Design and Evaluation of an Acoustic Modem for a Small Autonomous Unmanned Vehicle

Siyuan Zheng, Feng Tong, Bin Li, Qiuyang Tao, Aijun Song, Fumin Zhang2019 · DOI

TL;DR

This paper introduces a novel, low-complexity acoustic modem for small autonomous underwater vehicles (AUVs) that tackles the challenge of Doppler shift compensation. By dynamically adjusting the analog-to-digital (AD) sampling rate, they achieve efficient Doppler correction without requiring computationally intensive resampling, making it ideal for resource-constrained AUVs.

What Problem Does This Paper Address?

Underwater acoustic (UWA) communication is crucial for AUVs, enabling them to transmit data and coordinate with each other for tasks like environmental monitoring and ocean sampling. However, UWA channels are notoriously challenging due to multipath fading and Doppler spread, especially for small AUVs. The Doppler effect, caused by the relative motion between the transmitter and receiver, distorts the signal's frequency and can severely degrade communication performance. Traditional methods for correcting Doppler shift often involve complex real-time resampling algorithms, which are too computationally demanding for the limited processing power and battery life of small AUVs. This paper addresses this gap by proposing a hardware-based Doppler compensation method that is both effective and computationally efficient, suitable for integration into compact AUV platforms.

Methods

The core innovation of this paper lies in its hardware-based Doppler compensation approach. Instead of relying on software-based resampling, the authors propose to directly adjust the analog-to-digital (AD) sampling rate of the receiver. This is achieved by first estimating the Doppler shift using two linear frequency modulation (LFM) signals embedded within the communication frame. The estimated Doppler factor is then used to calculate an appropriate AD sampling rate. By tuning this sampling rate, the system effectively "resamples" the incoming signal in hardware, compensating for the Doppler-induced frequency shifts.

The acoustic modem itself utilizes a Direct Sequence Spread Spectrum (DSSS) modulation scheme. DSSS is chosen for its robustness against interference, multipath, and low signal-to-noise ratio (SNR) conditions, which are common in UWA environments. The modem also incorporates convolutional coding to further enhance error resilience. The system's frame structure includes a synchronization header (first LFM signal) for initial lock and Doppler estimation, followed by a second LFM signal for more precise Doppler estimation, and then the data packet.

Datasets

Not discussed in the paper. The paper focuses on experimental validation using real-world sea trials and a lake trial.

Key Insights and Ideas

  • Hardware-based Doppler Compensation: The most significant insight is the successful implementation of Doppler compensation by dynamically adjusting the AD sampling rate. This bypasses the computational bottleneck of software resampling, making it practical for low-power, low-cost embedded systems like small AUVs.
  • Low-Complexity Design: The proposed method significantly reduces the computational load on the AUV's processor, enabling the use of simpler, less expensive microcontrollers (specifically, the STM32F4 series).
  • Integration of DSSS and Adaptive Sampling: The synergistic combination of DSSS modulation for robust communication and the adaptive sampling rate for Doppler correction proves effective in challenging underwater environments.
  • Performance Trade-offs: While software-based Doppler compensation offers slightly better bit error rate (BER) performance due to higher precision, the hardware approach is deemed superior for AUV integration due to its drastically lower computational requirements and power consumption.

Key Results

  • Accurate Doppler Estimation: The hardware Doppler estimation closely matched software-based estimation, indicating the reliability of the approach.
  • Improved BER with Compensation: The bit error rate (BER) was significantly reduced after Doppler compensation. In sea trials, BER dropped to less than 0.06 after compensation and further improved to under 0.01 with channel coding.
  • Error-Free Communication at Low Speeds: During AUV integration trials, error-free communication (BER of zero) was achieved for Doppler shifts of approximately 2 Hz.
  • Significant Improvement at Higher Speeds: For a Doppler shift of 10 Hz, BER was reduced from approximately 40% to 7% after Doppler correction and channel encoding.

Limitations and Future Work

Limitations: * The paper acknowledges that the hardware sampling chip and system clock precision are not as good as software implementations, leading to slightly worse BER performance compared to software-based compensation. * The proposed system is designed for low data rates, which is a characteristic of DSSS and UWA communication in general.

Future Work: * The authors suggest that this proposed Doppler compensation method has the potential for use in other small-sized and low-cost underwater vehicles. Further research could explore higher data rates or more complex modulation schemes while maintaining the low-complexity hardware approach. * Investigating more advanced channel coding techniques could further enhance communication reliability.

Summary updated

BibTeX
@misc{zheng2019design,
  author = {Siyuan Zheng and Feng Tong and Bin Li and Qiuyang Tao and Aijun Song and Fumin Zhang},
  title = {{Design and Evaluation of an Acoustic Modem for a Small Autonomous Unmanned Vehicle}},
  year = {2019},
  doi = {10.3390/s19132923},
}
Figure