43 results about "Generalised likelihood ratio test" patented technology

A system and method for detecting network intrusions using one or more statistical models and a generalized likelihood ratio test (GLRT) is provided. The system includes a computer system and a network intrusion detection engine executed by the computer system. To detect network intrusions, the system receives network traffic data, computes a likelihood using one or more statistical models, such as an Markov-modulated Poisson process, and processes the traffic data using a GLRT. The statistical models are used to assess the likelihood of seeing a particular pattern of network traffic. The GLRT is used to classify a particular pattern as either indicative of an attack or not indicative of an attack. The system could apply one or more types of statistical models, such as in a flexible multi-tiered approach.

A method provides space-time adaptive processing (STAP) for target detection using adaptive matched filters (AMF). A generalized likelihood ratio test (GLRT) is determined where spatial and temporal correlation matrices Q and A are assumed. Then, the correlation matrices A and Q are replaced with maximum likelihood (ML) estimates obtained only from training signals subject to a persymmetric constraint.

A method and system realize reliable wireless communications in non-coherent multiple-input multiple-output (MIMO) doubly-selective channels. The method uses Grassmannian space-time-frequency block codes and an iterative generalized likelihood ratio test (GLRT) with a multi-dimensional basis expansion model (BEM), decision reordering, and a fixed number of surviving candidates. The computational complexity of a non-coherent MIMO equalizer becomes linear as a function of a code length and a size of a modulation alphabet. The codebook, the alphabet size, the bit labeling, and the block power are optimized using worst-case channel statistics or instantaneous channel states. The method can use soft-information feedback from error correction codes, such as low-density-parity-check codes to improve performance.

The invention discloses a frequency spectrum perceiving method based on the estimation of a signal arrival direction, which comprises the steps: in a cognitive radio system to which smart antenna technology is applied, the direction and the angle of signal transmission are used as a frequency spectrum opportunity, and a master user signal is subjected to two-step perception: firstly, the master user signal is detected by using a method based on a generalized likelihood ratio test, namely that the ratio of the maximum eigenvalue to the sum of the eigenvalues of a sample covariance matrix is used as test statistics; the test statistics are compared with a decision threshold which is set according to a given false alarm probability so as to decide whether the master user signal exists or not; if a master user does not exist, then the channel can be used, and if the master user exists, then a classical MUSIC (Multiple Signal Classification) algorithm is further used to estimate the arrival direction of the master user signal. A cognitive user can select other directions for communication, which do not cause disturbance to the master user, and the frequency spectrum perceiving method can effectively perceive time, frequency and the frequency spectrum vacancy of a space domain, perceive the frequency spectrum vacancy of an angle domain, and increase the frequency spectrum opportunity.

Codewords encoded using non-coherent codes and received at a receiver via non-coherent channels in a multi-input, multiple output (MIMO) network using orthogonal frequency demultiplexing (OFDM) are decode by concatenating multiple adjacent codewords of a received signal into a superblock at the receiver. A projector matrix based on a codebook is predetermining. Each codeword in the superblock is projected onto an orthogonal complement of a correspond transmitted codeword using the projector matrix to obtain a corresponding distance metric of a generalized likelihood ratio test (GLRT) codeword. A minimal distance metric is selected to obtain an estimate of the transmitted codeword corresponding to a transmitted signal.

The invention discloses a simulation method of MIMO (Multiple Input Multiple Output) radar target detection under non-Gaussian clutter environment. The method comprises the following steps of: establishing a target detection model based on the auxiliary data which is close to a detected unit and does not contain the target; substituting a likelihood function with a detection signal to get the detection statistic; under a condition that a non-Gaussian clutter covariance matrix is known, obtaining a theoretical MIMO radar target detector by using a generalized likelihood ratio test theory. Under a condition that the non-Gaussian clutter covariance matrix is unknown, an appropriate clutter covariance matrix is estimated based on the auxiliary data, and the generalized likelihood ratio is substituted by the estimated clutter covariance matrix, thus, a MIMO radar adaptive detector is proposed.

