• Journal of Institute of Control, Robotics and Systems
• /
• v.9 no.11
• /
• pp.917-924
• /
• 2003

Recurrent neural networks have been successfully applied to communications channel equalization. Major disadvantages of gradient-based learning algorithms commonly employed to train recurrent neural networks are slow convergence rates and long training sequences required for satisfactory performance. In a high-speed communications system, fast convergence speed and short training symbols are essential. We propose decision feedback equalizers using a recurrent neural network trained with Kalman filtering algorithms. The main features of the proposed recurrent neural equalizers, utilizing extended Kalman filter (EKF) and unscented Kalman filter (UKF), are fast convergence rates and good performance using relatively short training symbols. Experimental results for two time-varying channels are presented to evaluate the performance of the proposed approaches over a conventional recurrent neural equalizer.

• Journal of The Institute of Information and Telecommunication Facilities Engineering
• /
• v.9 no.4
• /
• pp.131-140
• /
• 2010

Recently, mobile wireless data traffic has been dramatically increased due to not only the popularization of digital convergence devices including smart phone, Net-book, and Tablet PC, but also the vitalization of wireless Internet related eco-systems such as AppStore. In addition, it is expected that a tremendous increase in mobile data is caused by the release of unlimited mobile data plans (flat-fee). In order to deal with such mobile data traffic explosion, it is necessary that network operators should make efforts to offload wireless data traffic. This paper reviews the condition of mobile wireless data traffic in domestic and international telecommunication industry and looks for various action plans to overcome the difficulty of network operators.

• Information and Communications Magazine
• /
• v.20 no.11
• /
• pp.1550-1559
• /
• 2003

The convergence of fixed and wireless networks in data communication is providing the necessary driver for M2M commerce to take-off. The opportunities provided by M2M Commerce areonly limited by imagination. Automotive Fleet and Freight, Tolling, Water and Power Metering, Supply Chain Management including Asset Management, Remote Monitoring and Diagnostics, Energy Management and Access Control and Security are among the many M2M applications that are currently getting rolled out. ARC Group expects the worldwide solutions market to be worth in excess of US$ 100 billion by 2007. In addition, operator revenues worldwide from the transport of Telematics data alone will rise from US$ 3.5 billion in 2002 to US$ 78 billion by 2007. This paper discusses some of the lifestyle and business opportunities provided by M2M Commerce in the new ear of network convergence. It also provides some case studies to demonstrate the benefits of M2M Commerce across the supply chain. The key focus of the paper is on achieving enhanced lifestyle, cost reduction, improved profitability and enhanced customer relationship management through M2M Commerce.

• ETRI Journal
• /
• v.28 no.6
• /
• pp.815-818
• /
• 2006

In the global Internet, a constraint-based routing algorithm performs the function of selecting a routing path while satisfying some given constraints rather than selecting the shortest path based on physical topology. It is necessary for constraint-based routing to disseminate and update link state information. The triggering policy of link state updates significantly affects the volume of update traffic and the quality of services (QoS). In this letter, we propose an adaptive triggering policy based on link-usage statistics in order to reduce the volume of link state update traffic without deterioration of QoS. Also, we evaluate the performance of the proposed policy via simulations.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.8 no.11
• /
• pp.3837-3859
• /
• 2014

Vehicular Ad hoc Networks (VANET) becomes more popular in industry, academia and government. However, Typical VANET is challenged by high speed mobility and insufficient spectrum resources over congested scenarios. To address those serious problems, some articles have introduced Cognitive Radio (CR) technology into VANET and formed CR-VANET. In this article, we propose an overview of CR-VANET by exploring different architectures and features. Moreover, we provide taxonomy of state-of-the-art papers in this emerging field and the key articles are well analyzed respectively. In addition, we illustrate both research and application frameworks of CR-VANET based on our works, and propose some open research issues for inspiring future work.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.8
• /
• pp.3749-3768
• /
• 2018

As one of the most potential types of low-density parity-check (LDPC) codes, CPM-QC-LDPC code has considerable advantages but there still exist some limitations in practical application, for example, the existing decoding algorithm has a low convergence rate and a high decoding complexity. According to the structural property of this code, we propose a new method based on a CPM-RID decoding algorithm that decodes block-by-block with weights, which are obtained by neural network training. From the simulation results, we can conclude that our proposed method not only improves the bit error rate and frame error rate performance but also increases the convergence rate, when compared with the original CPM-RID decoding algorithm and scaled MSA algorithm.

• Journal of Communications and Networks
• /
• v.4 no.4
• /
• pp.314-320
• /
• 2002

The various sectors of the International Telecommunication Union (ITU) have been addressing the evolution of thirdgeneration and future wireless systems in the context of a comprehensive International Mobile Telecommunications 2000 (IMT-2000) project, and within the ITU’s Telecommunication Standardization Sector (ITU-T) a Special Study Group on “IMT-2000 and Beyond” has been established to address the network aspects of these emerging wireless systems. The Special Study Group (SSG) is playing a global role in this general area, in which a number of regional standards development organizations and a variety of industry forums are also active. This paper provides background information on the SSG and describes the SSG’s ongoing work addressing medium-term issues relating to convergence of fixed and mobile systems and the harmonization of evolving IMT-2000 systems. The paper also addresses related mobility management aspects.

• ETRI Journal
• /
• v.26 no.1
• /
• pp.13-20
• /
• 2004

We propose a new optical code division multiple access (OCDMA) scheme for reducing multiple access interference (MAI) and enhancing performance for optical subscriber access networks using modified pseudorandom noise (PN)-coded fiber Bragg gratings with bipolar OCDMA decoders. Through the bipolar OCDMA decoder and the modified PN codes, MAI among users is effectively depressed. As the data are encoded either by a unipolar signature sequence of the modified PN code or its complement according to whether the data bit is 1 or 0, the bit error ratio (BER) can be more improved with the same signal to interference plus noise ratio over the conventional on-off shift keying-based OCDMA system. We prove by numerical analysis that the BER of the proposed bipolar OCDMA system is better than the conventional unipolar OCDMA system. We also analyze the spectral power distortion effects of the broadband light source.

• Proceedings of the KIEE Conference
• /
• 2004.05a
• /
• pp.89-91
• /
• 2004

Recently Neural Network techniques have widely used in adaptive and learning control schemes for production systems. However, generally it costs a lot of time for learning in the case applied in control system. Furthermore, the physical meaning of neural networks constructed as a result is not obvious. And in practice since it is difficult to the PID gains suitably lots of researches have been reported with respect to turning schemes of PID gains. A Neural Network-based PID control scheme is proposed, which extracts skills of human experts as PID gains. This controller is designed by using three-layered neural networks. The effectiveness of the proposed Neural Network-based PID control scheme is investigated through an application for a production control system. This control method can enable a plant to operate smoothy and obviously as the plant condition varies with any unexpected accident. This paper goal is convergence speed progress about applied gain of PID controller using the neural networks.

• Smart Media Journal
• /
• v.10 no.2
• /
• pp.48-54
• /
• 2021

Sepsis is one of the leading causes of mortality globally, and it costs billions of dollars annually. However, treating septic patients is currently highly challenging, and more research is needed into a general treatment method for sepsis. Therefore, in this work, we propose a reinforcement learning method for learning the optimal treatment strategies for septic patients. We model the patient physiological time series data as the input for a deep recurrent Q-network that learns reliable treatment policies. We evaluate our model using an off-policy evaluation method, and the experimental results indicate that it outperforms the physicians' policy, reducing patient mortality up to 3.04%. Thus, our model can be used as a tool to reduce patient mortality by supporting clinicians in making dynamic decisions.