security-driven access decisions. At its core, our framework combines mathematical threat assessment with fuzzy logic-based state management (FSM) to continuously adjust trust levels and access permissions. We validated our approach, through a proof-of-concept using a cluster of virtual machines (VMs) to simulate a controlled environment. This setup demonstrates the ZTA models effectiveness in small-scale networks and provides a foundation for testing various access scenarios and evaluating security policies.

Wireless industrial networks represent a key enabler for Industry 4.0. Since wireless links are known to be lossy, most deployments rely on scheduled transmissions to avoid collisions. Software Defined Networking (SDN) is a famous architecture where a controller centralizes all the configuration. In a scheduled wireless network, the SDN paradigm has to be extended to allocate both paths and bandwidth. Unfortunately, most solutions address only the initial configuration, which becomes suboptimal when the link quality evolves, or the conditions change. We propose here the mechanisms for continuous optimization. We define how control packets can reconfigure the data plane while guaranteeing a globally consistent state, even when the control plane exploits unreliable links. No data packet is dropped because of inconsistent forwarding rules. To be energy efficient, we also propose a scheduling algorithm that minimizes the network reconfiguration to minimize the overhead. Our simulation results highlight the strength of our proposition to handle topology and link quality changes.

The Industrial Internet of Things (IIoT) has now emerged for many industrial applications. However, many critical applications require high reliability and bounded end-to-end latency. Fortunately, scheduled wireless networks such as IEEE 802.15.4-TSCH try to orchestrate finely the transmissions. In particular, the Software-Defined Networking (SDN) paradigm is promising to concentrate the intelligence of the network in a controller and to simplify the functions achieved by the motes. SDN controller relies on link quality information of wireless devices to construct the routing topology of the network and to provision adequate radio resources for each link. We propose here an accurate link quality estimation scheme based on the SDN-TSCH, as an SDN solution for IEEE 802.15.4-TSCH network. We exploit the centralized view of the SDN controller to transmit packets of link quality estimation in a collision-free manner. Moreover, we investigate the performance of the control plane of SDN-TSCH through different shared, dedicated, and hybrid (mix of shared and dedicated cells) approaches. We figured out that the dedicated control plane provides high reliability and fast convergence for wireless SDN networks.

Industrial wireless sensor networks (IWSNs) play a key role in the Industry 4.0 revolution. The network infrastructure is critical to interconnect sensors and actuators and needs to respect key performance indicators. IEEE 802.15.4-TSCH is a candidate technology for IWSN since it relies on scheduled transmissions and frequency hopping to make the network more reliable. However, distributed scheduling solutions fail to provide high-reliability and low-end-to-end latency. Software defined network (SDN) tends now to emerge in wireless networks as well, where a controller is in charge of the whole network configuration. But scheduled wireless networks require to go beyond usual SDN forwarding rules by including the radio resource allocation (dedicated time-frequency blocks). Moreover, radio links are known to be unreliable, and we need to adapt the control and data planes to make the network efficient. We propose SDN-TSCH to orchestrate a scheduled network adapting the SDN paradigm. More specifically, the controller is in charge of 1) selecting the time source; 2) maintaining a tree structure for the control plane, with scheduled resources dedicated to the control plane; and 3) installing a new data flow while guaranteeing flow isolation. We also propose a very efficient link quality estimation technique tailored for scheduled TSCH networks. Our simulations highlight that the SDN controller can allocate in SDN-TSCH just-enough resources to respect both latency and reliability constraints.

The Industrial Internet of Things (IIoT) applications need to rely on a wireless infrastructure able to provide low end-to-end latency, and high reliability. Software-Defined Networking (SDN) is promising to make the network more agile, pushing the decision process to a controller. However, radio links are unstable while the controller needs to construct an accurate view of the network to schedule the transmissions efficiently. We propose here SDN-TSCH to separate the data and control planes for a scheduled network. We construct a reliable control plane, maintaining a collision-free path to and from the controller. Besides, SDN-TSCH guarantees flow isolation: each flow can reserve dedicated resources so that end-to-end reliability and latency constraints can be respected per flow. Finally, we also dedicate resources for best-effort traffic, to accommodate various applications. Our Cooja simulations highlight the flow isolation characteristics of SDN-TSCH: we provide very high reliability even in presence of best-effort traffic.