The invention discloses a radar target adaptive detection method based on linear fusion, which belongs to the field of radar signal processing. Aiming at non-Gaussian degree spatiotemporal variability of clutters in the actual environment, based on an adaptive match filter or a generalized likelihood ratio detection detector in the Gaussian background and adaptive normalized match filter in the non-Gaussian background, through simple and effective linear fusion rules, a suitable adaptive detector structure is designed by comprehensively adopting the optimal or sub-optimal clutter covariance matrix estimation method in the specific clutter background and rationally using clutter non-Gaussian feature information. The detector is simple in structure and has strong generalization ability. The detection performance of the detector, the amount of algorithm calculations and other needs are simultaneously taken into account. The constant false alarm rate robustness of the detector and the adaptability to the clutter environment are enhanced, and the radar target detection performance under unknown clutter environment is improved. The radar target adaptive detection method of the invention has popularization and application values.

An approach to determine cantilever movement is presented. An observer based state estimation and statistical signal detection and estimation techniques are applied to Atomic Force Microscopes. A first mode approximation model of the cantilever is considered and a Kalman filter is designed to estimate the dynamic states. The tip-sample interaction is modeled as an impulsive force applied to the cantilever in order to detect the presence of sample. A generalized likelihood ratio test is performed to obtain the decision rule and the maximum likelihood estimation of the unknown arrival time of the sample profile and unknown magnitude of it. The use of the transient data results in sample detection at least ten times faster than using the steady state characteristics.

The invention discloses a low-altitude target detection method for a frequency diversity array radar, belonging to the field of phased array radar low-altitude target detection. According to the method, due to the scanning characteristics of an FDA radar, the beam direction of the radar is related to a radial distance and a frequency difference, by introducing a frequency difference between different transmitting elements, a signal-to-noise ratio loss brought by a multipath effect in a low-altitude target environment can be effectively suppressed. Then, a generalized likelihood ratio test (GLRT) method is used to derive an FDA radar low-altitude target detector, and a target detection probability can be effectively improved under a certain false alarm rate. Finally, through the design of sub modules of matrix multiplication, matrix determinant calculation the like, an FDA radar low-altitude target detector is realized in an FPGA hardware platform, and the real-time detection of the FDAradar low-altitude target detection is effectively improved. In summary, according to the method, in a low-altitude target environment, the scanning characteristics of the FDA radar can be used to effectively complete the detection of a low-altitude target, and the method has high practicability in modern warfare.

The invention discloses a broadband radar target adaptive fusion detection method based on a skew-symmetry structure, belonging to the field of radar signal processing. Aiming at a broadband radar system for which auxiliary data is missing and a space symmetric linear array or a fixed coherent pulse recurrence interval is used, clutter background statistical characteristics are estimated through clutter covariance matrix skew-symmetry structured information in main data and the auxiliary data, target scattering point energy in all to-be-detected distance units is fused based on a one-step process generalized likelihood ratio detection criterion, the clutter information in the main data and the auxiliary data is used maximally, a broadband target skew-symmetry structured adaptive detector is designed, false alarm probability and detection probability analytical expression of the detector are derived, mathematical tool support is provided for detector threshold value setting and detection performance theoretical analysis, range spread target adaptive detection performance under an environment where the auxiliary data is missing is improved, and the method has a popularization and application value.

A method for decoding data symbols modulated with a corresponding codeword from a constellation set of codewords expands the constellation set of codewords with a set of basis functions to produce a basis-expanded constellation set and projects projecting a received modulated data symbol onto orthogonal complements of the basis expanded constellation set to obtain a set of distance metric of a generalized likelihood ratio test (GLRT) for each codeword of the constellation set. The set of basis functions includes a Fourier exponential basis function in a frequency domain, a Legendre polynomial basis function in a time domain, and a Fourier-Legendre product basis function in the frequency domain. The method selects a codeword corresponding to a minimal distance metric or a maximal correlation metric and decodes the data symbol from the received modulated data symbol using the codeword.

An improved system and method for detecting network anomalies comprises, in one implementation, a computer device and a network anomaly detector module executed by the computer device arranged to electronically sniff network traffic data in an aggregate level using a windowing approach. The windowing approach is configured to view the network traffic data through a plurality of time windows each of which represents a sequence of a feature including packet per second or flow per second. The network anomaly detector module is configured to execute a wavelet transform for capturing properties of the network traffic data, such as long-range dependence and self-similarity. The wavelet transform is a multiresolution transform, and can be configured to decompose and simplify statistics of the network traffic data into a simplified and fast algorithm. The network anomaly detector module is also configured to execute a bivariate Cauchy-Gaussian mixture (BCGM) statistical model for processing and modeling the network traffic data in the wavelet domain. The BCGM statistical model is an approximation of α-stable model, and offers a closed-form expression for probability density function to increase accuracy and analytical tractability, and to facilitate parameter estimations when compared to the α-stable model. Finally, the network anomaly detector module is further configured to execute a generalized likelihood ratio test for detecting the network anomalies.

