Alternative Initial Probability Tables for Elicitation of Bayesian Belief Networks

Bayesian Belief Networks are used in many fields of application. Defining the conditional dependencies via conditional probability tables requires the elicitation of expert belief to fill these tables, which grow very large quickly. In this work, we propose two methods to prepare these tables based on a low number of input parameters using specific structures and one method to generate the table using probability tables of each relation of a child node with a certain parent. These tables can be used further as a starting point for elicitation.

Sinkhole Attack Detection and Avoidance Mechanism for RPL in Wireless Sensor Networks

The security issue is one of the main problems in Wireless Sensor Network (WSN) and Internet of Things (IoTs). RPL (Routing protocol for low power and lossy networks) is a standard routing protocol for WSN, is not to be missed from being attacks. The performance of RPL is reduced significantly after being attacked. Sinkhole attack is one of the most common attacks to WSN and RPL, threatening the network capability by discarding packets and disrupting routing paths. Therefore, this paper proposes a new Secured-RPL routing protocol to detect and avoid sinkhole attacks in the network, which is called Cross Layers Secured RPL (CLS-RPL). This routing protocol is enhanced of the existing RPL routing protocol. CLS-RPL is a cross-layer routing protocol that uses information from the data link layer in its security mechanism. CLS-RPL uses a new technique and concept in detecting a sinkhole attack that is based on eave-listening (overhearing) that allows a child node to eave-listening its parent transmission. If the child node does not hear any transmission from its parent node after sending several packets, this means its parent node is a sinkhole attacker. Otherwise, if the node hears transmission from its parent node, this means that its parent node is legitimate and continues to send more packets. CLS-RPL implements a simple security mechanism that provides a high packet delivery ratio. The finding shows that CLS-RPL provides 52% improvement in terms of packet delivery ratio when compared to RPL protocol.

Taxonomy Proposal for Research on Reversible Logic

Reversible Logic has become a topic of interest among global researchers for its heat arresting attribute, the very reason for which CMOS devices are witnessing performance thresholds. Within a decade of substantial growth, reversible logic is quoted as one of the valued emerging technologies. The growth of research in this domain has exhibited enormous volumes of research publications. Searching the perfect published article of interest among the plethora of publications demand an enormous amount of literature study and hence is highly time-consuming. After accessing a substantial amount of articles, only a very few tend to address our needs. Hence, we propose a taxonomy for research on reversible logic in this paper which concentrates articles of relevant/similar nature into nodes so that researchers can get a first-hand reference for their choice of interest. Knowledge of pre-requisites for gaining insights into a particular domain also mandates a thorough amount of literature survey. Our taxonomy takes care of those issues by concentrating the nodes into a tree-like structure thereby providing the pre-requisite (parent node) for the node of interest (child node). For generating the taxonomy, we have used the Cosine Similarity function for measuring the distance between two author keywords. The taxonomy proposal in this paper will serve upcoming researchers in getting references for their choice of the domain rather than a comprehensive survey of the higher magnitude of published data. Researchers pursuing research in a certain domain too will benefit from the taxonomy.

Inter-node Hellinger Distance based Decision Tree

This paper introduces a new splitting criterion called Inter-node Hellinger Distance (iHD) and a weighted version of it (iHDw) for constructing decision trees. iHD measures the distance between the parent and each of the child nodes in a split using Hellinger distance. We prove that this ensures the mutual exclusiveness between the child nodes. The weight term in iHDw is concerned with the purity of individual child node considering the class imbalance problem. The combination of the distance and weight term in iHDw thus favors a partition where child nodes are purer and mutually exclusive, and skew insensitive. We perform an experiment over twenty balanced and twenty imbalanced datasets. The results show that decision trees based on iHD win against six other state-of-the-art methods on at least 14 balanced and 10 imbalanced datasets. We also observe that adding the weight to iHD improves the performance of decision trees on imbalanced datasets. Moreover, according to the result of the Friedman test, this improvement is statistically significant compared to other methods.

