Integer Arithmetic Algorithm for Fundamental Frequency Identification of Oceanic Currents.

Area of Science:

• Marine technology
• Sensor networks
• Edge computing

Background:

• Underwater sensor networks are vital for offshore aquaculture monitoring.
• Current data processing often relies on offline computation, increasing latency and cost.
• Limited bandwidth and high deployment costs necessitate efficient on-site data analysis.

Purpose of the Study:

• To develop an edge computing algorithm for extracting fundamental frequencies of water currents.
• To enable real-time data processing on ultra-low-power microcontrollers within underwater sensor networks.
• To reduce computational load and bandwidth requirements for offshore aquaculture monitoring.

Main Methods:

• An algorithm leveraging finite impulse response (FIR) filtering in the integer domain was developed.
• The algorithm utilizes only an integer arithmetic unit for computation.
• Implementation details focus on FIR tuning, memory optimization, and variable domain selection for reduced computational effort.

Main Results:

• The proposed algorithm successfully extracts fundamental frequencies of water currents.
• Validation using a shallow water model and real-world data showed a maximum error below 4%.
• The algorithm is suitable for deployment on commercial ultra-low-power microcontrollers.

Conclusions:

• The developed edge computing algorithm provides an efficient method for on-site current frequency analysis in underwater sensor networks.
• This approach significantly reduces the computational and communication overhead for offshore aquaculture monitoring.
• The algorithm's reliance on integer arithmetic makes it practical for resource-constrained sensor nodes.