• Journal of the Korea Society for Simulation
• /
• v.14 no.4
• /
• pp.55-68
• /
• 2005

The advancement in semiconductor technology is leading toward smaller critical dimension designs and larger wafer manufactures. Due to such phenomena, semiconductor industry is in need of an accurate control of the process. Photolithography is one of the key processes where the pattern of each layer is formed. In this process, precise superposition of the current layer to the previous layer is critical. Therefore overlay parameters of the semiconductor photolithography process is targeted for this research. The complex relationship among the input parameters and the output metrologies is difficult to understand and harder yet to model. Because of the superiority in modeling multi-nonlinear relationships, neural networks is used for the simulator modeling. For training the neural networks, conjugate gradient method is employed. An experiment is performed to evaluate the performance among the proposed neural network simulator, stepwise regression model, and the currently practiced prediction model from the test site.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.8
• /
• pp.3602-3620
• /
• 2016

Dynamic topology is one of the main influence factors on network performability. However, it was always ignored by the traditional network performability assessment methods when analyzing large-scale mobile ad hoc networks (MANETs) because of the state explosion problem. In this paper, we address this problem from the perspective of complex network. A two-layer hierarchical modeling approach is proposed for MANETs performability assessment, which can take both the dynamic topology and multi-state nodes into consideration. The lower level is described by Markov reward chains (MRC) to capture the multiple states of the nodes. The upper level is modeled as a small-world network to capture the characteristic path length based on different mobility and propagation models. The hierarchical model can promote the MRC of nodes into a state matrix of the whole network, which can avoid the state explosion in large-scale networks assessment from the perspective of complex network. Through the contrast experiments with OPNET simulation based on specific cases, the method proposed in this paper shows satisfactory performance on accuracy and efficiency.

• Journal of Korea Multimedia Society
• /
• v.23 no.12
• /
• pp.1540-1551
• /
• 2020

Aiming at the problem that the existing human behavior recognition algorithm cannot fully utilize the multi-level spatio-temporal information of the network, a human behavior recognition algorithm based on a dense three-dimensional residual network is proposed. First, the proposed algorithm uses a dense block of three-dimensional residuals as the basic module of the network. The module extracts the hierarchical features of human behavior through densely connected convolutional layers; Secondly, the local feature aggregation adaptive method is used to learn the local dense features of human behavior; Then, the residual connection module is applied to promote the flow of feature information and reduced the difficulty of training; Finally, the multi-layer local feature extraction of the network is realized by cascading multiple three-dimensional residual dense blocks, and use the global feature aggregation adaptive method to learn the features of all network layers to realize human behavior recognition. A large number of experimental results on benchmark datasets KTH show that the recognition rate (top-l accuracy) of the proposed algorithm reaches 93.52%. Compared with the three-dimensional convolutional neural network (C3D) algorithm, it has improved by 3.93 percentage points. The proposed algorithm framework has good robustness and transfer learning ability, and can effectively handle a variety of video behavior recognition tasks.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.9
• /
• pp.3152-3171
• /
• 2022

This work investigates the physical layer secrecy of a multi-antenna hybrid satellite-terrestrial relay networks (HSTRN) with jamming, in which a satellite aims to make communication with a destination user by means of a relay, along with spatially random eavesdroppers. In order to weaken the signals of eavesdroppers, the conventional relay can also generate intentional interference, besides forwarding the received signal. Shadowed-Rician fading is adopted in satellite link, while Rayleigh fading is adopted in terrestrial link, eavesdropper link and jamming link. The analytical and asymptotic formulas for the system secrecy outage probability (SOP) are characterized. Practical insights on the diversity order of the network are revealed according to the asymptotic behavior of SOP at high signal-to-noise ratio (SNR) regime. Then, analysis of the system throughput is examined to assess the secrecy performance. In the end, numerical simulation results are presented to validate the theoretical analysis and point out: (1) The secrecy performance of the considered network is affected by the channel fading scenario, the system configuration; (2) Decrease of the relay coverage airspace can provide better SOP performance; (3) Jamming from the relay can improve secrecy performance without additional network resources.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.4 no.3
• /
• pp.258-272
• /
• 2010

The low-rate (LR) UWB is a promising technology for the ubiquitous sensor network (USN) due to its extremely low power consumption and simple transceiver implementation. However the limited communication range is a bottleneck for its widespread use. This paper deals with a new frame structure of class 4 active RFID multi-hop relay system based on ISO/IEC 18000-7 standard integrating with IEEE 802.15.4a LR-UWB PHY layer specification, which sets up a connection to USN. As a result of the vital importance of the coverage and throughput in the application of USN, further we analyze the performance of the proposed system considered both impulse radio UWB (IR-UWB) and chirp spread spectrum (CSS). Our simulation results show that the coverage and throughput are remarkably increased.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2000.06a
• /
• pp.286-290
• /
• 2000

