Abstract:
An apparatus, a method, and a non-transitory computer readable medium for event detection of passive seismic data are disclosed. The apparatus includes processing circuitry extracts features from the passive seismic data based on a backbone subnetwork of a residual deep neural network. The processing circuitry generates bounding box proposals for a region of interest (ROI) in the passive seismic data based on the extracted features being input to a region proposal network of the residual deep neural network. The processing circuitry classifies the bounding box proposals into two groups. Each bounding box proposal in a first group indicates that a corresponding seismic signal presents in the ROI. Each bounding box proposal in a second group indicates that no seismic signal presents in the ROI. The processing circuitry determines at least one seismic signal in the ROI from the first group of bounding box proposals.
Abstract:
A method, non-transitory computer readable medium, and system for multiple-input multiple-output for blind identification that includes receiving an input signal, originated as an output signal of a transmitter, at a receiver. A signal processing module can obtain the input signal from the receiver. The signal processing module can use a finite impulse response filter and one or more matrices derived from the input signal to minimize a cost function and obtain a parameter matrix. The parameter matrix can then be used to estimate the output signal by generating one or more Toeplitz matrices using the parameter matrix.
Abstract:
A device, method, and non-transitory computer readable medium that for two-dimensional blind single-input multiple-output channel identification for image restoration. The method includes receiving, by a receiver having independent channels, a two-dimensional image data matrix then transforming the received two-dimensional image data matrix to a one-dimensional image vector. Channel parameters can then be estimated using the one-dimensional image vector. The method can then construct a restored image using the estimated channel parameters and the two-dimensional image data matrix.
Abstract:
A system, method, and non-transitory computer readable medium that perform blind signal estimation for single-input multiple-output systems. The method can include receiving, by the two or more receiver antennas of the receiver, an observed signal comprising the input signal and an additive noise term. The method can then form a data matrix using the observed signals from the two or more receiver antennas. The method can also include computing a singular value decomposition of the data matrix. The singular value decomposition can then be used to generate a parameter matrix. The method can then form a Toeplitz signal matrix using the parameter matrix. The method can estimating the input signal using the Toeplitz signal matrix.
Abstract:
A device includes circuitry configured to determine feedforward and feedback coefficients for an adaptive frequency-domain decision feedback equalizer (AFD-DFE) based on previously received signals. The equalizer output is determined by applying the feedforward and feedback coefficients of the AFD-DFE to a received signal, and the feedforward and feedback coefficients of the AFD-DFE are updated based on the equalizer output.
Abstract:
A method and apparatus for equalization in a communication system. The method includes receiving an input of a block of symbols, filtering the input in the frequency domain, calculating an error factor in the time domain based on the filtered input, updating tap weights of the equalizer in the frequency domain using circular correlation based on the error factor and the conjugate of the input in the frequency domain, and outputting the filtered input.
Abstract:
A device includes circuitry configured to determine feedforward and feedback coefficients for an adaptive frequency-domain decision feedback equalizer (AFD-DFE) based on previously received signals. The equalizer output is determined by applying the feedforward and feedback coefficients of the AFD-DFE to a received signal, and the feedforward and feedback coefficients of the AFD-DFE are updated based on the equalizer output.