-
塑料光纤上的光通信:集成光接收机技术
The Plastic Optical Fiber (POF) provides benefits compared to glass fiber. The best known POF is the step-index (SI) PMMA POF with a core diameter of 1 mm. It is simpler and less expensive than glass fibers. Large-core POF has greater flexibility and resilience to bending, shock, vibration, and it is easier in handling and connecting. The transmission windows of the POF are in the visible range. These advantages make POF very attractive for use in short-range communication and within in-building networks.
-
使用RF-DNA指纹识别操作环境的ZigBee设备
This research was performed to expand AFIT's Radio Frequency 'Distinct Native Attribute' (RF-DNA) fingerprinting process to support IEEE 802.15.4 ZigBee communication network applications. The fingerprints were constructed using a 'hybrid' pool of emissions collected under a range of conditions, including anechoic chamber and an indoor office environment where dynamic multi-path and signal degradation factors were present. The RF-DNA fingerprints were input to a Multiple Discriminant Analysis, Maximum Likelihood (MDA/ML) discrimination process and a 1 vs. many 'Looks most like' classification assessment made. The hybrid MDA model was also used for 1 vs. 1 'Looks how much like' verification assessment. ZigBee Device Classification performance was assessed using both full and reduced dimensional fingerprint sets. Reduced dimensional subsets were selected using Dimensional Reduction Analysis (DRA) by rank ordering (1) pre-classification KS-Test p-values and (2) post-classification GRLVQI feature relevance values. Assessment of Zigbee device ID verification capability.
-
使用机器学习算法的加密网络流量分类
This research evaluates the feasibility of using ML algorithms to classify web services within encrypted TLS flows. The ML algorithms are compared primarily based on classification accuracy. The runtimes of the classifiers are also considered, as classifiers must be able determine labels quickly in order to be used in near realtime network protection devices. Five ML algorithms are initially considered when analyzing only the first 12 packets: Naive Bayes, NBTree, LibSVM, J4.8, and AdaBoost+J4.8. AdaBoost+J4.8 and J4.8 produce the best accuracies and runtimes and are tested on flowlengths of 1-20 packets. J4.8 reaches a peak accuracy of 97.99at 14 packets. AdaBoost+J4.8 peaks later at 18 packets with 98.41accuracy. AdaBoost+J4.8 requires 21.55 microseconds to classify a single flow at peak accuracy, while J4.8 requires only 2.37 microseconds to classify at peak accuracy. The quick runtimes and high accuracies of the J4.8 and AdaBoost+J4.8 indicate that these ML algorithms are good choices for near real-time classification of web services within an encrypted TLS flow.
-
模型信息交换系统(MIXS)
This research invstigated the issue of facilitating network information exchange among models and more specifically concentrated on two primary objectives:(a) identify solutions to the model information exchange problem, focusing on the network, and (b) assess the feasibility of the implementation of the proposed solution and provide recommendation for its practical inmlementation. To address this issue, two options were explored:(a) keeping independent networks in place but developing associations among them in order to facilitate information exchange and (b) using one unified common network for all models.