• The Journal of the Acoustical Society of Korea
• /
• v.42 no.4
• /
• pp.357-363
• /
• 2023

In underwater signal processing, separating individual signals from mixed signals has long been a challenge due to low signal quality. The common method using Short-time Fourier transform for spectrogram analysis has faced criticism for its complex parameter optimization and loss of phase data. We propose a Triple-path Recurrent Neural Network, based on the Dual-path Recurrent Neural Network's success in long time series signal processing, to handle three-dimensional tensors from multi-channel sensor input signals. By dividing input signals into short chunks and creating a 3D tensor, the method accounts for relationships within and between chunks and channels, enabling local and global feature learning. The proposed technique demonstrates improved Root Mean Square Error and Scale Invariant Signal to Noise Ratio compared to the existing method.

• Korean Membrane Journal
• /
• v.2 no.1
• /
• pp.36-47
• /
• 2000

This paper deals with the separation of MTBE-methanol mixtures using crosslinked Poly(vinyl alcohol)(PVA) membranes with sulfur-succinic acid(SSA) as a crosslinking agent by pervaporation and vapor permeation technique. The operating temperatures, methanol concentration in feed mixtures, and SSA concentrations in PVA membranes were varied to investigate the separation performance of PVA/SSA membranes and the optimum separation characteristics by pervaporation and vapor permeation. And also, for PVA/SSA membranes, the swelling measurements were carried out to study the transport phenomena. The swelling measurements were carried out for pure MTBE and methanol, and MTBE/methanol=90/10, 80/20 mixtures using PVA/SSA membranes with varying SSA compositions. There are two factors of the membrane network and the hydrogen bonding. In pervaporation separation was also carried out for MTBE/methanol=90/10, 80/20 mixtures at various temperatures. The sulfuric acid group in SSA took an important role in the membrane performance. The crosslinking effect might be over the hydrogen bonding effect due to the sulfuric acid group at 3 and 5% SSA membranes, and this two factors act vice versa on 7% SSA membrane. In this case, the 5% SSA membrane shows the highest separation factor of 2,095 with the flux of 12.79g/㎡$.$hr for MTBE/methanol=80/20 mixtures at 30$^{\circ}C$ which this mixtures show near the azeotopic composition. Compared to pervaporation, vapor permeation showed less flux and similar separation factor. In this case, the flux decreased significantly because of compact structure and the effect of hydrogen bonding. In vapor permeation, density or concentration of methanol in vaporous feed is lower than that of methanol in liquid feed, as a result, the hydrogen bonding portion between the solvent and the hydroxyl group in PVA is reduced in vapor permeation. In this case, the 7% SSA membranes shows the highest separation factor of 2,187 with the flux of 4.84g/㎡$.$hr for MTBE/methanol=80/20 mixtures at 30$^{\circ}C$.

• Membrane Journal
• /
• v.2 no.1
• /
• pp.36-36
• /
• 1992

This paper deals with the separation of MTBE-methanol mixtures using crosslinked Poly(vinyl alcohol)(PVA) membranes with sulfur-succinic acid(SSA) as a crosslinking agent by pervaporation and vapor permeation technique. The operating temperatures, methanol concentration in feed mixtures, and SSA concentrations in PVA membranes were varied to investigate the separation performance of PVA/SSA membranes and the optimum separation characteristics by pervaporation and vapor permeation. And also, for PVA/SSA membranes, the swelling measurements were carried out to study the transport phenomena. The swelling measurements were carried out for pure MTBE and methanol, and MTBE/methanol=90/10, 80/20 mixtures using PVA/SSA membranes with varying SSA compositions. There are two factors of the membrane network and the hydrogen bonding. In pervaporation separation was also carried out for MTBE/methanol=90/10, 80/20 mixtures at various temperatures. The sulfuric acid group in SSA took an important role in the membrane performance. The crosslinking effect might be over the hydrogen bonding effect due to the sulfuric acid group at 3 and 5% SSA membranes, and this two factors act vice versa on 7% SSA membrane. In this case, the 5% SSA membrane shows the highest separation factor of 2,095 with the flux of 12.79g/㎡·hr for MTBE/methanol=80/20 mixtures at 30℃ which this mixtures show near the azeotopic composition. Compared to pervaporation, vapor permeation showed less flux and similar separation factor. In this case, the flux decreased significantly because of compact structure and the effect of hydrogen bonding. In vapor permeation, density or concentration of methanol in vaporous feed is lower than that of methanol in liquid feed, as a result, the hydrogen bonding portion between the solvent and the hydroxyl group in PVA is reduced in vapor permeation. In this case, the 7% SSA membranes shows the highest separation factor of 2,187 with the flux of 4.84g/㎡·hr for MTBE/methanol=80/20 mixtures at 30℃.

• Journal of Power System Engineering
• /
• v.4 no.4
• /
• pp.78-83
• /
• 2000

It is often necessary to observe the working environment of a robot to control it efficiently in the remote location. The remote sensing data and control commands are transmitted via various media such as radio, microwave, and computer network. The World Wide Web can be used as the infrastructure for teleoperation of mobile robots. In this paper we propose an advanced technique of the remote control of mobile robots on the web. For the real-time control, the image separation algorithm is proposed to transmit the current positions of mobile robots instead of transmitting the full frame image. It is shown by experiments that the proposed algorithm is effective.

