• International Journal of Computer Science & Network Security
• /
• v.22 no.7
• /
• pp.13-20
• /
• 2022

Basically Mobile Ad Hoc Network (MANET) is an autonomous system with the collection of mobile nodes, these nodes are connected to each other by using wireless networks. A mobile ad hoc network poses this quality which makes topology in dynamic manner. As this type of network is Ad Hoc in nature hence it doesn't have fixed infrastructure. If a node wishes to transfer data from source node to a sink node in the network, the data must be passed through intermediate nodes to reach the destination node, hence in this process data packet loss occurs in various MANET protocols. This research study gives a comparison of various Mobile Ad Hoc Network routing protocols like proactive (DSDV) and reactive (AODV, DSR) by using random topology with more intermediate nodes using CBR traffic. Our simulation used 50, 100, and 150 nodes variations to examine the performance of the MANET routing protocols. We compared the performance of DSDV, AODV and DSR, MANET routing protocols with the result of existing protocol using NS-2 environment, on the basis of different performance parameters like Packet Delivery Ratio, average throughput and average end to end delay. Finally we found that our results are better in terms of throughput and packet delivery ratio along with low data loss.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.5
• /
• pp.1755-1777
• /
• 2022

The development of wireless communication technology has led to the underutilization of radio spectra. To address this limitation, an intelligent cognitive radio network was developed. Specific emitter identification (SEI) is a key technology in this network. However, in realistic non-cooperative scenarios, the system may detect signal classes beyond those in the training database, and only a few labeled signal samples are available for network training, both of which deteriorate identification performance. To overcome these challenges, a meta-learning-based open-set identification system is proposed for SEI. First, the received signals were pre-processed using bi-spectral analysis and a Radon transform to obtain signal representation vectors, which were then fed into an open-set SEI network. This network consisted of a deep feature extractor and an intrinsic feature memorizer that can detect signals of unknown classes and classify signals of different known classes. The training loss functions and the procedures of the open-set SEI network were then designed for parameter optimization. Considering the few-shot problems of open-set SEI, meta-training loss functions and meta-training procedures that require only a few labeled signal samples were further developed for open-set SEI network training. The experimental results demonstrate that this approach outperforms other state-of-the-art SEI methods in open-set scenarios. In addition, excellent open-set SEI performance was achieved using at least 50 training signal samples, and effective operation in low signal-to-noise ratio (SNR) environments was demonstrated.

• Journal of the Korea Society of Computer and Information
• /
• v.29 no.2
• /
• pp.1-12
• /
• 2024

Predicting path loss is one of the important factors for wireless network design, such as selecting the installation location of base stations in cellular networks. In the past, path loss values were measured through numerous field tests to determine the optimal installation location of the base station, which has the disadvantage of taking a lot of time to measure. To solve this problem, in this study, we propose a path loss prediction method based on machine learning (ML). In particular, an ensemble learning approach is applied to improve the path loss prediction performance. Bootstrap dataset was utilized to obtain models with different hyperparameter configurations, and the final model was built by ensembling these models. We evaluated and compared the performance of the proposed ensemble-based path loss prediction method with various ML-based methods using publicly available path loss datasets. The experimental results show that the proposed method outperforms the existing methods and can predict the path loss values accurately.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.26 no.12B
• /
• pp.1656-1664
• /
• 2001

When a packet loss occurs in a communication network operating a TCP protocol, the TCP protocol regards it that the loss has resulted from network congestion. Then the TCP protocol performs congestion control. When it is applied to the wireless network having quite a high BER characteristics, the performance of TCP protocol is degraded very much. In this paper, we propose an Explicit Error Notification(EEN) algorithm to improve the performance of the wireless TCP When a packet loss occurs in the wireless network, the TCP receiver decodes the TCP segment sequence number and the address of the TCP sender and receiver, and then informs the TCP sender of the error in wireless network by sending a NACK. It is to distinguish packets in error from losses of network congestion. In this paper, the performance of the proposed EEN algorithm is analyzed and simulated. In fact, as more errors are corrected, the proposed algorithm shows a larger improvements in performance.

• KIPS Transactions on Software and Data Engineering
• /
• v.8 no.3
• /
• pp.123-128
• /
• 2019

This paper proposes the stock price prediction based on the artificial intelligence, where the model with recurrent convolution neural network (RCNN) layers is adopted. In the motivation of this prediction, long short-term memory model (LSTM)-based neural network can make the output of the time series prediction. On the other hand, the convolution neural network provides the data filtering, averaging, and augmentation. By combining the advantages mentioned above, the proposed technique predicts the estimated stock price of next day. In addition, in order to emphasize the recent time series, a custom weighted loss function is adopted. Moreover, stock data related to the stock price index are adopted to consider the market trends. In the experiments, the proposed stock price prediction reduces the test error by 3.19%, which is over other techniques by about 19%.