The software architecture of most applications is more and more fragmented, and relying on micro-services. Moreover, some parts may be specialized, and a customer may choose to delegate a task to a service provider. In this situation, the customer must be sure to get results that comply with the task when they pay the service provider, and inversely. We propose a framework based on atomic swaps to enable such simultaneous exchanges. Our scheme is based on exchanging a transactional key during an atomic swap. Our framework protects both actors, and enables non-repudiation, from both sides, even in an asynchronous environment.

Many IoT applications rely on data streams, flowing from producers to consumers. Typically, Named Data Networking has been designed to manipulate directly data chunks, and is particularly relevant in IoT networks. However, in multi-tenant networks, privacy is a major concern, and producers may refuse to share the personal data they generate with non trusted stakeholders. To guarantee k-anonymity, producers can require their data to be aggregated with the one of other producers. We propose here a routing scheme based on aggregation, relying on a pub-sub approach. By appropriately constructing and querying the set of offers, i.e. the list of data streams that are collected, aggregated and transformed together, our routing aggregation scheme provides a privacy aware large-scale interconnection, where the consumer does not access directly to individual measurements. Our performance evaluation highlights the flexibility of our solution to accommodate a large set of queries, while still respecting privacy.

Smart Cities rely on a broad variety of services generating an heterogeneous traffic load, and having diverse privacy and performance requirements. In this paper, we focus on mobility services for an Intelligent Transportation System (ITS), where several applications inter-operate by sharing information. In particular, we consider three main services: ride-hailing, smart parkings and traffic regulation. To investigate the load generated by these mobility services, we rely on two real datasets. Our goal is to study whether a given technology can scale with realistic conditions related to the envisioned smart mobility scenario. In particular, we analyze here how many LoRa gateways are necessary and how they should be deployed to support these mobility services. We highlight the bottlenecks occurring in most dense areas of the network and argue that heterogeneous deployments can efficiently handle such hot spots.

Many Internet of Things (IoT) applications have increasingly stringent requirements: messages have to be delivered to the destinations before a given deadline. Unfortunately, radio networks are known to be lossy, and retransmissions and acknowledgements help to improve the end-to-end reliability. To provide high-reliability, most wireless industrial networks schedule the transmissions to reduce the collisions, and to make the medium access deterministic. In these conditions, duocast helps to improve both the reliability and the fault-tolerance: two receivers are associated with one transmission, so that the transmission fails only if both receivers fail to decode the packet. While anycast has been widely used in simulations, we provide here a thorough experimental evaluation. Indeed, radio links present practically variations, may be asymmetrical, and the hidden receiver problem may practically reduce the gain of duocast. We demonstrate in our experimental evaluation that duocast is really efficient to provide high-reliability while limiting the number of (re)transmissions. It also helps the network to be fault-tolerant: even if a device crashes, or if a given link quality degrades, an alternative path exists to forward the packets, without any additional delay.

The Internet of Things (IoT) is expected to integrate a large number of sensors, and actuators to the Internet. Multiple concurrent applications may cohabit on top of the same IoT infrastructure, and may re-use the same data for various purpose. However, privacy represents a major concern for many IoT applications, such as in smart building and healthcare. We propose here a multi-domain IoT framework where each domain aggregates distinct data-streams to respect their privacy concerns. We argue that removing sensitive meta-data and aggregating values reported by each data-stream is sufficient to hide individual private measurements. Moreover, relying on the Named Data Networking (NDN) paradigm, we can exploit caching strategies and perform in-network processing to ensure both scalability and privacy. In this paper, we discuss the necessary mechanisms to design a scalable inter-domain, privacy aware NDN scheme.

The Internet of Things (IoT) interconnects a large collection of low power devices to the Internet. Instead of relying on the usual approach where IoT devices push their data to a cloud, we envision a decentralized approach, where multiple IoT domains cooperate to exchange data safely. As privacy is critical regarding IoT data, each domain should define what it accepts to export. We propose to create a multi-domain overlay of border routers, defining their own privacy constraints. Border routers export their available data streams and policies to their peering domains. Such policies define how streams should be aggregated and transformed to be compliant with the privacy requirements. We take advantage of the Named Data Networking (NDN) architecture to enable data re-usability and in-network transformations. Our evaluation highlights the scalability of our approach: NDN content stores supporting data aggregation and transformation reduce both the network and the cache load while enforcing privacy natively.

