A novel method for blind signal separation of single-channel and time-frequency overlapped multi-component signal

Pang, Lihui; Tang, Bin

doi:10.1504/ijict.2016.074838

Cited by 3 publications

(4 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The output layer with one neuron is used to provide the sum value of cost function. (40) m(w, t + 1) = β 1 m(w, t) + (1 − β 1 )∇J n (w), (41)…”

Section: Neural Networkmentioning

confidence: 99%

“…BSS problem has three mainstream methods, such as ICA, SCA and NMF, but not limited to those three methods, taking source signal characteristics-based BSS method proposed in reference [38][39][40], for example. Szu et al [38] proposed an effective singlechannel BSS method based on the limited character set feature of digital communication signal.…”

Section: Introductionmentioning

confidence: 99%

“…Warner et al [39] presented a single-channel separation approach based on the differences between shaping filters. Pang et al [40] proposed a novel BSS method for single-input multi-output (SIMO) system based on the periodicity of original signal, which can separate time-frequency overlapped multi-component signal effectively. Recently, a BSS method, combining the maximum likelihood estimation (MLE) criterion and a neural network (NN) with a bias term, is proposed in reference [41].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A MLE-based blind signal separation method for time–frequency overlapped signal using neural network

Pang

Tang

Tan

et al. 2022

EURASIP J. Adv. Signal Process.

View full text Add to dashboard Cite

The blind signal separation (BSS) algorithm obtains each original/source signal from the observed signal collected by the receiving antenna or sensor. Objective/loss/cost function and optimization method are two key parts of BSS algorithm. Modifying the objective function and optimization from the perspective of neural network (NN) is a novel concept in BSS domain. $$L_2$$ L 2 regularization is adopted as a term of maximum likelihood estimation (MLE)-based objective function like in Liu et al. (Sensors 21(3):973, 2021); however, we modified the probability density function (PDF) term of the objective function and used the kernel density estimation method for time–frequency overlapped digital communication signal. Multiple optimizers are studied in this paper, and we figure out the right optimizer for our application scenario. A varies of comparison experiments—whose separation results will be provided in forms of correlation coefficient and performance index—are carried out, which indicate our method can converge quickly and achieve satisfactory separation results with performance index (PI) lower than 0.02 when signal-to-noise ratio (SNR) no less than 10dB. Additionally, it demonstrates performance of our method is better than that of typical separation—FastICA, especially for the lower SNR environment, and it shows that our method is not sensitive to the frequency overlap level (FOL) of the source signal, even FOL as high as $$100\%$$ 100 % ; it still can get high-precision separation results with $$\textrm{PI}<0.02$$ PI < 0.02 .

show abstract

“…The output layer with one neuron is used to provide the sum value of cost function. (40) m(w, t + 1) = β 1 m(w, t) + (1 − β 1 )∇J n (w), (41)…”

Section: Neural Networkmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A MLE-based blind signal separation method for time–frequency overlapped signal using neural network

Pang

Tang

Tan

et al. 2022

EURASIP J. Adv. Signal Process.

View full text Add to dashboard Cite

show abstract

“…Adaptive array processing can modify weighting coefficients dynamically to maximise SINR, so that detection ability of targets can be enhanced greatly. If only coefficients of N array antenna elements was adaptive, corresponding filter can decrease spatial filtering, which is also called adaptive digital beam forming (ADBF) (Nuteson et al, 2002;Pang and Tang, 2016), and if time domain filter is also adaptive, we can obtain space-time adaptive processing (STAP) (Peckhan et al, 2000), which can filter jamming and clutter more efficiently. ADBF and STAP are typical adaptive array processing applications, meaning adaptive null steering for interface with unknown orientation (both spatial and time domain are included), while still keep constant gain on target's DOA.…”

Section: Introductionmentioning

confidence: 99%

Robust adaptive array processing based on modified multistage Wiener filter algorithm

Wang

Gong

Lian

et al. 2019

IJICT

View full text Add to dashboard Cite

Multistage Wiener filter (MSWF) is a very efficient algorithm for adaptive array processing because of low-complexity and prominent rank-reduction advantage. However, if training sample data was contaminated by outliers, especially when outliers having the same DOA with target emerge, the MSWF results will be decreased severely. In this paper, MSWFs backward iteration was improved, and median cascaded canceller (MCC) strategy was adopted so that optimal weighting calculation can be obtained via sorting and median processing, meaning impact of outliers were removed effectively. Blocking matrix solving of MSWF forward iteration was completed by Householder transform to enhance fix-point format performance. The new-designed algorithm attained excellent compromise between robustness and complexity. To verify presented algorithm's performance, array with 50 elements was established in simulation platform, and the simulated results also proved it can cope with outlier-contaminated applications effectively.

show abstract

Optimal Variable Momentum Factor Algorithm for NPCA in Blind Source Separation

Gao

Dong

et al. 2021

2021 IEEE 18th International Conference on Mobile Ad Hoc and Smart Systems (MASS)

View full text Add to dashboard Cite

A novel method for blind signal separation of single-channel and time-frequency overlapped multi-component signal

Cited by 3 publications

References 10 publications

A MLE-based blind signal separation method for time–frequency overlapped signal using neural network

A MLE-based blind signal separation method for time–frequency overlapped signal using neural network

Robust adaptive array processing based on modified multistage Wiener filter algorithm

Optimal Variable Momentum Factor Algorithm for NPCA in Blind Source Separation

Contact Info

Product

Resources

About