Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userKernel Learning for High-Resolution Time-Frequency Distribution
238
0
0.0
(
0
)
Added by
Haijian Zhang
Publication date
2020
fields
Electronic Engineering
and research's language is
English
Ask ChatGPT about the research
No Arabic abstract
The design of high-resolution and cross-term (CT) free time-frequency distributions (TFDs) has been an open problem. Classical kernel based methods are limited by the trade-off between TFD resolution and CT suppression, even under optimally derived parameters. To break the current limitation, we propose a data-driven kernel learning model directly based on Wigner-Ville distribution (WVD). The proposed kernel learning based TFD (KL-TFD) model includes several stacked multi-channel learning convolutional kernels. Specifically, a skipping operator is utilized to maintain correct information transmission, and a weighted block is employed to exploit spatial and channel dependencies. These two designs simultaneously achieve high TFD resolution and CT elimination. Numerical experiments on both synthetic and real-world data confirm the superiority of the proposed KL-TFD over traditional kernel function methods.
rate research
Read More
In this paper, we develop two high-resolution channel estimation schemes based on the estimating signal parameters via the rotational invariance techniques (ESPRIT) method for frequency-selective millimeter wave (mmWave) massive MIMO systems. The first scheme is based on two-dimensional ESPRIT (TDE), which includes three stages of pilot transmission. This scheme first estimates the angles of arrival (AoA) and angles of departure (AoD) and then pairs the AoA and AoD. The other scheme reduces the pilot transmission from three stages to two stages and therefore reduces the pilot overhead. It is based on one-dimensional ESPRIT and minimum searching (EMS). It first estimates the AoD of each channel path and then searches the minimum from the identified mainlobe. To guarantee the robust channel estimation performance, we also develop a hybrid precoding and combining matrices design method so that the received signal power keeps almost the same for any AoA and AoD. Finally, we demonstrate that the proposed two schemes outperform the existing channel estimation schemes in terms of computational complexity and performance.
Time-frequency distributions (TFDs) play a vital role in providing descriptive analysis of non-stationary signals involved in realistic scenarios. It is well known that low time-frequency (TF) resolution and the emergency of cross-terms (CTs) are two main issues, which make it difficult to analyze and interpret practical signals using TFDs. In order to address these issues, we propose the U-Net aided iterative shrinkage-thresholding algorithm (U-ISTA) for reconstructing a near-ideal TFD by exploiting structured sparsity in signal TF domain. Specifically, the signal ambiguity function is firstly compressed, followed by unfolding the ISTA as a recurrent neural network. To consider continuously distributed characteristics of signals, a structured sparsity constraint is incorporated into the unfolded ISTA by regarding the U-Net as an adaptive threshold block, in which structure-aware thresholds are learned from enormous training data to exploit the underlying dependencies among neighboring TF coefficients. The proposed U-ISTA model is trained by both non-overlapped and overlapped synthetic signals including closely and far located non-stationary components. Experimental results demonstrate that the robust U-ISTA achieves superior performance compared with state-of-the-art algorithms, and gains a high TF resolution with CTs greatly eliminated even in low signal-to-noise ratio (SNR) environments.
In audio signal processing, probabilistic time-frequency models have many benefits over their non-probabilistic counterparts. They adapt to the incoming signal, quantify uncertainty, and measure correlation between the signals amplitude and phase information, making time domain resynthesis straightforward. However, these models are still not widely used since they come at a high computational cost, and because they are formulated in such a way that it can be difficult to interpret all the modelling assumptions. By showing their equivalence to Spectral Mixture Gaussian processes, we illuminate the underlying model assumptions and provide a general framework for constructing more complex models that better approximate real-world signals. Our interpretation makes it intuitive to inspect, compare, and alter the models since all prior knowledge is encoded in the Gaussian process kernel functions. We utilise a state space representation to perform efficient inference via Kalman smoothing, and we demonstrate how our interpretation allows for efficient parameter learning in the frequency domain.
In this work, we present a neuromorphic system that combines for the first time a neural recording headstage with a signal-to-spike conversion circuit and a multi-core spiking neural network (SNN) architecture on the same die for recording, processing, and detecting High Frequency Oscillations (HFO), which are biomarkers for the epileptogenic zone. The device was fabricated using a standard 0.18$mu$m CMOS technology node and has a total area of 99mm$^{2}$. We demonstrate its application to HFO detection in the iEEG recorded from 9 patients with temporal lobe epilepsy who subsequently underwent epilepsy surgery. The total average power consumption of the chip during the detection task was 614.3$mu$W. We show how the neuromorphic system can reliably detect HFOs: the system predicts postsurgical seizure outcome with state-of-the-art accuracy, specificity and sensitivity (78%, 100%, and 33% respectively). This is the first feasibility study towards identifying relevant features in intracranial human data in real-time, on-chip, using event-based processors and spiking neural networks. By providing neuromorphic intelligence to neural recording circuits the approach proposed will pave the way for the development of systems that can detect HFO areas directly in the operation room and improve the seizure outcome of epilepsy surgery.
Given nonstationary data, one generally wants to extract the trend from the noise by smoothing or filtering. However, it is often important to delineate a third intermediate category, that we call high frequency (HF) features: this is the case in our motivating example, which consists in experimental measurements of the time-dynamics of depolymerising protein fibrils average size. One may intuitively visualise HF features as the presence of fast, possibly nonstationary and transient oscillations, distinct from a slowly-varying trend envelope. The aim of this article is to propose an empirical definition of HF features and construct estimators and statistical tests for their presence accordingly, when the data consists of a noisy nonstationary 1-dimensional signal. We propose a parametric characterization in the Fourier domain of the HF features by defining a maximal amplitude and distance to low frequencies of significant energy. We introduce a data-driven procedure to estimate these parameters, and compute a p-value proxy based on a statistical test for the presence of HF features. The test is first conducted on simulated signals where the ratio amplitude of the HF features to the level of the noise is controlled. The test detects HF features even when the level of noise is five times larger than the amplitude of the oscillations. In a second part, the test is conducted on experimental data from Prion disease experiments and it confirms the presence of HF features in these signals with significant confidence.
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