The Industrial Internet of Things (IIoT) is expected to be a key enabler for the Industry 4.0. However, networked control automation often requires high reliability and bounded latency to react properly. Thus, modern wireless protocols for industrial networks, such as IEEE 802.15.4-2015 time-slotted channel hopping (TSCH), rely on a strict schedule of the transmissions to avoid collisions and to make the end-to-end traffic deterministic. Unfortunately, guaranteeing a bounded end-to-end latency is particularly challenging since transmissions have to be temporally chained. Even worse, potential degradation of the link quality may result in reconstructing the whole TSCH schedule along the path. In this article, we propose the low-latency distributed scheduling function (LDSF) that relies on the organization of the slotframe in smaller parts, called blocks. Each transmitter selects the right set of blocks, depending on its hop distance from the border router, so that retransmission opportunities are automatically scheduled. To save energy, a node can still turn off its radio as soon as its packet is correctly acknowledged. Our mathematical analysis as well as our simulation evaluation show the efficiency of the proposed LDSF algorithm compared to three state-of-the-art scheduling functions (SFs): 1) the minimal SF (MSF); 2) low-latency SF (LLSF); and 3) stratum.

Industrial Internet of Things (IIoT) networks are considered the large-scale deployment of IoT devices for industrial applications such as smart manufacturing, harvesting and supply chain management. The Internet of Things (IoT) devices are typically connected over a wireless medium, given the large geographical distribution area and the increasing demand for flexible installations. In some cases, a combination of wired and wireless connectivity can be assumed as common practice. In both scenarios, wireless communications for IIoT networks is a fundamental component of the system architecture that needs to satisfy stringent requirements such as reliable connectivity and minimal delays. Therefore, the dependability of wireless communications for IIoT networks should be carefully studied to provide new solutions, which can guarantee that applications can meet their real-time and reliability requirements while optimizing the control capability of the overall network. This paper focuses on the dependable wireless communications in the IIoT networks, where wireless control and monitoring tasks need to meet stringent real-time and reliability constraints. After reviewing recent solutions and discussing their suitability for IIoT networks, we highlight the yet open challenges that needs to be tackled by both academia and industry.

Industrial applications require more and more low-power operation and high-reliability (close to 100%). Since traditional low-power radio technologies are sensitive to external interference, many recent standards implement frequency hopping schemes. For instance, IEEE 802.15.4-2015 time slotted channel hopping (TSCH) relies on a deterministic schedule of data transmissions combined with a pseudo-random frequency hopping scheme to improve the reliability. Unfortunately, specific radio channels keep on increasing the average number of retransmissions. Using a subset of the best radio channels (whitelisting) helps to improve the reliability, but may create collisions when used improperly. We here investigate the most accurate techniques to use only the best radio channels while still providing deterministic performance. We propose to group the links per timeslot, allocating them either to the same whitelist or even appropriately reordering them to avoid collisions. Finally, we evaluate the performance of the different whitelisting schemes using an experimental dataset from FIT IoT-LAB platform, proving the relevance of such approach to improve the reliability.

Industrial wireless networks are now used in many applications, and require to fulfill a certain set of requirements to operate properly. IEEE802.15.4-TSCH is considered a suitable solution to provide real time multihop transmissions in noisy and harsh environments. The standard relies on a strict schedule of the transmissions to reduce the radio duty cycle ratio. While constructing a schedule for periodic traffic has been widely studied in the past, we focus here on the aperiodic, sporadic case. We have to multiplex the transmissions in the schedule to reduce the energy consumption while limiting the number of collisions to provide still high reliability. We propose here to study experimentally the performance of TSCH with shared access and bursty arrivals. Then, we demonstrate how to re-adapt the scheduler to better deal with unpredictable traffic. By performing experiments, we can predict the optimal number of transmitters in a shared cell. Thus, we can overcome collisions and packets drops in complex scenarios where bursty traffic is required.

Many applications for the Internet of Things require strict guarantees to operate properly. IEEE802.15.4-TSCH has been proposed to design a real-time access for industrial wireless networks. The standard relies on a strict schedule of the transmissions., and combines frequency hopping to provide high-reliability. While most of the propositions focus on a periodic traffic, we propose here to address the non-periodic case. In particular, shared access would be more efficient to reduce the power duty cycle by multiplexing the transmissions through the same cells. We propose here a Markov model for a shared access and sporadic traffic. By formulating the collision probability, the scheduler can decide how many transmitters to assign to a given shared cell according to the amount of traffic each node generates on average. The simulations validate the accuracy of our model to capture the collision rate for non-periodic traffic.

