34 results about "Neyman pearson" patented technology

The invention discloses an epilepsia electroencephalogram signal classified detection device which comprises a module for carrying out small wave analysis on a normal electroencephalogram signal and an epilepsia electroencephalogram signal, a module for calculating approximate entropy of each layer of detailed signal obtained after small wave decomposition of the small wave analysis device, and a module for further carrying out classified detection utilizing a Neyman-Pearson criteria. According to the invention, various different types of epilepsia activities are distinguished from the obtained epilepsia signal by clinical data, electroencephalogram signal characteristics are extracted by a small wave analysis and approximate entropy combined mode, judgment and detection are carried out by the NEYMAN-PEARSON criteria, and as compared with the obtained result with unclassified detection result, the detection rate of the epilepsia electrical activity of brain is improved.

The invention discloses a method for determining the quality of an iris image based on machine learning. The method comprises the following steps of: pre-processing the iris image; extracting quality factors of the iris image; fitting probability density functions of positive and negative samples of single quality factors by using multiple Gaussian models; performing fusion by using an improved Neyman-Pearson method to acquire the quality score of the iris image; and determining an optimal quality level by a hypothesis testing method. The invention provides a robust detecting method aiming at defocusing, motion blur and heterotropia; and multiple quality factors are fused by introducing the Neyman-Pearson method to form the quality score, and the image quality level with statistical significance is acquired by the hypothesis testing method. The method can be used for quality determination when the iris image is acquired and performance prediction aiming at a recognition algorithm.

The invention provides a method for detecting SAR image change based on two-dimensional Gamma distribution, comprising the following steps: according to the input SAR image to be detected and reference image data, the parameters of the two-dimensional Gamma distribution is estimated by a moment estimation method; likelihood ratio statistics are formed by Neyman-Pearson criterion; based on the two-dimensional distribution, clutter suppression is carried out according to the dependency of the image data, thus gaining the image after clutter suppression; CFAR normalization is carried out on the image after clutter suppression; furthermore, global thresholds are set so as to binarize the image, thus obtaining an initial detection result; the binary image after detection is processed morphologically, counting-filtered, and target-clustered so as to further eliminate isolated false alarm points, thus gaining the final detection result. The method reaches higher detection rate based on the low false alarm point, and is applicable to detect artificial objects under various clutter environments, more especially under the strong clutter environments.

A method of diagnosing pathologic heart conditions in which a time series of heart sounds is filtered and parsed into a sequence of individual heart cycles. A systolic interval as well as systolic sub-intervals are identified for each heart cycle. The systolic intervals and ECG peaks are then digitally filtered to optimize for click detection. For each heartcycle, systole time limits are determined, a time series of the transform at specific wavelet scales are input to a Neyman-Pearson “constant false alarm rate” (CFAR) detector to identify anomalously high wavelet coefficients, and a vector of detections vs. time is created. The series of anomalously high detections (one series for each heart cycle) are then assembled into a matrix and convolved with an averaging vector yielding detection statistics across heart cycles and time intervals consistent with an observed spread of click occurrence times. A click score is then determined as the maximum element of the vector formed by the median wavelet coefficient amplitude across heart cycles squared at each time sample multiplied element-wise by the vector formed by the sum across heart cycles of the number of detections at each time sample. The click score is compared to a threshold value set by a desired probability of detection vs. a probability of false alarm tradeoff. If the click score is less than the threshold then a “no click” indicator is displayed. If the click score is greater than the threshold then a “click present” indicator is displayed.

The invention discloses a spectrum detection method based on the number of MTM-SVD adaptive sensor, and relates to the technical field of cognitive radio spectrum detection method. The MTM-SVD spectrum detection technology has the advantages that the prior information of a primary user is not needed, the detection and the reconstruction of signals submerged in colour noises are processed effectively, algorithm velocity is fast and the like, so the method of the invention can derive a threshold formula according to Neyman-Pearson standard based on MTM-SVD algorithm and determine the optimal number of real needed sensors. The receiving signals on each sensor are sampled, the characteristic spectrum of each sampled signal is calculated, and a new judgment amount TM(fi) is calculated according to the obtained characteristic spectrum; the new judgment amount TM(fi) is compared with a threshold th satisfying a false-alarm probability Pf to judge whether the spectrum is occupied or not. Compared with traditional energy detection, the method of the invention reduces the complexity of system while detecting the signal of the primary user correctively.