One Core Task of Interpretability in Machine Learning — Expansion of Structural Equation Modeling

Structural equation modeling (SEM) is a system of two kinds of equations: a linear latent structural model (SM) and a linear measurement model (MM). The latent structure model is a causal model from the latent parent node to the latent child node. Meanwhile, MM’s link is from latent variable parent node to observed variable child node. However, researchers should determine the initial causal order between variables based on experience when applying SEM. The main reason is that SEM does not fully construct causal models between observed variables (OVs) from big data. When the artificial causal order is contrary to the fact, the causal inference from SEM is doubtful, and the implicit causal information between the OVs cannot be extracted and utilized. This study first objectively identifies the causal order of variables using the DirectLiNGAM method widely accepted in recent years. Then traditional SEM is converted to expanded SEM (ESEM) consisting of SM, MM and observation model (OM). Finally, through model testing and debugging, ESEM with good fit with data is obtained.

Design and Development of Network Reliability based Secure Multicast Routing Protocol for MANET

In Mobile Ad hoc network (MANET), link quality and stability of links as well as nodes play a major role. In ad hoc network, links are often changing which could affect the node mobility and integrity of data packets. In this research work, Network Reliability based Secure Multicast Routing Protocol (NRSMRP) is proposed to achieve network reliability by means of creation of reliable multicast tree. This multicast tree is constructed based on link quality and reliability trust metric. In first phase, node categorization and reliability metric calculation are implemented with the help of link quality. In second phase, reliable multicast tree is formed based on parent node and child node. Parent node must have god capacity and signal strength to communicate with child node. In last phase, authentication based multicast routes are established based on the calculation of direct reputation of mobile nodes. From the results, proposed protocol achieves better performance than existing schemes.

An Advanced AOI-Cast Algorithm Based on PCA

Voronoi-based Overlay Network (VON) has been proposed that promises to maintain high overlay topology consistency in a bandwidth-efficient manner. VoroCast constructs a spanning tree across all AOI neighbors based on Voronoi diagrams, while FiboCast dynamically adjusts the messaging range by a Fibonacci sequence. VoroCast improves the AOI scalability of P2P-based NVEs. However, one potential drawback of the schemes is that the child node degrees are only based on peers' positions, but not node capacities. Since each node in VoroCast has different capacity, packet loss will be unavoidable. VoroCast may lead to lower multicast efficiency in AOI. To these problems, an Advanced AOI-cast algorithm based on PCA is presented. In the algorithm, node capacity is related to node CPU (c), node bandwidth (b) and node memory (m), and the node capacity is calculated according to the PCA. An undirected graph is formed through all the nodes in the AOI and the edge weights are calculated by the Gaussian function. Through the prim algorithm, to generate the minimum spanning tree of the weighted undirected graph, and the minimum spanning tree is used as the final multicast tree. The message is delivered through the multicast tree. The simulation results show that the algorithm gains a greater improvement on multicast efficiency, and achieves better scalability.

Design and Realization of a Wireless Data Acquisition System Based on Multi-Nodes and Multi-Base-Stations

Multi-nodes inertial data acquisition has the widespread applications in medical rehabilitation research and biomechanical analysis. In the traditional method, the inertial data of every child-node is sent to a center node by wired line firstly. Then the center node transfers the data of all child-nodes to PC through a single serial port. This approach has some obvious drawbacks, including the fact that the sampling rate is low, the communication protocols are complex and installation is inconvenient. This article proposed a multi-node and multi-base-station wireless data acquisition system. The communication between child-node and the base-station is performed in a point-to-point manner. The data of micro inertial sensors on the child-node circuit board is collected by the MCU then is sent to the matching base-station by the RF module. The base-station send data packet to the upper computer through serial port. The results showed that the upper computer software can collect data packets synchronously from three child-nodes (the data packet of each child-node contain 19 Bytes) with high-speed sampling rates up to 118 Hz. And there is no obvious deferring time under this condition. The technique has the certain applications in the data collecting system of motion capture. Key words: Pcomm; multi-thread; multiple serial ports; nRF905; MPU6000