4results about How to "Improve authentication accuracy" patented technology

This invention discloses a behavior-adaptive continuous identity authentication method and system. Through component attribution analysis, the basic continuous identity authentication model is deconstructed into multiple computational units. These computational units are used as components, and an ablation vector is designed to determine whether the component's parameters are set to zero during inference. Ablation experiments are conducted on the components based on a user behavior dataset to obtain a set of optimal-performing components. Through parameter search adaptation, scaling factors are applied to the parameters of key components based on the current user's behavior data. With the optimization objective of maximizing identity authentication accuracy, an optimization search is performed to obtain the optimal scaling factor. This optimal scaling factor is then applied to the key components of the basic continuous identity authentication model to generate an adaptive continuous identity authentication model for the current user's behavior, enabling continuous identity authentication for the current user. This invention improves the authentication accuracy and computational efficiency of continuous identity authentication, and is particularly suitable for mobile devices.

The invention discloses a distributed optical fiber sensing event noise reduction method based on a denoising mask auto-encoder. The method comprises the following steps: step 1, carrying out DAS data acquisition; step 2, data structuring; step 3, data enhancement; step 4, constructing a pre-training network based on a de-noising mask auto-encoder; step 5, non-label data pre-training: taking the pre-processed non-label image as input, adding noise and masking, reconstructing an original image through an encoder E1 and a decoder D, carrying out repeated training to enable the reconstructed image to approach to a real image, and storing the weight of the pre-trained encoder E1; step 6, constructing a downstream classification task network; step 7, performing fine tuning on the downstream classification task network: using a small amount of labeled images, performing fine tuning on an encoder E2 under a random smooth framework, so that the encoder E2 keeps consistent prediction under noise input, and the authentication robustness is improved; according to the invention, end-to-end high-fidelity noise reduction is realized, and the authentication accuracy is improved.

The invention relates to an unsupervised multi-user physical layer authentication method based on a twin converter neural network, and belongs to the technical field of wireless communication security. The method specifically comprises the following steps: system modeling: constructing a dynamic multi-user wireless communication system consisting of a plurality of legal sending ends in a mobile state, at least one attacker and a single legal receiving end, and collecting channel state information (CSI) of each device under a multipath and time-varying channel through wireless links such as WiFi; carrying out expansion, amplitude calculation and normalization on the collected complex CSI to obtain CSI amplitude samples arranged according to a time sequence; the method comprises the following steps: constructing CSI of the same legal equipment at adjacent moments into a positive sample pair based on short-time invariance characteristics of a dynamic channel, and constructing CSI with a relatively large time interval and / or CSI from other equipment into a negative sample pair, so as to form an unsupervised contrast learning sample set; the method comprises the following steps: constructing a lightweight twin Transform model CSI-UST, carrying out linear embedding, position coding and self-attention feature extraction on a CSI amplitude sequence, and training by comparing a loss function, so that the CSI feature distance of the same equipment within a short time is converged, and the CSI feature distance of different equipment or moments is greater than a preset boundary; in an authentication stage, CSI to be authenticated and historical CSI of each legal device are sequentially input into CSI-UST to calculate a feature distance, a source device is determined based on a distance threshold, a CSI sample of a corresponding device is updated when authentication succeeds, and a new device is incorporated into a legal device set when threshold determination fails and upper layer confirmation passes, thereby realizing dynamic expansion without updating model parameters. The method does not need to depend on the CSI of an attacker or large-scale annotation data, is low in model parameter quantity and small in memory occupation, can still keep high authentication accuracy and low misjudgment rate in various dynamic scenes, and is suitable for being deployed in resource-limited Internet of Things equipment.

This invention discloses a radio frequency fingerprinting method and apparatus based on multi-dimensional time series feature fusion, comprising: extracting sampled data corresponding to the signal data of a wireless communication device and performing standardization processing; extracting multi-dimensional signal features from the processed sampled data and constructing a multi-dimensional feature vector; constructing a multi-dimensional feature sequence and corresponding labels based on the multi-dimensional feature vectors of consecutive frames to obtain a dataset for wireless communication device authentication; training a classifier using the dataset to obtain a trained classifier; inputting the signal feature sequence to be identified into the trained classifier; determining the optimal sequence length based on the obtained initial identification results; using the identification results with the optimal sequence length corresponding to the initial identification results as the target identification results; and performing multiple authentications based on the target identification results to obtain the final identification result. This invention improves the fingerprint recognition accuracy of devices by fully exploiting the sequential characteristics of signals in the time dimension.