Industrial networks differ from others kinds of networks because they require real-time performance in order to meet strict requirements. With the rise of low-power wireless standards, the industrial applications have started to use wireless communications in order to reduce deployment and management costs. IEEE802.15.4-TSCH represents currently a promising standard relying on a strict schedule of the transmissions to provide strong guarantees. However, the radio environment still exhibits time-variable characteristics. Thus, the network has to provision sufficient resource (bandwidth) to cope with the worst case while still achieving high energy efficiency. The 6TiSCH IETF working group defines a stack to tune dynamically the TSCH schedule. In this paper, we analyze in depth the stability and the convergence of a 6TiSCH network in an indoor testbed. We identify the main causes of instabilities, and we propose solutions to address each of them. We show that our solutions improve significantly the stability.

Smart Cities need real time information to improve the efficiency of their transportation systems. In particular, crowd sensing may help to identify the current speed in each street, the congested areas, etc. In this context, map matching techniques are required to map a sequence of GPS waypoints into a set of streets on a common map. Unfortunately, most map matching approaches are probabilistic. We propose rather an unambiguous algorithm, able to identify all the possible paths that match a given sequence of waypoints. We need an unambiguous identification for each waypoints set. For instance, the actual speed should be assigned to the correct set of streets, without error. To identify all the possible streets, we construct the set of candidates iteratively. We identify all the edge candidates around each waypoint, and reconstruct all the possible sub-routes that connect them. We then verify a set of constraints, to eliminate impossible routes. The road segments common to all computed routes form an unambiguous match. We evaluate the matching ratio of our technique on real city maps (London, Paris and Luxembourg). We also validate our approach with a real GPS trace in Seattle.

With the large growth of the Internet of Things (IoT), a strong focus has been put on designing and developing energy efficient and high performance protocols. Industrial-type wireless networks require strict and on-time delivery guarantees, such as close to 100% network reliability and ultra low delay. To this aim, standards such as IEEE 802.15.4-TSCH or Wireless HART, aim to guarantee high-level network reliability by keeping nodes time-synchronized and by employing a slow channel hopping pattern to combat noisy environments and external interference. In wireless networks, since all the radio channels are not impacted in a similar manner, blacklisting bad channels may improve performance of the whole wireless infrastructure. In this paper, we perform a thorough experimental study to characterize the radio (for all IEEE 802.15.4 channels) and connectivity among the nodes of an indoor testbed. More precisely, we investigate the locality of these blacklisting techniques and we highlighted: the fact that some channels perform poorly only in a small set of locations, for certain radio links. Our study tends to justify the need for local blacklisting techniques, demanding more control packets, but dealing more efficiently with spectral re-use.

Industrial applications require more and more low-power operations, low-delay, deterministic communications as well as end-to-end reliability close to 100%. However, traditional radio technologies are sensitive to external interference, which degrades the reliability and introduces unpredictable delays due to collision detection and retransmissions. Therefore, recent standardization efforts focus on slow channel hopping strategies to provide strict Quality of Service (QoS) for the Industrial Internet of Things (IIoT). By keeping nodes time-synchronized and by employing a channel hopping approach, IEEE 802.15.4-TSCH (Time-Slotted Channel Hoping) aims at providing high-level network reliability. However, some radio channels still suffer from high external interference and need to be blacklisted. Since the interference pattern is rather dynamic, unpredictable and highly localized, we here propose heuristics to decide which channels to blacklist. To avoid deafness, the transmitter and the receiver must also agree on a consistent blacklist. Furthermore, since the external interference may be time-dependent as well, we also propose mechanisms to decide when a channel has to be blacklisted or on the contrary recovered. Our thorough experimental evaluation based on OpenWSN and FIT IoT-LAB highlight the relevance of this approach: with a localized blacklisting strategy, we increase by 20% packet delivery rate for the worst links.

The IoT expects to exploit IEEE802.15.4e-TSCH, designed for wireless industrial sensor networks. This standard relies on techniques such as channel hopping and bandwidth reservation to ensure both energy savings and reliable transmissions. The 6TiSCH working group currently proposes to exploit the RPL routing protocol on top of the IEEE802.15.4-2012-TSCH layer. Since many applications may require low end-to-end delay (e.g. alarms), we propose here a distributed algorithm to schedule the transmissions while upper bounding the end-to-end delay. Our strategy is based on stratums to reserve time-bands for each depth in the routing structure constructed by RPL. By allocating a sufficient number of timeslots for the possible retransmissions, we guarantee that any packet is delivered during one single slotframe, wherever the source is located. Experiments on a large scale testbed prove the relevance of this approach to reduce the end-to-end delay while minimizing the number of collisions, prejudicial to the reliability in multihop networks.

