Cross-layer multipath approach for critical traffic in duty-cycled wireless sensor networks

2021 
Abstract Ensuring that an active path is provided at all times in a Wireless Sensor Network (WSN) is fundamental for securing the timely delivery of critical data and maintaining the quality of service required to effectively support delay-sensitive applications. In duty cycling mode, ensuring such a path becomes a real challenge as the activity scheduling of different nodes should be regulated and adjusted to provide a pipe-lined node activity from source to sink. Furthermore, some transmission delays may be incurred in waiting for a next-hop node to be awakened. The data transfer time is mainly influenced by the performance of both MAC and routing protocols; an inter-layer design allowing efficient coordination between the MAC and routing is then necessary to make the appropriate decisions relating to the selection of the next hop and the adjustment of the duty cycle. In this article, we provide a cross-layer multi-path routing approach designed to support the transfer of critical data over duty-cycled WSN networks. This approach establishes multi-node disjoint paths with complementary duty-cycling schedules to ensure the availability of an active path capable of transmitting critical data at all times. It relies on a tight cooperation between the routing and MAC layers in order to appropriately adjust and regulate the wake-up schedules of involved nodes in the forwarding phase. We analytically proved the correctness of the proposed scheduling adjustment rules. We conducted extensive simulations to show the ability of the CL-NDRECT to provide a viable trade-off between energy conservation and real time responsiveness. Simulation results show that CL-NDRECT provides around 65% cumulative energy saving during route establishment and data transfer compared to the always-on multipath approach at the cost of around 10% increase in response time.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    53
    References
    1
    Citations
    NaN
    KQI
    []