The invention discloses a Neyman-Pearson criterion-based zero speed detection method, which comprises the following steps of receiving measured information output by sensors in real time during footstep motions in a single-soldier navigation system by using a handheld computer; determining a window function N according to system sampling frequency and data transmission speed; converting a zero speed detection problem into a modeled mathematical problem by utilizing a double-hypothesis testing theory, and obtaining a Neyman-Pearson criterion-based zero detection inequality; determining mathematical models of output signals of miniature inertia measurement unit sensors and received signals of the handheld computer; calculating a joint probability density function of the output signals of the miniature inertia measurement unit sensors; replacing unknown elements in the inequality with maximum likelihood estimation values of unknown signal elements to obtain an extensive probability likelihood ratio inequality; and substituting the output data of a miniature inertia measurement unit into the extensive probability likelihood ratio inequality, and further detecting a zero speed state. According to the method, problems are mathematized and modeled, and the detection accuracy is improved.

To provide an anomalous sound detection training technique by which a feature amount extraction function for detecting anomalous sound can be generated irrespective of whether training data for anomalous signals is available or not. An anomalous sound detection training apparatus includes: a first function updating unit 3 that updates a feature amount extraction function and an feature amount inverse transformation function, which are input, based on an optimization index of a variational autoencoder; an acoustic feature extraction unit 4 that extracts an acoustic feature of normal sound based on training data for normal sound; a normal sound model updating unit 5 that updates a normal sound model by using the acoustic feature that is extracted; a threshold updating unit 6 that obtains a threshold φ_ρ corresponding to a false positive rate ρ, which has a predetermined value, by using the training data for normal sound and the feature amount extraction function that is input; and a second function updating unit 8 that updates the feature amount extraction function that is updated, based on a Neyman-Pearson-type optimization index defined by the threshold φ_ρ that is obtained, and repeatedly performs processing of each of the above-mentioned units.

The invention provides a waveform optimization method of static clutter environment moving target detection based on priori knowledge. The waveform optimization method of static clutter environment moving target detection based on priori knowledge can utilize the priori knowledge to obtain the priori information of target response and clutter response, can determine the moving target echo signal model, can design a transmitted waveform frequency domain according to the Neyman-Pearson criterion so as to optimize the waveform, facing the moving target, based on the verification and measurement ratio, can analyze different noise types, can design the optimum transmitted waveform energy spectrum ESD, and then can perform inverse transformation on the time domain for the next impulse transmission. Therefore, in the same false alarm probability, the detection probability facing the moving target can achieve maximum.

The invention discloses a method for performing waveform optimization on a motion target in a complex clutter environment. The method includes calculating a motion target model having constant radialvelocity, and determining static clutter response and motion clutter response in a complex clutter environment; obtaining the optimal waveform of a noise characteristic based echo signal facing detection probability according to a Neyman-Pearson criterion; and obtaining the optimal energy spectral density (ESD) of an emission signal according to the minimum value of power spectral density (PSD) inthe noise characteristics. Through prior information such as static clutter, motion clutter and targets, the method can design an energy spectrum of the emission signal and detect a motion target byutilizing the optimal waveform in the case of different noise; and compared with traditional linear frequency modulation and phase encoding signals, the method can make the detection probability facing to the motion target reach to the maximum in the case of the same false alarm probability.

The invention discloses a URLLC system performance evaluation method based on physical layer authentication. The method comprises the steps that a transmitting end transmits a frame to a receiving end, the frame comprises a pilot signal, an authentication signal and an information signal, the authentication signal is superposed on the information signal, the information signal is obtained by performing channel coding and modulation on an initial signal, and the authentication signal is obtained based on the information signal, a Hash function and a secret key; The receiving end calculates theframe error probability based on the frame, further calculates the decoding probability of data transmission, obtains the false alarm probability based on the frame and hypothesis inspection conditions, and further obtains the average false alarm probability; based on the Neyman-Pearson theory theory, an average false alarm probability is set to be equal to an upper limit of the false alarm probability, an optimal threshold is obtained, a detection probability is obtained based on the optimal threshold, and an average detection probability is obtained based on the detection probability; And when the decoding probability and the average detection probability meet system requirements, the frame passes the authentication, and the receiving end obtains the throughput based on the decoding probability and the average detection probability so as to evaluate the performance of the URLLC system.