The present invention discloses a sea-surface small target detection method based on front-back revenue reference particle filter which mainly solves the problem that a conventional technology is not suitable for detecting the sea-surface low-speed floating small targets. The method comprises the realization steps of 1) obtaining and partitioning the echo data; 2) selecting a to-be-detected distance unit Sd in an echo data block and dividing the to-be-detected distance unit Sd into the to-be-detected sub-units; 3) calculating the instantaneous frequency curve function estimation of the to-be-detected distance unit Sd; 4) calculating a Doppler steering vector h and the covariance matrix estimation of the k-th to-be-detected sub-unit zk; 5) utilizing the h and the covariance matrix estimation to calculate the generalized likelihood ratio test statistic amount of the sub-unit zk; 6) accumulating the generalized likelihood ratio test statistic amount of all to-be-detected sub-units to obtain the test statistic amount xi k of the to-be-detected distance unit Sd; 7) calculating a detection threshold T xi; 8) comparing the xi k and the T xi to determine the existence of the targets. The sea-surface small target detection method of the present invention enables the radar detection performance to be improved, and can be used to detect the sea-surface floating small targets.

The invention discloses a wireless microwave dew intensity monitoring method based on wet antenna cause discrimination. The method comprises the following steps: monitoring wireless microwave signal attenuation intensity; deducing dry period and wet period by using a Markov transformation model; using a generalized likelihood ratio test method to detect whether there is a wet antenna phenomenon caused by dew in the dry period; if the wet antenna phenomenon is detected, applying a trained ICA algorithm to separate out attenuation signals caused by a wet antenna from microwave signal attenuationintensity data received by the wireless microwave receiver; and establishing a model relationship between the attenuation value caused by the wet antenna and the dew intensity by using a Fresnel reflection formula to realize monitoring of the dew intensity. According to the method, a direct monitoring method and an indirect model method are combined, high-resolution monitoring of the dew intensity is achieved, a traditional dew monitoring method is innovated, meanwhile, an existing high-temporal-spatial-resolution microwave communication network is effectively utilized, manpower and materialresources are saved, and the monitoring precision of the dew intensity is improved.

A method for initial synchronisation in a random access procedure in LTE standards provides dividing the Physical Random Access Channel (PRACH) into a plurality of tiles, each comprising by a adjacent subcarriers, over which the channel response is assumed to be substantially flat, i.e. the frequency response is replaced over each subcarrier of a tile with an average frequency response. In an embodiment, an algorithm for initial synchronization is provided on the basis of a generalized likelihood ratio test (GLRT).

The invention discloses a self-adaptive target detection method of a frequency diversity array multiple-input-multiple-output radar, and mainly solves the problem that the existing frequency diversityarray multiple-input-multiple-output radar cannot realize self-adaptive target detection. The method comprises the steps of 1) constructing an equivalent received signal of the frequency diversity array multiple-input-multiple-output radar; 2) constructing a binary hypothesis test problem and generalized likelihood ratio test detection statistics according to the received signal; 3) carrying outoptimization solution on the test statistics in the step 2) by adopting an improved interval search method based on a Newton-like method to obtain improved interval search generalized likelihood ratiodetection statistics based on the Newton-like method; and 4) setting a detection threshold according to the actual situation, and comparing the improved interval search generalized likelihood ratio detection statistics based on the Newton-like method with the detection threshold to obtain a detection result. The method can achieve the self-adaptive detection of the target, improves the target detection performance of the radar, and can be used for the target recognition of the frequency diversity array multiple-input-multiple-output radar.

The invention relates to a repeating false target discriminating method based on angular glint. The method is characterized by establishing an angular glint statistical model according to a monopulse amplitude-comparing angle measurement method and performing correct discrimination by analyzing a difference between the azimuth error variances of an authentic target and a repeating false target and by using a Neyman-Pearson theorem and a generalized likelihood ratio test (GLRT) method, and comprises a step 1 of establishing angular glint statistical model according to the monopulse amplitude-comparing angle measurement method; a step 2 of estimating the parameter of angular observations in order to prepare for further target detection; a step 3 of acquiring a detection threshold according to a set false alarm probability, and then obtaining a detector output by using an echo signal in order to determine a threshold and finally obtain a detection result.

