SAPM: Self-Adjusting Pipelining Mechanism for Efficient Bulk Data Dissemination in Smart Homes

Wireless Sensor Networks (WSNs) are popular in smart homes to implement context-aware and intelligent services. Due to fixing bugs and adding new functions, the software on sensor nodes needs to be updated depending on efficient bulk data dissemination. Existing page transfer of data dissemination protocols employ pipelining technique to achieve spatial multiplexing. The conventional pipelining inefficiency, however, becomes a bottleneck in terms of dissemination delay and communication overhead. In this paper, we propose SAPM, a novel Self-Adjusting Pipelining Mechanism for efficient bulk data dissemination. By considering the impact of link correlation, detecting accumulated pages in sensor nodes, and carefully selecting the number of pages to request, SAPM shortens the dissemination delay effectively and reduces communication overhead to improve current bulk data dissemination protocols' performance. Our evaluation results on test bed experiments and extensive simulations show that SAPM reduces the completion time of state-of-the-art bulk data dissemination protocols by 27.8% on average compared to naive solution and the number of transmissions by 17.4% on average under high link correlation while having a low memory overhead.