Section: New Results

QoS in Wireless Sensor Networks

Participants : François Despaux, Abdelkader Lahmadi, Evangelia Tsiontsiou, Kévin Roussel, Moutie Chehaider, 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. Our research on WSN QoS is thoroughly organized in four topics:

• self-adaptive MAC protocol for both QoS and energy efficiency

  By combining our two previous MAC protocols called Queue-MAC and CoSenS, we extended Queue-MAC to iQueue-MAC to support multi-hop transmission [23] , [6] . iQueue-MAC provides immediate yet energy-efficient throughput enhancement for dealing with burst or heavy traffic. Combined with CSMA/CA, iQueue-MAC makes use of queue length of each sensor node and allocates suitable TDMA slots to them for packets transmission. During light traffic period, no extra slots will be allocated; iQueue-MAC acts like other low duty-cycle MACs to conserve power. While in burst or heavy traffic period, iQueue-MAC senses the build up of packet queues and dynamically schedules adequate number of slots for packet transmission. Within ANR QUASIMODO project, we have implemented iQueue-MAC on STM32W108 chips that offer IEEE 802.15.4 standard communication. We set up several real-world experimental scenarios, including a 46 nodes multi-hop test-bed for simulating a general application, and conducted numerous experiments to evaluate iQueue-MAC, in comparison with other traffic adaptive duty-cycle protocols, such as multi- channel version RI-MAC and CoSenS. Results clearly show that iQueue-MAC outperforms multi-channel version of RI-MAC and CoSenS in terms of packet delay and throughput.

• 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 lower complexity and can deal with dynamic topology changes (which is the case in duty-cycled WSN). Through intensive numerical experiments, we have shown that OC has much less complexity compared with SAMCRA, known as one of the best existing algorithms. Sub-optimal paths can be obtained with a distributed version of OC, and following this principle, a first OC-based routing protocol is implemented over Contiki rime stack on TelosB motes. Its improvement and performance evaluation, as well as its integration to uIP/RPL stack is our ongoing work.

• Systems and middleware for supporting QoS in wireless sensor networks

  For supporting new protocols implementation which require to interact with low level services (MAC, Radio drivers, hardware timers) and integration to the Internet of Things approach, we focused on the OS for WSN. Several contributions have been made available for both ContikiOS (https://github.com/contiki-os/contiki/pull/519) and RiotOS (https://github.com/RIOT-OS/RIOT/pull/408, https://github.com/RIOT-OS/RIOT/pull/459). This allows to preparing for the next step towards the implementation of iQueue-MAC on both ContikiOS and RiotOS and compare experimentally with other protocols. In parallel and as part of LAR project, we also investigated the integration of different types of WSN using a gateway to make the data access transparent following RESTful webservice through CoAP/UPD/6loWPAN [24] .

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

  Probabilistic end-to-end performance guarantee may be required when dealing with real-time applications. As part of ANR QUASIMODO project, we are dealing with Markov modeling of multi-hop networks running slotted CSMA/CA (beacon enabled mode of IEEE 802.15.4). One of the problem of the existing models resides in their strong assumptions that may not be directly used to assess the end-to-end delay in practice. In particular, realistic radio channel, capture effect and OS-related implementation factors are not taken into account [15] , [14] . We proposed to explore a new approach which is based on process mining to extract the Markov chain model from the execution of the protocol code.