Methods and related systems are described for modified semblance criterions based on the approach of thresholding the signal energy. A first criterion is derived by posing the problem as that of detecting a signal with energy (or amplitude) greater than the specified threshold and deriving the generalized likelihood ratio test statistic. A second criterion is derived using the same method by posing the problem as that of rejecting any signal with energy (or amplitude) below a specified threshold and detecting it if its energy is above another threshold greater than or equal to the first. These appropriately modify the original semblance criterion which is shown to be equivalent to the GLRT test statistic in the absence of any threshold on the signal amplitude. In addition simpler modifications are also described. Tests on synthetic data illustrate the effectiveness of all these modifications which perform comparably well at suppressing unwanted arrivals while accurately processing the desired signals.

An approach to determine cantilever movement is presented. An observer based state estimation and statistical signal detection and estimation techniques are applied to Atomic Force Microscopes. A first mode approximation model of the cantilever is considered and an observer is designed to estimate the dynamic states. The cantilever-sample interaction is modeled as an impulsive force applied to the cantilever in order to detect the presence of sample. A generalized likelihood ratio test is performed to obtain the decision rule and the maximum likelihood estimation of the unknown arrival time of the sample profile and unknown magnitude of it. The use of the transient data results in sample detection at least ten times faster than using the steady state characteristics.

An approach to determine cantilever movement is presented. An observer based state estimation and statistical signal detection and estimation techniques are applied to Atomic Force Microscopes. A first mode approximation model of the cantilever is considered and an observer is designed to estimate the dynamic states. The cantilever-sample interaction is modeled as an impulsive force applied to the cantilever in order to detect the presence of sample. A generalized likelihood ratio test is performed to obtain the decision rule and the maximum likelihood estimation of the unknown arrival time of the sample profile and unknown magnitude of it. The use of the transient data results in sample detection at least ten times faster than using the steady state characteristics.

A method provides space-time adaptive processing (STAP) for target detection using adaptive matched filters (AMF). A generalized likelihood ratio test (GLRT) is determined where spatial and temporal correlation matrices Q and A are assumed. Then, the correlation matrices A and Q are replaced with maximum likelihood (ML) estimates obtained only from training signals subject to a persymmetric constraint.

The invention provides a time sequence concept drift detection method and system, a medium and equipment, and the method comprises the steps: carrying out decomposition of an obtained original time sequence signal through employing an empirical mode decomposition method based on extreme value symmetric extension, and obtaining an intrinsic mode component containing the feature information of different time scales of the original signal; performing coarse graining processing on the obtained intrinsic mode component through fuzzy entropy to obtain an intrinsic mode component converted by using the fuzzy entropy; monitoring the intrinsic mode component after fuzzy entropy conversion by adopting a non-parametric statistical process control model based on generalized likelihood ratio test, calculating the maximum degree of freedom of the generalized likelihood ratio test, and comparing the maximum degree of freedom with a preset control threshold to determine whether a mean value, a variance or drifting of the mean value and the variance occurs or not; conceptual drift detection is realized from the perspective of time domain characteristics of different frequencies.

The invention discloses an airplane icing on-line detection method based on statistical test and filtering. The airplane icing on-line detection method at least comprises the following steps that S1,flight state measurement data, engine thrust data and rudder deviation input data in the flight process are collected; S2, an icing starting moment is detected by utilizing a generalized likelihood ratio test method; S3, a disturbance signal is generated to be superposed on the rudder deviation input data, and joint state estimation and aerodynamic derivative identification influenced by icing areconducted on the flight state measurement data by utilizing an H-infinity filtering method; S4, an icing end moment is detected by utilizing the generalized likelihood ratio test method; and S5, generating and superposing of the rudder deflection disturbance signal are stopped, and joint state estimation and aerodynamic derivative identification influenced by icing continue to be conducted on themeasurement data by utilizing H-infinity filtering until the flight is finished. By combining quick icing detection and a parameter estimation method, an airplane icing detection algorithm with stronger functions is obtained.

The invention discloses a simulation method of MIMO (Multiple Input Multiple Output) radar target detection under non-Gaussian clutter environment. The method comprises the following steps of: establishing a target detection model based on the auxiliary data which is close to a detected unit and does not contain the target; substituting a likelihood function with a detection signal to get the detection statistic; under a condition that a non-Gaussian clutter covariance matrix is known, obtaining a theoretical MIMO radar target detector by using a generalized likelihood ratio test theory. Under a condition that the non-Gaussian clutter covariance matrix is unknown, an appropriate clutter covariance matrix is estimated based on the auxiliary data, and the generalized likelihood ratio is substituted by the estimated clutter covariance matrix, thus, a MIMO radar adaptive detector is proposed.

