Blind Source Separation in Perspective of ICA Algorithms: A Review

Agrawal, Jharna; Gupta, Manish Kumar; Garg, Hitendra

doi:10.1109/cises54857.2022.9844373

Cited by 4 publications

(3 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…with training batch size B, predict label p i ∈ R 1×8 and the true label t i ∈ R 1×8 . L is denoted as the MSE loss function represented asL(x i , y i ) = (x i − y i ) 2 , where x i and y i are the predict label and true label, respectively.…”

Section: Network Architecturementioning

confidence: 99%

“…BSS algorithms, such as independent component analysis (ICA) and non‐negative matrix factorization (NMF) [1], can be used to separate overlapping radio signals and extract the individual components, enabling their recognition and analysis. However, in general, ICA lacks the ability to accurately determine the exact number of source signals or the appropriate scaling of those source signals in [2]. NMF has proven to be a powerful approach for addressing the challenges of source recognition and separation with scaling preservation.…”

Section: Introductionmentioning

confidence: 99%

“…Thus, we summarize two key challenges in this scenario as following: (1) In low SNR, characteristics of the overlapping signals spectrum are strongly obscured by the noise, leading to compromising identification accuracy. (2) In the propagation of wireless signals transmission, each individual component of the overlapping signals experiences diverse propagation loss, which means the scaling among each component signal exists large differences. That will introduce the difficulty on the scaling preservation.…”

mentioning

confidence: 99%

See 2 more Smart Citations

JCCM: Joint conformer and CNN model for overlapping radio signals recognition

Liang,

Li,

Liang

et al. 2023

Electronics Letters

View full text Add to dashboard Cite

Overlapping radio signals recognition is attracting more attention as the development and ubiquitous application of radio technologies. The traditional blind signal separation (BSS) method is mostly not effective when both radio propagation effects and low signal‐to‐noise ratio (SNR) scenarios are taken into consideration. In this letter, joint conformer and CNN model (JCCM) is proposed to separate and recognize the overlapping radio signals which are also unknown by the monitor node. JCCM utilizes the attention mechanism of conformer to encode signal spectrum into feature maps and decodes the feature maps into signal component proportions by perceiving the global and local features through convolutional neural networks (CNN). In addition, a signal preprocessing module is designed including MinPool and AvgPool layers for signal denoising. JCCM is evaluated in terms of signal separation, scaling preservation and principal signal classification experiments. The results show that JCCM has better performance especially in low SNR conditions.

show abstract

Section: Network Architecturementioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

mentioning

confidence: 99%

See 1 more Smart Citation

JCCM: Joint conformer and CNN model for overlapping radio signals recognition

Liang,

Li,

Liang

et al. 2023

Electronics Letters

View full text Add to dashboard Cite

show abstract

Blind Source Separation of Electromagnetic Signals Based on Swish-Tasnet

Chen,

Liu,

Mao

et al. 2024

Circuits Syst Signal Process

View full text Add to dashboard Cite

A Robust Hybrid Neural Network Architecture for Blind Source Separation of Speech Signals Exploiting Deep Learning

Ansari,

Alnajjar,

Khater

et al. 2023

IEEE Access

View full text Add to dashboard Cite

In the contemporary era, blind source separation has emerged as a highly appealing and significant research topic within the field of signal processing. The imperative for the integration of blind source separation techniques within the context of beyond fifth-generation and sixth-generation networks arises from the increasing demand for reliable and efficient communication systems that can effectively handle the challenges posed by high-density networks, dynamic interference environments, and the coexistence of diverse signal sources, thereby enabling enhanced signal extraction and separation for improved system performance. Particularly, audio processing presents a critical domain where the challenge lies in effectively handling files containing a mixture of human speech, silence, and music. Addressing this challenge, speech separation systems can be regarded as a specialized form of human speech recognition or audio signal classification systems that are leveraged to separate, identify, or delineate segments of audio signals encompassing human speech. In various applications such as volume reduction, quality enhancement, detection, and identification, the need arises to separate human speech by eliminating silence, music, or environmental noise from the audio signals. Consequently, the development of robust methods for accurate and efficient speech separation holds paramount importance in optimizing audio signal processing tasks. This study proposes a novel three-way neural network architecture that incorporates transfer learning, a pre-trained dual-path recurrent neural network, and a transformer. In addition to learning the time series associated with audio signals, this network possesses the unique capability of direct context-awareness for modeling the speech sequence within the transformer framework. A comprehensive array of simulations is meticulously conducted to evaluate the performance of the proposed model, which is benchmarked with seven prominent state-of-the-art deep learning-based architectures. The results obtained from these evaluations demonstrate notable advancements in multiple objective metrics. Specifically, our proposed solution showcases an average improvement of 4.60% in terms of short-time objective intelligibility, 14.84% in source-to-distortion ratio, and 9.87% in scale-invariant signal-to-noise ratio. These extraordinary advancements surpass those achieved by the nearest rival, namely the dual-path recurrent neural network time-domain audio separation network, firmly establishing the superiority of our proposed model's performance.

show abstract

Blind Source Separation in Perspective of ICA Algorithms: A Review

Cited by 4 publications

References 26 publications

JCCM: Joint conformer and CNN model for overlapping radio signals recognition

JCCM: Joint conformer and CNN model for overlapping radio signals recognition

Blind Source Separation of Electromagnetic Signals Based on Swish-Tasnet

A Robust Hybrid Neural Network Architecture for Blind Source Separation of Speech Signals Exploiting Deep Learning

Contact Info

Product

Resources

About