Singular Value Decomposition (SVD) for PPG Signal Processing

Singular Spectrum Analysis (SSA) for PPG

Singular Spectrum Analysis (SSA) is a structured framework that builds on SVD and is particularly well-suited to quasi-periodic signals like PPG. SSA extends basic SVD denoising by adding a grouping step where individual SVD components are combined into meaningful groups (trend, oscillatory components, noise) and a diagonal averaging step that converts the grouped matrix components back into time series.

The SSA procedure for PPG processing follows five stages: embedding (trajectory matrix construction), SVD computation, grouping of elementary components, reconstruction via diagonal averaging, and optional forecasting for gap-filling or prediction.

Because SSA adaptively separates quasi-periodic components, it is a natural fit for PPG during transitions between rest and activity, where fixed frequency-domain filters struggle.

The window length L in SSA controls the resolution of the decomposition. For PPG applications, L should be chosen to span at least 2-3 complete cardiac cycles. At a heart rate of 60 BPM and 100 Hz sampling rate, one cardiac cycle is 100 samples, suggesting L >= 200-300. Golyandina and Zhigljavsky (2013) discuss window length selection in detail; in PPG practice, shorter windows are often used to reduce computational cost (doi:10.1007/978-3-642-34913-3).

SVD for PPG Feature Extraction

Beyond denoising, SVD provides a compact representation of PPG waveform morphology that is valuable for classification and regression tasks. The singular values and singular vectors themselves serve as features that capture the essential shape characteristics of the PPG pulse wave.

Waveform Morphology Analysis

The left singular vectors from SVD of a PPG trajectory matrix represent the principal waveform shapes present in the signal. The first left singular vector approximates the average pulse shape, while subsequent vectors capture systematic variations in morphology across beats. This representation can be used as a compact morphology feature set, for example in cuffless blood pressure estimation, where singular values and vectors from a beat-by-beat matrix of normalized pulses can complement pulse transit time features. For more on PPG-based blood pressure algorithms, see our pulse wave analysis guide.

The singular value spectrum itself may carry diagnostic information. Allen and Murray (2003) showed that PPG pulse shape changes with age at the ear, finger, and toe in 116 healthy subjects, with elongation of the systolic rising edge and damping of the dicrotic notch (doi:10.1088/0967-3334/24/2/306). In stiffer arteries, the singular value spectrum would therefore be expected to show a faster decay (energy more concentrated in the first component) due to reduced dicrotic notch amplitude and simplified waveform morphology. Conversely, healthy young subjects with compliant arteries show a more gradual singular value decay, reflecting richer waveform features including a prominent dicrotic notch and reflected wave components.

Dimensionality Reduction for Machine Learning

SVD provides an efficient dimensionality reduction technique for PPG-based machine learning models. Rather than feeding raw PPG samples or hand-crafted time-domain features into a classifier, the SVD representation captures the most informative aspects of the signal in a small number of components. Reducing a pulse of roughly 100 samples to a few components makes SVD preprocessing particularly valuable for models deployed on resource-constrained wearable devices.

SVD Combined with Other Methods

SVD rarely operates in isolation in modern PPG processing pipelines. Its strengths complement those of other signal processing techniques, and hybrid approaches often achieve superior performance.

SVD + Adaptive Filtering

A common pipeline uses SVD as a preprocessing step before adaptive filtering. SVD first separates the signal into a reduced set of components, and the adaptive filter (typically NLMS or RLS) then operates on the SVD-reconstructed signal with an accelerometer reference. The SVD step improves adaptive filter convergence by removing broadband noise, effectively increasing the signal-to-noise ratio of the input. This approach is described in the context of the motion artifact removal pipeline.

SVD + Wavelet Decomposition

Combining SVD with wavelet decomposition exploits the complementary strengths of both methods. Wavelets provide time-frequency localization, while SVD provides energy-based component separation. A typical hybrid approach applies wavelet decomposition first to isolate the frequency band of interest, then applies SVD within that band for fine-grained noise removal.

SVD + Deep Learning

SVD features can serve as inputs to deep learning models, combining the interpretability and efficiency of SVD with the nonlinear modeling power of neural networks. The SVD representation provides a compact, noise-reduced input that improves training efficiency and generalization compared to raw PPG input.