An approach to determine cantilever movement is presented. An observer based state estimation and statistical signal detection and estimation techniques are applied to Atomic Force Microscopes. A first mode approximation model of the cantilever is considered and a Kalman filter is designed to estimate the dynamic states. The tip-sample interaction is modeled as an impulsive force applied to the cantilever in order to detect the presence of sample. A generalized likelihood ratio test is performed to obtain the decision rule and the maximum likelihood estimation of the unknown arrival time of the sample profile and unknown magnitude of it. The use of the transient data results in sample detection at least ten times faster than using the steady state characteristics.

The invention discloses a non-intrusive household appliance load identification method based on adaptive feature selection, which is specifically implemented according to the following steps: data preprocessing: denoising a selected REDD data set; performing event detection on the processed data through an improved generalized likelihood ratio test; extracting multi-dimensional load characteristics of the detected event change points; carrying out segmentation according to the power, and for multi-dimensional characteristics, using an adaptive mRmR algorithm to extract characteristics corresponding to each segment; taking the selected characteristics as load marks, and establishing a load feature library through an improved K-means algorithm; and on the basis of the load characteristic library, using the kNN algorithm to identify the load working state of the household electrical appliances in the user. The problem that in the prior art, misjudgment is likely to happen is solved, and the recognition accuracy is improved.

The invention discloses a self-adaptive millimeter wave beam layered search method. The method comprises the following steps: 1, assuming a beam search scene and modeling a multi-layer beam codebook; 2, modeling a beam hierarchical search pilot signal; 3, laying self-adaptive wave beams into any layer, and using a generalized likelihood ratio for checking and judging whether the signal-to-noise ratio of the received pilot signals reaches a target value or not; if the target value is reached or the wave beam search time of the layer reaches a preset maximum value, stopping the search of the layer; otherwise, continuing the layer search; 4, searching any layer except the last layer in a layered manner, and selecting the to-be-searched angle space of the next layer after the single-layer search stopping condition in the step 3 is met; and 5, after the search of the last layer is stopped, selecting a final beam direction. According to the method, the beam search time can be automatically adjusted to ensure the beam search accuracy under the condition that the path loss and the SNR are unknown, the search time is automatically prolonged to ensure the search precision under the condition of low SNR, and the search time is reduced and the system overhead is reduced under the condition of high SNR.

The invention discloses an airborne radar moving target detection method and device, equipment and a storage medium. The method comprises the steps: receiving an echo signal of a single snapshot distance unit, obtaining a covariance matrix of the echo signal of the single snapshot distance unit, obtaining a radar emission signal parameter, and obtaining a clutter basis matrix according to the radar emission signal parameter; calculating Gaussian white noise under a first condition and Gaussian white noise under a second condition according to the covariance matrix and the clutter basis matrix; calculating to obtain a generalized likelihood ratio detector; and detecting a moving target through a generalized likelihood ratio detector. According to the radar target detection method based on the generalized likelihood ratio test, clutter components in single-snapshot echo signals are approximately expressed as subspace vector weighted sums; then improved least square and maximum likelihood estimation methods are adopted to estimate clutter, noise and target amplitude; and finally, the detector suitable for the airborne radar to detect the moving target in the non-uniform environment is obtained.

The invention discloses a wireless microwave dew intensity monitoring method based on wet antenna cause discrimination. The method comprises the following steps: monitoring wireless microwave signal attenuation intensity; deducing dry period and wet period by using a Markov transformation model; using a generalized likelihood ratio test method to detect whether there is a wet antenna phenomenon caused by dew in the dry period; if the wet antenna phenomenon is detected, applying a trained ICA algorithm to separate out attenuation signals caused by a wet antenna from microwave signal attenuationintensity data received by the wireless microwave receiver; and establishing a model relationship between the attenuation value caused by the wet antenna and the dew intensity by using a Fresnel reflection formula to realize monitoring of the dew intensity. According to the method, a direct monitoring method and an indirect model method are combined, high-resolution monitoring of the dew intensity is achieved, a traditional dew monitoring method is innovated, meanwhile, an existing high-temporal-spatial-resolution microwave communication network is effectively utilized, manpower and materialresources are saved, and the monitoring precision of the dew intensity is improved.