Understanding the flooding in low-duty-cycle wireless sensor networks

In low-duty-cycle networks, sensors stay dormant most of time to save their energy and wake up based on their needs. Such a technique, while prolonging the network lifetime, sets excessive challenges for efficient flooding within the network. Tailored for obtaining short delay in low-duty-cycle networks, recently proposed flooding protocols have achieved some initial success. Many fundamental problems of flooding in low-dutycycle networks, however, are still not well understood. In this paper, we thoroughly investigate how the flooding behaviors are fundamentally affected from theory to practice in a lowduty-cycle sensor network. We study how practical factors like duty cycle length and link loss affect the flooding delay. We mathematically quantify the performance deterioration caused by those factors and present initial learning in achieving efficient flooding against them. Our theoretical analysis brings us not only an in-depth understanding of several fundamental trade-offs in low-duty-cycle sensor networks, but also insights on the design of flooding protocols that can approach excellent performance.