Optimal neuro-control algorithm is extended to the control of a multi-degree-of-freedom structure. An active mass driver(AMD) system on the top roof is used as an exciter. The control signals are made by a multi-layer perceptron(MLP) which is trained by minimizing a sub-optimal performance index. The performance index is a function of both the output responses and the control signals. Structure having nonlinear hysteretic behavior is also trained and controlled by using proposed control algorithm. In training neuro-controller, emulator neural network is not used. Instead, sensitivity-test data are used. Therefore, only one neural network is used for the control system. Both the time delay effect and the dynamics of hydraulic actuator are included in the simulation. Example shows that optimal neuro-control algorithm can be applicable to the multi-degree of freedom structures.

• Proceedings of the Acoustical Society of Korea Conference
• /
• 1993.06a
• /
• pp.117-120
• /
• 1993

In this paper, we explore the possibility of URAN(Universally Reconstructable Artificial Neural-network) VLSI chip for speech recognition. URAN, a newly developed analog-digital hybrid neural chip, is discussed in respects to its input, output, and weight accuracy and their relations to its performance on speaker independent digit recognition. Multi-layer perceptron(MLP) nets including a large frame input layer are used to recognize a digit syllable at a forward retrieval. The simulation results using the full and limited floating precision computations for the input, output, and weight variables of the network give the comparable classification performance. An MLP with piecewise linear hidden and output units is also trained successfully using low accuracy computation.

• International Journal of Concrete Structures and Materials
• /
• v.1 no.1
• /
• pp.63-73
• /
• 2007

Optimum multi-layered feed-forward neural network (NN) models using a resilient back-propagation algorithm and early stopping technique are built to predict the shear capacity of reinforced concrete deep and slender beams. The input layer neurons represent geometrical and material properties of reinforced concrete beams and the output layer produces the beam shear capacity. Training, validation and testing of the developed neural network have been achieved using 50%, 25%, and 25%, respectively, of a comprehensive database compiled from 631 deep and 549 slender beam specimens. The predictions obtained from the developed neural network models are in much better agreement with test results than those determined from shear provisions of different codes, such as KBCS, ACI 318-05, and EC2. The mean and standard deviation of the ratio between predicted using the neural network models and measured shear capacities are 1.02 and 0.18, respectively, for deep beams, and 1.04 and 0.17, respectively, for slender beams. In addition, the influence of different parameters on the shear capacity of reinforced concrete beams predicted by the developed neural network shows consistent agreement with those experimentally observed.

• Journal of the Korea Society for Simulation
• /
• v.12 no.4
• /
• pp.73-81
• /
• 2003

An Ad Hoc network is a dynamic multi-hop wireless network that is established by a group of mobile hosts on a shared wireless channel by virtue of their proximity to each other. Since wireless transmissions are locally broadcast in the region of the transmitting host, hosts that are in close proximity can hear each other and are said to be neighbors. The transitive closure of the neighborhood of all the hosts in the set of mobile hosts under consideration forms an Ad Hoc network. Thus, each host is potentially a router and it is possible to dynamically establish routes by chaining together a sequence of neighboring hosts from a source to a destination in the Ad Hoc network. In a network, various real-time services require the network to guarantee the Quality of Services provided to the receiver. End-to-end QoS can be provided most efficiently when each layer of the protocol stack translates the requirements of the application into layer classified requirements and satisfies them. In this study, a mechanism to guarantee the QoS in Ad Hoc networks with buffer nodes is proposed. They effectively prevent traffic congestion and yield better transmission rate. In this way QoS is enhanced.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.5 no.4
• /
• pp.185-194
• /
• 2010

In this paper, we propose an application layer-based vertical handover management protocol, called multihoming RTP (mhRTP), for real-time applications with seamless mobility across heterogeneous wireless access networks. The proposed multi-homing RTP provides a soft handover by utilizing multiple available wireless access network interfaces simultaneously. The newly available path is dynamically added to the ongoing session by the mhRTP session manager. Also the decision making of QoS-improving or QoS-guaranteed handover is possible based on the estimation of available bandwidth in each candidate network. The performances of the proposed mhRTP have been analyzed through a series of simulations on OPNET network simulator. From the performance analysis, we confirmed that the proposed mhRTP can provide QoS-guaranteed vertical handover with efficient session managements.