Applied Mechanics and Materials Vols. 602-605

Abstract: In this paper, a trustworthy QoS unicast routing scheme based on trustworthy network is proposed. First, a cognitive network model is designed, referring to the intelligence of artificial immune system. Then, trustworthiness of each node is evaluated by deploying sliding window mechanism and analyzing the behavior record of each node. Finally, a population adaptive based immune of trustworthy QoS unicast routing algorithm is proposed, referring to the adaptive process of antibody identifying antigen. We verify the effectiveness of our scheme by using large-scale computer emulation experiments and compare it to existing classical algorithms. The results show that the proposed algorithm is both feasible and effective for providing users with better QoS, etc.

Abstract: The block-by-block method extended by Kumar and Agrawal to fractional differential equations. Cao et al. proposed a high order schema which is based on an improved block-by-block approach, which consists in finding 4 unknowns simultaneously at each step block through solving a 4 × 4 system. We rigorously analytically prove that this method is convergent with order for , and order 6 for .

Abstract: A mathematical model for the growth of gallium nitride in a vertical impinging metalorganic chemical vapor deposition (MOCVD) reactor is developed. The dependence of the GaN film and the uniformity of the deposited layers on the inlet flow is investigated. Based on the simulation results, the uniformity of the precursor distribution is better, with the increasment of the inlet flow, so the quality of GaN film is superior.

Abstract: In this paper, the data mining technology and the mining process was explained; and several common methods of data mining were described. Based on the characteristics of the target value, application of text classification and textual association in the target value mining were discussed, and the process model of data mining concerning target value was also expressed.

Abstract: With the increasingly salient social influence of virtual practice, personalized lifelong learning concept could be realized through network platform. The new learning environment and platform generate the new evaluation system and model. The working experience and features of network learning of the meteorological technical personnel determine the difference between network evaluation model and the traditional summative evaluation system.. The nonlinearity between input vector and output vector also determines the neural network learning evaluation model orientation. The training and simulation show that the network learning evaluation achieves its best effect when the hidden node is 4.

Abstract: This paper introduced points and directed line segment relation judgment method, the characteristics of generation and Graham method using the original convex hull generation algorithm of convex hull discrete points of the convex hull, an improved algorithm for planar discrete point set is proposed. The main idea is to use quadrilateral to divide planar discrete point set into five blocks, and then by judgment in addition to the four district quadrilateral internally within the point is in a convex edge. The result shows that the method is relatively simple program, high computational efficiency.

Abstract: Ad Hoc network, a self-organization network system, can establish a communication to a plurality of mobile users without any infrastructure. All nodes in this type of network system exist via portable sources. The hot spot of Ad Hoc network is to realize further energy consumption reduction. In the work, an improved protocol, by which energy consumption is reduced, is proposed based on AODV protocol; named NL-AODV, it choose the optimal routing by the capacity, the nodes energy and the hop. It is verified that the improved protocol can effectively prolong the original lifetime of the network and obviously improve the routing efficiency by simulations. Consequently, it is more suitable for dynamic topology changes in the network system at present.

Abstract: Medical image compression technology has great significance in the field of medical image engineering and clinical application. The current image compression technology is mainly for gray image, Few people study on medical image color compression algorithm. This paper presents a compression algorithm for color image based on DCT. First, the algorithm change the color medical image of RGB space to SAT space, this transformation ensures the medical image without distortion, and effectively reduce the mean image of each dimension of component. Then, the algorithm make the DCT transform for color image compression in the SAT space. Experimental shows that the improved algorithm in color medical image compression has achieved good results.

Abstract: A new approach to the generation of an initial point is proposed for discrete combined shape, which improves fully the local searching capability of discrete combined shape algorithm. Combined shape algorithm is embedded into genetic algorithm as a combined shape operator. Consequently a hybrid genetic algorithm for structural optimization with discrete variables is proposed. The constrained optimization problems were dealt with by adaptive annealing penalty factors and penalty function. The numerical results show that improved combined shape genetic algorithm for structural optimization with discrete variable problems has a faster convergence speed, which has advantages of local searching capability and globally searching capability of genetic algorithm. Improved combined shape genetic algorithm is an efficient optimal design method for engineering structure.

Abstract: Regarding the scheduling problem for real-time business over Bluetooth sensor networks, an adaptive scheduling algorithm for real-time business over Bluetooth sensor networks was proposed. By evaluating the current channel quality and data arrival rate, the best scheduling interval was computed for each link with real-time business. Master maintained a slot counter for each link and determined whether to schedule this link by comparing if the value of slot counter reached the best scheduling interval. The simulation results indicate that this algorithm has better bandwidth utilization, higher fairness and lower scheduling delay than traditional scheduling algorithms.