• Journal of Information Processing Systems
• /
• v.18 no.3
• /
• pp.443-456
• /
• 2022

An efficient and reasonable resource allocation strategy can greatly improve the service quality of Internet of Vehicles (IoV). However, most of the current allocation methods have overestimation problem, and it is difficult to provide high-performance IoV network services. To solve this problem, this paper proposes a network resource allocation strategy based on deep learning network model DDQN. Firstly, the method implements the refined modeling of IoV model, including communication model, user layer computing model, edge layer offloading model, mobile model, etc., similar to the actual complex IoV application scenario. Then, the DDQN network model is used to calculate and solve the mathematical model of resource allocation. By decoupling the selection of target Q value action and the calculation of target Q value, the phenomenon of overestimation is avoided. It can provide higher-quality network services and ensure superior computing and processing performance in actual complex scenarios. Finally, simulation results show that the proposed method can maintain the network delay within 65 ms and show excellent network performance in high concurrency and complex scenes with task data volume of 500 kbits.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.11 no.12
• /
• pp.5927-5942
• /
• 2017

Many real networks exhibit overlapping community structures. Recent studies have been performed that analyze the impact of overlapping community structure on information propagation, but few of them concerned with individual behaviors. From this point of view, we propose a Markov process model to evaluate the performance of information propagation in social networks with overlapping community structures. In addition, many individual social behaviors are combined in the model. For example, individuals may exhibit selfish behaviors, such as individual and social selfishness, and people may discard the information after they have used it. The accuracy of the model is verified by simulation. Furthermore, the numerical results show that both overlapping community structure of the network and individual behaviors have a significant impact on the outbreak size and propagation speed of the information. Additionally, the overlapping community structure of the social network can reduce the impact of selfishness on information propagation.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.32B no.5
• /
• pp.765-776
• /
• 1995

This paper describes a speaker adaptation method to improve the recognition performance of MLP(multiLayer Perceptron) based HMM(Hidden Markov Model) speech recognizer. In this method, we use lst-order linear transformation network to fit data of a new speaker to the MLP. Transformation parameters are adjusted by back-propagating classification error to the transformation network while leaving the MLP classifier fixed. The recognition system is based on semicontinuous HMM's which use the MLP as a fuzzy vector quantizer. The experimental results show that rapid speaker adaptation resulting in high recognition performance can be accomplished by this method. Namely, for supervised adaptation, the error rate is signifecantly reduced from 9.2% for the baseline system to 5.6% after speaker adaptation. And for unsupervised adaptation, the error rate is reduced to 5.1%, without any information from new speakers.

• Proceedings of the IEEK Conference
• /
• summer
• /
• pp.57-60
• /
• 2004

Next generation wireless network is evolving toward IP-based network that can various provide multimedia services. A challenge in wireless mobile Internet is support of quality of service over wireless access networks. DiffServ architecture is proposed for evolving wireless mobile Internet. In this paper we propose an algorithm for optimal buffer partitioning which requires the minimal channel capacity to satisfy the QoS requirements of input traffic. We used a partitioned buffer with size B to serve a layered traffic at each DiffServ router. We consider a traffic model with a single source generates traffic having J $(J\geq2)$ quality of service (QoS) classes. QoS in this case is described by loss probability $\varepsilon_j$. for QoS class j. Traffic is admitted or rejected based on the buffer occupancy and its service class. Traffic is generated by heterogeneous Markov-modulated fluid source (MMFS).

• International Journal of Computer Science & Network Security
• /
• v.24 no.8
• /
• pp.139-144
• /
• 2024

VANETs have become one of the most attractive research areas in the world of wireless networks in recent years. Indeed, vehicular networks have become capable of optimizing road traffic, which significantly reduces the number of accidents through notifications exchanged between nearby vehicles. The routing function based on the opportunistic algorithm is a critical part of the vehicle's communication system and will therefore be an ideal target for attacks that could aim to prevent alert messages from reaching their destination, and thus endanger human lives. The black hole attack is a major threat to the security of VANETs. The main idea of this paper focuses on the analysis of this type of attack in VANETs using Discrete-Time Markov Chains (DTMC). and deduce at the end the effect of the number of malicious nodes on the delivery rate in the network.

• Journal of Internet Technology
• /
• v.20 no.1
• /
• pp.301-313
• /
• 2019

In this paper, we propose a new MAC protocol for Cognitive Radio (CR)-based IoT networks, called MSMA/CA. We extend the standard CSMA/CA, adopted in IEEE 802.11 WLANs, to the CR networks with the minimal modification since it works well in the real world. We resolve the classical hidden/exposed terminal problems by a variant of RTS/CTS mechanism and further, the hidden primary terminal problem by the mutual spectrum sensing at the transmitter and the receiver. We also modify the backoff process of CSMA/CA to incorporate the blocking of secondary transmitters, with the aim of protecting ongoing primary transmissions from aggressive secondary users. We analyze the throughput and delay of our proposed scheme using the Markov chain model on the backoff procedure, and verify its accuracy by simulations. Simulation results demonstrate that our protocol is suitable for IoT networks since the performance is insensitive to the number of users or devices.

• The Journal of the Acoustical Society of Korea
• /
• v.38 no.6
• /
• pp.703-709
• /
• 2019

This paper proposes speech recognition systems employing Convolutional Neural Network (CNN) and Long Short Term Memory (LSTM) structures combined with Hidden Markov Moldel (HMM) to effectively recognize the speech of VeloPharyngeal Insufficiency (VPI) patients, and compares the recognition performance of the systems to the Gaussian Mixture Model (GMM-HMM) and fully-connected Deep Neural Network (DNNHMM) based speech recognition systems. In this paper, the initial model is trained using normal speakers' speech and simulated VPI speech is used for generating a prior model for speaker adaptation. For VPI speaker adaptation, selected layers are trained in the CNN-HMM based model, and dropout regulatory technique is applied in the LSTM-HMM based model, showing 3.68 % improvement in recognition accuracy. The experimental results demonstrate that the proposed LSTM-HMM-based speech recognition system is effective for VPI speech with small-sized speech data, compared to conventional GMM-HMM and fully-connected DNN-HMM system.

• Proceedings of the Korean Information Science Society Conference
• /
• 2004.10c
• /
• pp.526-528
• /
• 2004

고속 네트워크와 이질적인 자원의 결합으로 구성되어 보안에 취약할 수 밖에 없는 네트워크 기반 컴퓨팅 환경을 대상으로 내.외부적 공격이나 결함이 발생하더라도 중요한 서비스를 지속적으로 제공하여, 시스템의 피해를 최소화하는 침입감내시스템(Intrusion Tolerant Systems)에 대한 연구가 활발하다. 본 논문에서는 주-백업 구조를 갖는 침입감내시스템 구조를 제안하였으며, 마코브 분석(Markov Analysis)을 통해, 시스템 생존성을 정량적으로 정의하였다.

• Korean Management Science Review
• /
• v.15 no.1
• /
• pp.63-76
• /
• 1998

Reliability evaluation methodologies of the multipath MIN system are reviewed and critically compared. Some guidelines are proposed to select efficient evaluation method for the system designers to use. Considering the switch failure and repair characteristics of the MIN system, three types of Markov models are proposed for the MIN system availability models. These models can be used for the MIN performance analysis. The performance of the MIN system are supposed to vary according to the failure state of the system.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.28 no.1
• /
• pp.37-49
• /
• 2003

Using the concept of closed queueing network, we present a consistent way of interpreting existing duality relations among queueing systems. Also, using embedded Markov chains, we present a few new duality relations for the queueing systems with negative customers.