• The Transactions of the Korea Information Processing Society
• /
• v.4 no.9
• /
• pp.2342-2353
• /
• 1997

According to the outstanding development of information industry, a study of firewall is progressing as one of methods to cope with threat and loss of the data through computer network. For the secure network, this paper proposes the method diminishing threat and loss of the network using the correlation firewall with network audit trail system. Also, this paper suggests not only the audit analyzer execution model but also the type of databases used in audit analyzer to analyze the audit data. Network audit trail system has the function of identifing and analyzing of all intruder actions using audit records created by users.

• The Journal of the Korea Contents Association
• /
• v.5 no.3
• /
• pp.201-208
• /
• 2005

To guarantee the quality of service for user request, we should consider various kinds of things. The important thing of QoS is that response time of service is transparently suggested 'to network users. We can know the response time of service using the information of network latency, system latency, and software component latency, In this paper, we carried out the modeling of network latency and analyzed the effects of packets loss to the network latency, Also, we showed the effectiveness of modeling using the NS-2. This research can help to provide the effective methods in case of SLA(Service Level Agreement) agreement between service provider and user.

• International Journal of Control, Automation, and Systems
• /
• v.6 no.3
• /
• pp.351-363
• /
• 2008

This paper is concerned with tasking and guidance of networked airborne sensors to achieve fault-tolerant sensing. The sensors are coordinated to locate hostile transmitters by intercepting and processing their signals. Faults occur when some sensor-carrying vehicles engaged in target location missions are lost. Faults effectively change the network architecture and therefore degrade the network performance. The first objective of the paper is to optimally allocate a finite number of sensors to targets to maximize the network life and availability. To that end allocation policies are solved from relevant Markov decision problems. The sensors allocated to a target must continue to adjust their trajectories until the estimate of the target location reaches a prescribed accuracy. The second objective of the paper is to establish a criterion for vehicle guidance for which fault-tolerant sensing is achieved by incorporating the knowledge of vehicle loss probability, and by allowing network reconfiguration in the event of loss of vehicles. Superior sensing performance in terms of location accuracy is demonstrated under the established criterion.

• Journal of Internet Computing and Services
• /
• v.21 no.5
• /
• pp.21-29
• /
• 2020

Accurate estimation of human pose relies on backbone method in which its role is to extract feature map. Up to dated, the method of backbone feature extraction is conducted by the plain convolutional neural networks named by CNN and the residual neural networks named by Resnet, both of which have various architectures and performances. The CNN family network such as VGG which is well-known as a multiple stacked hidden layers architecture of deep learning methods, is base and simple while Resnet which is a bottleneck layers architecture yields fewer parameters and outperform. They have achieved inspired results as a backbone network in human pose estimation. However, they were used then followed by different pose estimation networks named by pose parsing module. Therefore, in this paper, we present a comparison between the plain CNN family network (VGG) and bottleneck network (Resnet) as a backbone method in the same pose parsing module. We investigate their performances such as number of parameters, loss score, precision and recall. We experiment them in the bottom-up method of human pose estimation system by adapted the pose parsing module of openpose. Our experimental results show that the backbone method using VGG network outperforms the Resent network with fewer parameter, lower loss score and higher accuracy of precision and recall.

• The Transactions of the Korean Institute of Electrical Engineers A
• /
• v.50 no.2
• /
• pp.76-82
• /
• 2001

This paper presents a approach for a service restoration and optimal reconfiguration of distribution network using Genetic algorithm(GA) and Tabu search(TS) method. Restoration and reconfiguration problems in distribution network are difficult to solve in short times, because distribution network supplies power for customers combined with many tie-line switches and sectionalizing switches. Furthermore, the solutions of these problems have to satisfy radial operation conditions and reliability indices. To overcome these time consuming and sub-optimal problem characteristics, this paper applied Genetic-Tabu algorithm. The Genetic-Tabu algorithm is a Tabu search combined with Genetic algorithm to complement the weak points of each algorithm. The case studies with 7 bus distribution network showed that not the loss reduction but also the reliability cost should be considered to achieve the economic service restoration and reconfiguration in the distribution network. The results of suggested Genetic-Tabu algorithm and simple Genetic algorithm are compared in the case study also.