A Low Complexity Aggregation Method for Underwater On-Pipe Sensor Network

Abstract

This paper considers the sensor aggregation for underwater pipe-assisted stress wave communication (SWC). Although the SWC is able to support the on-pipe sensor network in short range, the spectrum of long-range SWC is limited due to the effect of reflections at the pipe joints. To address this issue, the orthogonal frequency-division multiplexing (OFDM)-based consensus data aggregation is proposed. Furthermore, the consensus data can be received via time-domain sampling after ’compute-during-transmit’ in multiple access channels (MAC). The simulation results show 0.26%, 0.03%, and 0.08% mean-square-error (MSE) in 5, 10, and 20 sensor aggregation respectively with 5 dB average SNR. A vanishing mean-square-error (MSE) can be observed when increasing either the signal-to-noise ratio (SNR) or the number of sensors. This design can obtain the aggregated data by a simple time domain sampling with no need of down-conversion and demodulation.