With the large deployment of smart and heterogeneous devices, interest of researchers to define new protocols to meet Internet of Things (IoT) requirements is growing. A particular interest was accorded to define a robust MAC layer for wireless sensor networks, in order to reduce interference caused by other co-located networks and applications using the ISM band. This paper gives a comprehensive study of the Time Slotted channel Hopping IEEE802.15.4, part of the 6TiSCH stack and explains how its TDMA approach improves the reliability with performance guarantees. We also investigate analytically and experimentally the impact of the scheduling algorithm on the reliability. Then, we provide an experimental evaluation of co-located WSN using the FiT-IoT LAB testbed and the OpenWSN Stack. Performance analysis of IEEE802.15.4e-TSCH is achieved with a variable number of co-located synchronized or unsynchronized instances. While this standard is robust for lightly loaded networks, new mechanisms have to be proposed when we have too much traffic or too many interfering networks.

Time Slotted Channel Hopping (TSCH) is among the proposed Medium Access Control (MAC) layer protocols of the IEEE 802.15.4-2015 standard for low-power wireless communications in Internet of Things (IoT). TSCH aims to guarantee high network reliability by exploiting channel hopping and keeping the nodes time-synchronized at the MAC layer. In this paper, we focus on the traffic isolation issue, where several clients and applications may cohabit under the same wireless infrastructure without impacting each other. To this end, we present an autonomous version of 6TiSCH where each device uses only local information to select their timeslots. Moreover, we exploit 6TiSCH tracks to guarantee flow isolation, defining the concept of shared (best-effort) and dedicated (isolated) tracks. Our thorough experimental performance evaluation campaign, conducted over the open and large scale FIT IoT-LAB testbed (by employing the OpenWSN), highlight the interest of this solution to provide reliability and low delay while not relying on any centralized component.

The industrial Internet of Things (IoT) relies on multi-hop radio paths. Synchronized nodes follow a Frequency-Time Division Multiple Access (FTDMA) schedule, but even using channel-hopping to mitigate interference, the radio links suffer packet losses. Resource allocation algorithms must consider the requirements of the applications in terms of delivery and allocate extra resource to compensate for anticipated losses. We propose a hop-by-hop allocation mechanism that extends the Traffic-Aware Scheduling Algorithm (TASA) by enabling retransmissions. We give each flow on the network the possibility to satisfy its applicative end-to-end delivery constraint. We keep the amount of resource allocated for retransmissions low, and balance the allocations on the relay nodes. By means of simulations, we show the gain in terms of reliability, and the cost in terms of number of allocations.

The devices composing Wireless Sensor Networks (WSN) are very limited in terms of memory, processing power and battery. RPL has emerged as the de facto routing standard in low-power and lossy networks. While most of the proposals focus on minimizing the global energy consumption, we aim here at designing an energy-balancing routing protocol: each node should efficiently consume the same quantity of energy to improve the network lifetime. To this end, we exploit an Expected Lifetime metric, denoting the residual time of the nodes (time until the node will run out of energy). We propose mechanisms to detect the energy-bottleneck nodes and to spread the traffic load uniformly among them. While RPL constructs a Destination-Oriented Directed Acyclic Graph (DODAG) structure, it only implements single path. We propose here to exploit its natural multipath structure. This multipath approach helps reducing the number of DODAG reconstructions that leads to instabilities and convergence problems. Simulations highlight we improve both the routing reliability and the network lifetime.

Fingerprinting based localization is one of the most widely used indoor localization methods. This method is divided into two phases: during the off-line training phase, fingerprints within the area of interest are collected and stored in a fingerprint database; during the on-line mapping phase, the real-time location of a device is estimated by mapping itself to the most accurate fingerprint in the database. The efficiency of the mapping process is one of the key challenges of the on-line phase, and is mostly characterized by the localization accuracy and the response time. Clustering methods have been introduced to reduce computational overhead. In this paper, we propose a dynamic clustering method leveraging the inertial information of the target device. An active area is dynamically computed around the prior position. The target mapping space is significantly reduced with this active area. This method can be integrated with other clustering algorithms to overcome the edge problem and remove outliers. We evaluate this method compared with the state-of-the-art methods on a body sensor based localization system. The results show that the accuracy, precision and response time of the system are improved greatly.