For more on how deep learning is applied to PPG analysis, see our guide on PPG signal processing algorithms. For details on atrial fibrillation detection from PPG, see our AFib detection guide.

Computational Considerations and Real-Time Implementation

The computational complexity of SVD is $O(mn \cdot \min(m,n))$ for an m x n matrix, which can be substantial for long PPG recordings. Several strategies make SVD practical for real-time wearable applications.

Sliding window SVD applies the decomposition to short overlapping segments (typically 5-10 seconds at 100 Hz, yielding matrices of 500-1000 x L). This bounds the computation and enables quasi-real-time operation with a latency equal to the window length.

Incremental SVD algorithms update the decomposition as new samples arrive, avoiding full recomputation. Brand (2006) described fast low-rank updates of the thin SVD, which allow the decomposition to be refreshed as new data arrive (doi:10.1016/j.laa.2005.07.021). With the small ranks typical of PPG, this makes near-sample-by-sample processing feasible.

Randomized SVD algorithms (Halko et al., 2011) find the k dominant singular components of a dense m x n matrix in $O(mn \log(k))$ floating-point operations, compared with $O(mnk)$ for classical algorithms, which is attractive when only a few components are needed (doi:10.1137/090771806).

On microcontrollers common in wearable devices, such as the ARM Cortex-M4, the cost of a sliding-window SVD depends on the window and embedding sizes and should be profiled for the target hardware; incremental updates reduce per-sample cost, enabling integration into existing heart rate estimation pipelines with minimal computational overhead.

Benchmarks and Performance Comparison

Systematic comparison of SVD-based methods against alternative PPG denoising approaches across published benchmarks reveals consistent patterns.

On motion-heavy benchmarks such as the IEEE Signal Processing Cup 2015 dataset (12 subjects, wrist PPG during intensive arm exercises), SVD-based methods generally sit between simple bandpass filtering and state-of-the-art deep learning methods in accuracy, while requiring far less training data and computation than neural network approaches.

On resting PPG, the primary advantage of SVD in the resting case is its ability to preserve waveform morphology, including the dicrotic notch and reflected wave components, which are often distorted by aggressive frequency-domain filtering.

The key tradeoff with SVD is the balance between denoising performance and morphological fidelity. Retaining too few singular values produces a clean but over-smoothed signal that loses subtle waveform features. Retaining too many preserves features but leaves residual noise. Adaptive rank selection methods (AIC, MDL, gap-based) provide the best balance but add computational overhead. For applications focused purely on heart rate extraction, aggressive truncation (k = 2-3) is appropriate. For morphology-dependent applications like blood pressure estimation or arterial stiffness assessment, more conservative truncation (k = 5-8) preserves the waveform details needed for accurate clinical measurements.

Conclusion

Singular Value Decomposition provides a mathematically principled and computationally tractable approach to PPG signal processing that occupies a valuable middle ground between simple filtering and complex machine learning methods. Its ability to separate signal components by energy without requiring frequency-domain assumptions makes it particularly effective for PPG signals where cardiac and noise components share overlapping spectral characteristics. When combined with complementary techniques such as adaptive filtering, wavelet decomposition, or deep learning feature extraction, SVD-based methods achieve state-of-the-art performance while maintaining the interpretability and efficiency needed for practical wearable device deployment. Researchers and engineers working on PPG systems should consider SVD as both a standalone denoising tool and a preprocessing component in more complex signal processing pipelines.

References

• Allen, J., & Murray, A. (2003). Age-related changes in the characteristics of the photoplethysmographic pulse shape at various body sites. Physiological Measurement, 24(2), 297-307. https://doi.org/10.1088/0967-3334/24/2/306
• Golyandina, N., & Zhigljavsky, A. (2013). Singular Spectrum Analysis for Time Series. Springer. https://doi.org/10.1007/978-3-642-34913-3
• Brand, M. (2006). Fast low-rank modifications of the thin singular value decomposition. Linear Algebra and its Applications, 415(1), 20-30. https://doi.org/10.1016/j.laa.2005.07.021
• Halko, N., Martinsson, P. G., & Tropp, J. A. (2011). Finding structure with randomness: probabilistic algorithms for constructing approximate matrix decompositions. SIAM Review, 53(2), 217-288. https://doi.org/10.1137/090771806