Modeling and Simulation of Signal To Noise Ratio in Non Stationary Acoustic
Implementation Plan:
—————————
Step 1: Initially, we collect and load audio data from Hear my ship dataset.
Step 2: Then, we separate noise from signal and reconstruct clean sound using the Dually Regularized Non Negative Matrix Factorization (DR-NMF) method .
Step 3:Next, we optimize SNR and remove residual noise using the VG-ISAC hybrid denoising strategy for clearer signal reconstruction.
Step 4: Next, we enhance weak target detection by extracting meaningful features using the CEA-EITD-FAE hybrid Net method based on collected data.
Step 5: Next, we perform reverberation process using the ERCL-Attention Net model with wavelet-based reconstruction for accurate echo signals based on collected data.
Step 6: Finally, we plot performance metrics for the following
6.1: Number of Epochs vs. Accuracy (%)
6.2: Number of Epochs vs. Precision (%)
6.3: Number of Epochs vs. F1-Score (%)
6.4: Number of Epochs vs. Recall (%)
6.5: SNR (dB)
6.6: Frequency (kHz)
Software Requirements:
—————————–
1. Development Tool: Matlab-R2023a or above
2. Operating System: Windows-10 (64-bit) or above
Dataset:
———–
Link : Hear My Ship (https://hearmyship.fer.hr/)
Note:
——-
1) If the proposed plan does not fully align with your requirements, please provide all necessary details—including steps, parameters, models, and expected outcomes—in advance.
2) Kindly ensure that any missing configurations or specifications are clearly outlined in the plan before confirming.
3) If there’s no built-in solution for what the project needs, we can always turn to reference models, customize our own, different math models or write the code ourselves to fulfil the process.
4) If the plan satisfies your requirement, Please confirm with us.
5) Project based on Simulation only.
We perform with an Existing Approach Ref 1 – Title: Blind Source Separation and Denoising of Underwater Acoustic Signals