The era of the Internet of Things brings complexity and deployment costs in smart cities, particularly in Wireless Sensor Networks (WSNs). Utilities such as gas or water providers are keen on delegating the management of the communications to specialized firms, namely WSN Operators, that will share the WSN resource among their various clients. WSN operators will use a functional architecture to manage the Service Level Agreements (SLAs), i.e. the Quality of Service (QoS) clauses they contract with their clients. WSN operators will need a robust and reliable technology in order to guarantee QoS constraints in a wireless environment, as in the industrial world. IEEE 802.15.4e Time Slotted Channel Hopping (TSCH) [1] is one good candidate. Moreover, the IETF experience in IP networks management is an important input for monitoring and QoS control over WSNs. This article gives formal guidelines for the implementation of a SLA architecture for operated WSNs. It distinguishes the various formal algorithms that are necessary to operate a WSN according to SLAs, and determines which functional entities are necessarily technology-dependent. Detailed examples of such entities are developed in an IPv6 over IEEE 802.15.4e TSCH context, such as advocated in the IETF 6TiSCH Working Group [2].

IEEE 802.15.4 defines a popular MAC standard for wireless sensor and actuator networks. With the default parameters, under medium to high load, 802.15.4 generates excessive collisions and packet losses. Low duty cycles even exacerbate the problem, because more nodes become active after long periods of sleep and contend for channel access. In this paper, we have applied the models that led to the Idle Sense access method for 802.11 to the 802.15.4 slotted CSMA/CA, taking into account the central role of the coordinator and also the bursty nature of the traffic. Surprisingly, the approach perfectly applies to 802.15.4 even if the principles of the two access methods fundamentally differ. Based on the model, we propose ABE, an adaptation method that adjusts the contention window to optimal values so that the network obtains high throughput along with low duty cycles leading to low energy consumption. The method converges to near-optimal back off values even under bursty traffic and for any number of contending nodes.

The IPv6 Routing Protocol for Low-Power and Lossy Networks (RPL) is one of the emerging routing standards for multihop Wireless Sensor Networks (WSN). RPL is based on the construction of a Destination-Oriented Directed Acyclic Graph (DODAG), which offers a loop-free topology to route data packets. While several routing metrics have been proposed in the literature, it is unclear how they perform with RPL. In this paper, we analyze the impact of different PHY and MAC metrics on the stability and efficiency of RPL. We highlight the fact that realistic conditions lead to instabilities and oscillations in the routing structure. While minimizing the hop length leads to a stable but poor routing structure, more sophisticated link metrics such as ETX reflect more clearly the radio link quality but increase the number of DODAG reconfigurations. We also provided a detailed methodology to measure the DODAG stability and to implement efficiently each routing metric with RPL.

The dynamic throughput optimal policy that supports all incoming traffic while minimizing outage probability for cooperative ad-hoc network is consider in this paper. The algorithm operates without knowledge of traffic rates or the memory historical channel states. Using the technique of Lyapunov optimization, we present a framework that can tradeoff between the capacity region the total outage probability of network. Distance this minimum outage probability is controlled by a parameter V effecting an explicit tradeoff in average queue length. We also use simulation to compare the performance of our policy with two-hop relay policy as well as non-cooperative policy.

We propose MCCT (Multi-Channel Cluster Tree), a cluster-tree construction protocol for nodes in IEEE 802.15.4 beacon-enabled mode. By multiplexing transmissions across orthogonal channels, we reduce collisions between control and data frames, which leads to better packet delivery rate and fairness. We propose a method for constructing a cluster-tree suitable for minimizing beacon collisions. The protocol builds on a neighbor discovery procedure that uses a dedicated control channel while still sticking to the superframe structure of IEEE 802.15.4. We also specify a channel assignment and superframe scheduling method that takes into account channel diversity. We evaluate the proposed protocol through simulation and compare with other proposals: standard 802.15.4 and a representative of distributed solutions to the superframe scheduling problem — MeshMAC. The simulation results show that MCCT significantly improves packet delivery ratio, delay, and fairness. It also results in very good packet delivery ratio for increased network density.

We can improve the performance of wireless mesh networks by using multiple interfaces tuned to non-overlapping channels. A Channel and Interface Assignment (CIA) decides when to switch interfaces and which channel to use. Surprisingly, the impact of CIAs on connectivity has received little attention so far. In this paper, we present a comparison and performance evaluation of the existing CIA strategies addressing the connectivity issues: network topology, density of connections, and neighbor discovery. The results presented in this paper provide guidelines for network designers in planning multi-channel multi-interface network deployments.

