Congestion management in delay tolerant networks
Proceedings of the 4th Annual International Conference on Wireless Internet
| Hi-index | 0.00 |
This work focuses on the Delay/Fault-Tolerant Mobile Sensor Networks (DFT-MSN's) for pervasive information gathering, which play a key role in many promising applications, such as air quality monitoring, ubiquitous healthcare, and biological research. Due to the unique characteristics of DFT MSN, such as sensor mobility, loose connectivity, fault tolerability, delay tolerability, and buffer limit, designing an efficient data delivery scheme is challenging. We first study two basic approaches, namely, direct transmission and flooding, by using queuing theory and statistics. Based on the analytic results that show the tradeoff between data delivery delay/ratio and transmission overhead, we propose two simple and effective DFT-MSN data delivery schemes, namely Replication-Based Efficient Data Delivery Scheme (RED) and Message Fault Tolerance-Based Adaptive Data Delivery Scheme (FAD). The RED scheme utilizes the erasure coding technology in order to achieve the desired data delivery ratio with minimum overhead, while the FAD scheme employs the message fault tolerance, which indicates the importance of the messages, for the same purpose. Our results show that both schemes achieve high message delivery ratio with acceptable delay. We also observe that without end-to-end connections due to sparse network density and sensor node mobility, routing in DFT-MSN becomes localized and ties closely to medium access control, which naturally calls for merging Layer 3 and Layer 2 protocols in order to reduce overhead and improve network efficiency. To this end, we develop a cross-layer data delivery protocol for DFT-MSN, with various optimization approaches proposed to address the tradeoff between link utilization and energy efficiency. In order to better understand the characteristics of DFT-MSN, we also introduce a generic queuing analytic model based on queuing theory and statistic analysis. To validate our analytic models, we have carried out extensive simulations and observed a good match between analytic and simulation results. Beyond theoretical work, we also implement the FAD data delivery scheme and establish a DFT-MSN testbed by using Crossbow sensors. A small scale experiment is carried out to further evaluate the effectiveness and efficiency of DFT-MSN.