• Journal of KIISE:Computing Practices and Letters
• /
• v.10 no.1
• /
• pp.89-98
• /
• 2004

Conversational agent is a system that provides users with proper information and maintains the context of dialogue on the natural language. Analyzing and modeling process of user's query is essential to make it more realistic, for which Bayesian network is a promising technique. When experts design the network for a domain, the network is usually very complicated and is hard to be understood. The separation of variables in the domain reduces the size of networks and makes it easy to design the conversational agent. Composing Bayesian network as two stages, we aim to design conversational agent easily and analyze user's query in detail. Also, previous information of dialogue makes it possible to maintain the context of conversation. Actually implementing it for a guide of web pages, we can confirm the usefulness of the proposed architecture for conversational agent.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.5 no.1
• /
• pp.157-176
• /
• 2011

An interest-based P2P constructs the peer connections based on similarities for efficient search of resources. A clustering technique using peer similarities as data is an effective approach to group the most relevant peers. However, the separation of groups produced from clustering lowers the scalability of a P2P network. Moreover, the interest-based approach is only concerned with user-level grouping where topology-awareness on the physical network is not considered. This paper proposes an efficient scalable search for the interest-based P2P system. A scalable multi-ring (SMR) based on fuzzy clustering handles the grouping of relevant peers and the proposed scalable search utilizes the SMR for scalability of peer queries. In forming the multi-ring, a minimized route function is used to determine the shortest route to connect peers on the physical network. Performance evaluation showed that the SMR acquired an accurate peer grouping and improved the connectivity rate of the P2P network. Also, the proposed scalable search was efficient in finding more replicated files throughout the peer network compared to other traditional P2P approaches.

• Journal of the Korea Society for Simulation
• /
• v.11 no.2
• /
• pp.45-53
• /
• 2002

The main idea behind the Intelligent Networks(IN) concept is the separation of switching functionality from the service control, in order to meet various service requirements of subscribers and development of new services in time. In (N+1) type Intelligent Network with FEP-BEP framwork, each SCP-BEP system maintains its own subscibers' database respectively. In this case, DB updating operations at each SCP-BEP should be peformed concurrently such that DB updating method could affect the overall system performance. Moreover, it is not easy to predict the current system capacity to satisfy the future IN subscribers' service needs. In this paper, we discuss how much DB updating method affects the performance trends of (N+1) type Intelligent Network with FEP-BEP system by means of the simulation technique as the number of calls increase. The average turnaround time is used as a system performance measure.

• Journal of the Korea Society of Computer and Information
• /
• v.24 no.4
• /
• pp.169-176
• /
• 2019

Wireless sensor networks are used to monitor and control areas in a variety of military and civilian areas such as battlefield surveillance, intrusion detection, disaster recovery, biological detection, and environmental monitoring. Since the sensor nodes are randomly placed in the area of interest, separation of the sensor network area may occur due to environmental obstacles or a sensor may not exist in some areas. Also, in the situation where the sensor node is placed in a non-relocatable place, some node may exhaust energy or physical hole of the sensor node may cause coverage hole. Coverage holes can affect the performance of the entire sensor network, such as reducing data reliability, changing network topologies, disconnecting data links, and degrading transmission load. It is possible to solve the problem that occurs in the coverage hole by finding a coverage hole in the sensor network and further arranging a new sensor node in the detected coverage hole. The existing coverage hole detection technique is based on the location of the sensor node, but it is inefficient to mount the GPS on the sensor node having limited resources, and performing other location information processing causes a lot of message transmission overhead. In this paper, we propose an Adjacent Matrix-based Hole Coverage Discovery(AMHCD) scheme based on connectivity of neighboring nodes. The method searches for whether the connectivity of the neighboring nodes constitutes a closed shape based on the adjacent matrix, and determines whether the node is an internal node or a boundary node. Therefore, the message overhead for the location information strokes does not occur and can be applied irrespective of the position information error.

• International Journal of Computer Science & Network Security
• /
• v.24 no.5
• /
• pp.119-128
• /
• 2024

The moveable ad hoc networks are untrustworthy and susceptible to any intrusion because of their wireless interaction approach. Therefore the information from these networks can be stolen very easily just by introducing the attacker nodes in the system. The straight route extent is calculated with the help of hop count metric. For this purpose, routing protocols are planned. From a number of attacks, the wormhole attack is considered to be the hazardous one. This intrusion is commenced with the help of couple attacker nodes. These nodes make a channel by placing some sensor nodes between transmitter and receiver. The accessible system regards the wormhole intrusions in the absence of intermediary sensor nodes amid target. This mechanism is significant for the areas where the route distance amid transmitter and receiver is two hops merely. This mechanism is not suitable for those scenarios where multi hops are presented amid transmitter and receiver. In the projected study, a new technique is implemented for the recognition and separation of attacker sensor nodes from the network. The wormhole intrusions are triggered with the help of these attacker nodes in the network. The projected scheme is utilized in NS2 and it is depicted by the reproduction outcomes that the projected scheme shows better performance in comparison with existing approaches.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.17 no.11 s.114
• /
• pp.1058-1064
• /
• 2006

In this paper, the channel capacity of a specified wireless indoor multiple-input multiple-output(MIMO) channel is estimated by analyzing spatial characteristics of this channel using the three-dimensional ray tracing method, and a technique for deriving an optimized separation of multi-antenna elements is proposed. At first, the ray paths, the path losses, and the time-delay profile are computed using the three-dimensional ray tracing method in an indoor corridor environment, which has the line of sight(LOS) and non-line of sight(NLOS) regions. The ray tracing method is verified by a comparison between the computation results and the measurements which are obtained with dipole antennas, an amplifier and a network analyzer. Then, an MIMO system is positioned in the indoor channel environment and the ray paths and path losses are computed for four antenna-position combinations and various values of the antenna separation to obtain the channel capacity for the MIMO system. An optimum antenna-separation is derived by averaging the channel capacities of 100 receiver positions with four different antenna combinations.