We consider the problem of running RPL on top of the IEEE 802.15.4 MAC layer-the two layers operate over two different structures, a directed acyclic graph in the case of RPL and a cluster-tree for IEEE 802.15.4. We propose to adapt the cluster-tree of IEEE 802.15.4 so that it can efficiently work coupled with RPL. Nodes in our modified cluster-tree can associate with several parent nodes by taking advantage of an adequate organization of superframes at the MAC layer. Building on this modified MAC layer, we define an opportunistic forwarding scheme that extends RPL with the possibility of forwarding packets over multiple paths. Instead of always using a preferred parent, a node opportunistically forwards packets through other parents as long as their routes towards the sink are better. We take advantage of the opportunistic forwarding to support higher-priority delay-sensitive alarms that need to arrive in sink before a given deadline along with low-intensity monitoring data considered as best-effort. We compare our opportunistic version of RPL to its basic version through detailed simulations in terms of packet delivery ratio, incurred delay, and overhead.

We propose a new energy efficient network structure for maintaining synchronization in access methods based on Synchronous Preamble Sampling. Our scheme limits the number of synchronization messages and increases network capacity through the use of multiple non-interfering virtual channels. It consists in constructing independent clusters based on the Weakly Connected Dominating Set (WCDS) so that they can use different virtual channels and only need to maintain internal synchronization, while still offering global connectivity. We define a distributed and self-stabilizing algorithm for constructing and maintaining the clusters. Our simulation results show that the proposed scheme has comparable energy consumption to Scheduled Channel Polling, but results in better network capacity. Moreover, it achieves better energy savings and network capacity than recently proposed Crankshaft access method.

In this paper, we briefly describe an interactive roulette game enabled over a wireless sensor network platform. It basically consists in a train speeding in one way (the spinning roulette) and in a message hopping along some deployed sensors in the other way (the wheeling ball). This demonstration aims at illustrating in an interactive manner the high radio channel randomness in dense wireless sensor networks, as well as its impact on the communication protocols.

One way of improving performance of wireless mesh networks is to use multiple non-overlapping channels. At the same time, the mesh network must continuously self-adapt to varying radio conditions and topology changes. Thus, we propose a neighbor discovery protocol that fits the requirements of multichannel networks while dealing at the same time with the deafness problem. The proposed scheme also takes advantage of neighbor discovery to continually monitor channel activity. We provide a theoretical analysis of the average discovery time. Our simulations show that the solution integrated within Molecular MAC, a multichannel MAC, results in efficient discovery and channel assignment.

Multichannel solutions are increasingly used to cope with the problem of low capacity in sensor networks resulting from high contention during wake-up periods of nodes. We can consider wake-up schedules as virtual channels, because nodes using different schedules cannot communicate with each other. In this paper, we show that multiple virtual channel solutions come at the cost of an increased energy consumption in broadcasts. We analyze the problem of broadcasting over multiple virtual channels under different classes of MAC methods and propose Clustered Virtual Channels (CVC), a new network structure that limits the number of frames needed for maintaining multiple virtual channels synchronized. Our simulation results show that CVC reduces the cost of maintaining synchronization and increases capacity while making all types of broadcasts possible.

We present a novel geographical routing scheme for spontaneous wireless mesh networks. Greedy geographical routing has many advantages, but suffers from packet losses occurring at the border of voids. In this paper, we propose a flexible greedy routing scheme that can be adapted to any variant of geographical routing and works for any connectivity graph, not necessarily Unit Disk Graphs. The idea is to reactively detect voids, backtrack packets, and propagate information on blocked sectors to reduce packet loss. We also propose an extrapolating algorithm to reduce the latency of void discovery and to limit route stretch. Performance evaluation via simulation shows that our modified greedy routing avoids most of packet losses.

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.

A multi-channel MAC is a promising approach for improving network throughput by multiplexing transmissions over orthogonal channels. Molecular MAC has recently adopted this approach by proposing to modify the standard IEEE 802.11 DCF. It requires role and channel assignment to nodes: some of them use a static channel while others dynamically switch to neighbor channels on-demand. To assign roles and channels, we extend the notion of the Weakly Connected Dominating Set, a structure already used in clustering. We adapt it by introducing new constraints to define what we call a reversible WCDS (r-WCDS), which is particularly suitable for Molecular MAC. We propose a divide-and-conquer scheme that partitions the network into clusters with one leader per cluster solving a MILP formulation to assign roles in its cluster. By appropriately defining the roles at the border of clusters, we maintain global connectivity in the r-WCDS. Our simulations show that the performance of our strategy is close to a centralized algorithm.

