Section: New Results

QoS in Wireless Sensor Networks

Participants : François Despaux, Abdelkader Lahmadi, Bilel Nefzi, Hugo Cruz-Sanchez, Ye-Qiong Song [contact] .

WSN research focus has progressively been moved from the energy issue to the QoS issue. Typical example is the MAC protocol design, which cares about not only low duty-cycle, but also high throughput with self-adaptation to dynamic traffic changes [21] . Our research on WSN QoS is thoroughly organized in three topics:

  • MAC protocol design for both QoS and energy efficiency

    The main result that we obtained in 2012 is a new hybrid CSMA/TDMA MAC protocol, called Queue-MAC, that dynamically adapts the duty-cycle according to the current network traffic. The queue length of nodes is used as the network traffic indicator. When the traffic increases, the active CSMA period is accordingly extended by adding dynamic TDMA slots, allowing thus to efficiently handle burst traffic under QoS constraints. This protocol is implemented on the STM32W108 SOC chips and compared with both a fixed duty-cycle reference protocol and an optimized IEEE802.15.4 MAC protocol. Through extensive experimental measurements, we showed that our queue-length aware hybrid CSMA/TDMA MAC protocol largely outperforms the compared protocols. The proposed protocol can be easily implemented through slight adaptation of the IEEE802.15.4 standard [25] .

    Many industrial WSN are based on IEEE802.15.4 standard. One of the critical issues is the scheduling of neighboring coordinators beacons. In [20] , we presented TBoPS, a novel technique for scheduling beacons in the cluster-tree topology. TBoPS uses a dedicated period called beacon only period (BOP) to schedule beacons at the beginning of IEEE 802.15.4 superframe. The advantage of TBoPS is that every beacon-enabled node distributively selects a beacon schedule during association phase.

  • QoS routing

    For supporting different QoS requirements, routing in WSN must simultaneously consider several criteria (e.g., minimizing energy consumption, hop counts or delay, packet loss probability, etc.). When multiple routing metrics are considered, the problem becomes a multi-constrained optimal path problem (MCOP), which is known as NP-complete. In practice, the complexity of the existing routing algorithms is too high to be implemented on the low cost and power constrained sensor nodes. Recently, Operator calculus (OC) has been developed by Schott and Staples with whom we collaborate. OC can be applied to solving MCOP problem with much lower complexity and can deal with dynamic topology changes (which is the case in duty-cycled WSN). The OC approach has been successfully applied to a concrete routing problem [13] . Its implementation over Contiki on TelosB motes has also been achieved, confirming thus its great potential for developing new QoS routing protocols for WSN.

  • End-to-end performance in multi-hop networks

    Probabilistic end-to-end performance guarantee may be required when dealing with real-time applications. For instance, in our ANR QUASIMODO project, we considered an intrusion detection and tracking scenario and analyzed the application requirements with respect to the network QoS. Assuming the use of the extended Kalman filter based tracking technique, we derived the tradeoff relationship between the tracking precision and the delay (from the target position and speed sampling to mobile nodes moving to cover the estimated next step area). In [5] we proposed a novel coordinative moving algorithm for autonomous mobile sensor networks to guarantee that the target can be detected in each observed step while minimizing the amount of moving sensors (so saving energy). In such kind of application context, we aim to provide methods for both network resource allocation and estimating the end-to-end delay in multi-hop WSN. Assuming IEEE802.15.4 WSN with cluster-tree routing, in [16] we addressed the problem of allocating and reconfiguring the available bandwidth using an Admission Control Manager that guarantees that the nodes respect their probabilistic bandwidth assignment when generating data traffic. It has been shown by simulation that using the proposed method, one can obtain desired probabilistic guarantee in both bandwidth and energy efficiency.

    In a more general context of meshed networks, we present an empirical support of an analytical approach, which employs a frequency domain analysis for estimating end-to-end delay in multi-hop networks. The proposed analytical results of the end-to-end delay distribution are validated through simulation and compared with queuing theory based analysis. Our results demonstrate that an analytical prediction schema is insufficient to provide an adequate estimation of the end-to-end delay distribution function, but it requires to be combined with simulation methods for detailed link and node latency distribution [15] .