A new energy saving framework for long lasting wireless sensor nodes

TLDR: A new data processing framework for energy saving, based on a synergy between data compression and data aggregation techniques, which allows a more compact representation of the transmitted data in the clusters head nodes compared with the individual use of each technique.

Abstract: This paper proposes a new data processing framework for energy saving, based on a synergy between data compression and data aggregation techniques. This combination allows a more compact representation of the transmitted data in the clusters head nodes compared with the individual use of each technique. For data compression, we use the static Huffman algorithm with Extrapolation prediction that exploits Temporal correlation (ET) and static Huffman algorithm with Differential prediction that exploits Spatial correlation (DS). For data aggregation we use a Bit Aggregation Technique (BAT) to efficiently represent the data carrying bits from a byte. To validate the synergetic combination between ET, DS and BAT, we developed two platforms that allow us to simulate and practical implement the algorithms. The performances are compared with those of the classical Adaptive Huffman algorithm with Differential prediction that exploits Temporal correlation (DT). The results show an important decrease of energy consumption for the synergetic solution obtained both on software and hardware platforms.