We propose a novel view on efficient packet forwarding in wireless mesh networks based on a molecular analogy in which mesh routers are either nuclei or electrons in an atom that corresponds to a 802.11 cell. In this view, a mesh network appears as a collection of spatially distributed 802.11 cells operating on different channels. We define Molecular MAC that uses dynamic channel switching at neighbor mesh routers to efficiently forward packets over multiple hops. To avoid deafness, nuclei notify electrons about pending packets before electrons pull them from nuclei for reception or further forwarding. We evaluate the proposed scheme through simulation and compare with other proposals. Our results show that Molecular MAC obtains much better performance in terms of throughput, packet delivery rate, end-to-end delay, and fairness.

We propose an efficient way of constructing the wireless mesh structure associated with Molecular MAC, a multichannel access method designed for efficient packet forwarding. We base our role assignment on a restricted Weakly Connected Dominating Set structure. After presenting a formal definition of the role assignment problem, we prove its NP-completeness. Then, we propose a centralized 2-approximation algorithm that maximizes the sum of radio link capacities in the molecular structure. Finally, we extend this protocol so that it can operate in a distributed way still providing the same guarantee. This distributed protocol is self-stabilizing thus robust to topology changes. Our simulation results show that the 2-approximation distributed protocol provides an improvement in throughput with respect to other protocols.

International audience

We analyze greedy geographical routing in spontaneous wireless mesh networks to show several interesting properties. First, we can approximate the dependence of packet loss probability on the mean node rank with a Fermi-Dirac function. When the mesh network grows, it becomes opaque to packets regardless of the average node rank. We also show that packet loss probability in mesh networks with greedy geographical routing does not exhibit the behavior of percolating systems. Finally, we propose an analytical model of greedy geographical routing and use it to derive packet loss probability.

MANET protocols are often evaluated through simulations, not in a real radio environment. We describe here the implementation and deployment of a complete testbed for hybrid networks, allowing a seamless integration of an ad-hoc network in the Internet. In particular, a self-organization protocol and the mobility management protocol benefiting from this organization were implemented. The performances demonstrate the feasibility and usefulness of this scheme. Besides, this testbed offers a detailed description of the requirements to constitute a wireless testbed and to test any protocol for ad hoc or hybrid networks.

Assuming a given network topology and a routing protocol, this work is focused on the capacity evaluation of routing protocols based on either a self-organization scheme or a flat approach. To reach this goal, we propose to use linear-programming formulation to model radio resource sharing as linear constraints. Four models are detailed to evaluate the capacity of any routing scheme in wireless multihops networks. First, two models of fairness are proposed: either each node has a fair access to the channel, or the fairness is among the radio links. Besides, a pessimistic and an optimistic scenarios of spatial re-utilization of the medium are proposed, yielding a lower bound and an upper bound on the network capacity for each fairness case. Finally, using this model, we provide a comparative analysis of some flat and self-organized routing protocols

Routing protocols are the main issue of ad hoc networks. Because flat propositions (reactive, proactive) are not sufficient and suffer from a lack of performance, new solutions should he investigated and proposed. On the other hand, virtual topologies propose to structure the network and to give a hierarchy between the strongest and the weakest nodes. We propose a new routing protocol, virtual structure routing (VSR), based on a virtual topology including both a backbone and clusters. The backbone is used to collect control traffic and to reduce overhead for route discovery. VSR uses clusters to define a route as a list of cluster IDs. This cluster topology is more stable than the physical topology. Hence, routes are more robust. VSR combines the assets of both flat approaches; intra-cluster routing is proactive while inter-cluster routing is reactive. Finally, routes are computed dynamically and a mechanism for route repair is proposed.

Hybrid networks are heterogeneous networks merging both wireless and ad hoc nodes and where the interconnection to the IP world is an important topic through gateways called AP (access point). Indeed, each node can be contacted and can contact another node in the Internet. To reach that, architectures that support mobility management are studied. The solutions inspired by wired networks are not particularly suited to hybrid networks. We propose to use a virtual dynamic infrastructure including both backbone and clusters. A backbone is suited to spare energy, optimize control traffic diffusion and hierarchize participants. The clusters are intended to create service areas and to handle particularly the mobility management. We present algorithms to both construct and maintain such structure. This dynamic topology is robust according to mobility, and is well suited to implement mobility management and localization procedure. Finally, the number of backbone members and clusters are completely parameterizable according to the environment.