Browsing by Author "Zhao, Jingjun"

Filter results by typing the first few letters
Now showing 1 - 2 of 2
  • No Thumbnail Available
    Item
    Bayesian Sparse Factor Analysis of High Dimensional Gene Expression Data
    (North Dakota State University, 2019) Zhao, Jingjun
    This work closely studied fundamental techniques of Bayesian sparse Factor Analysis model - constrained Least Square regression, Bayesian Lasso regression, and some popular sparsity-inducing priors. In Appendix A, we introduced each of the fundamental techniques in a coherent manner and provided detailed proof for important formulas and definitions. We consider provided introduction and detailed proof, which are very helpful in learning Bayesian sparse Factor Analysis, as a contribution of this work. We also systematically studied a computationally tractable biclustering approach in identifying co-regulated genes, BicMix, by proving all point estimates of the parameters and by running the method on both simulated data sets and a real high-dimensional gene expression data set. Missed derivation of all point estimates in BicMix has been provided for better understanding variational expectation maximization (VEM) algorithm. The performance of the method for identifying true biclusters has been analyzed using the experimental results.
  • No Thumbnail Available
    Item
    A Two-phase Security Mechanism for Anomaly Detection in Wireless Sensor Networks
    (North Dakota State University, 2013) Zhao, Jingjun
    Wireless Sensor Networks (WSNs) have been applied to a wide range of application areas, including battle fields, transportation systems, and hospitals. The security issues in WSNs are still hot research topics. The constrained capabilities of sensors and the environments in which sensors are deployed, such as hostile and non-reachable areas, make the security more complicated. This dissertation describes the development and testing of a novel two-phase security mechanism for hierarchical WSNs that is capable of defending both outside and inside attacks. For the outside attacks, the attackers are usually malicious intruders that entered the network. The computation and communication capabilities of the sensors restrict them from directly defending the harmful intruders by performing traditionally encryption, authentication, or other cryptographic operations. However, the sensors can assist the more powerful nodes in a hierarchical structured WSN to track down these intruders and thereby prevent further damage. To fundamentally improve the security of a WSN, a multi-target tracking algorithm is developed to track the intruders. For the inside attacks, the attackers are compromised insiders. The intruders manipulate these insiders to indirectly attack other sensors. Therefore, detecting these malicious insiders in a timely manner is important to improve the security of a network. In this dissertation, we mainly focus on detecting the malicious insiders that try to break the normal communication among sensors, which creates holes in the WSN. As the malicious insiders attempt to break the communication by actively using HELLO flooding attack, we apply an immune-inspired algorithm called Dendritic Cell Algorithm (DCA) to detect this type of attack. If the malicious insiders adopt a subtle way to break the communication by dropping received packets, we implement another proposed technique, a short-and-safe routing (SSR) protocol to prevent this type of attack. The designed security mechanism can be applied to different sizes of both static and dynamic WSNs. We adopt a popular simulation tool, ns-2, and a numerical computing environment, MATLAB, to analyze and compare the computational complexities of the proposed security mechanism. Simulation results demonstrate effective performance of the developed corrective and preventive security mechanisms on detecting malicious nodes and tracking the intruders.