- Advanced Photonic Communication Systems
- Advanced Optical Network Technologies
- Optical Network Technologies
- Interconnection Networks and Systems
- Scheduling and Optimization Algorithms
- Advanced Queuing Theory Analysis
- Embedded Systems Design Techniques
- Network Traffic and Congestion Control
- Software-Defined Networks and 5G
- Ultra-Wideband Communications Technology
- Optimization and Search Problems
- Wireless Networks and Protocols
- Simulation Techniques and Applications
- VLSI and FPGA Design Techniques
- Parallel Computing and Optimization Techniques
- Manufacturing Process and Optimization
- Advanced Fiber Optic Sensors
- Assembly Line Balancing Optimization
- Semiconductor Lasers and Optical Devices
- Advanced Wireless Network Optimization
- DNA and Nucleic Acid Chemistry
- Supply Chain and Inventory Management
- Cybersecurity and Information Systems
- Biofuel production and bioconversion
- Advanced Data and IoT Technologies
Mokpo National University
2008-2024
Tsinghua University
2021
Youngsan University
2019
Universidade do Estado do Rio de Janeiro
2012
Istituto Nazionale di Fisica Nucleare, Sezione di Genova
2012
Istituto Nazionale di Fisica Nucleare, Laboratori Nazionali di Frascati
2012
Technical University of Darmstadt
2012
Hanyang University
2010-2012
University of California, Berkeley
2012
Halbleiterlabor of the Max-Planck-Society
2012
We suggest an immediate allocation with colorless grant (IACG) method for dynamic bandwidth of a gigabit passive optical network (GPON) system. The suggested is based on the GigaPON Access Network (GIANT) media access control (MAC) method. IACG can assign transmission regardless value scheduling counter, whereas GIANT MAC only when counter has 1. Also, unlike method, utilizes unallocated remainder bandwidth. At end scheduling, assigns to each unit (ONU) equal share. Using computer...
In this paper, we propose a dynamic bandwidth allocation with high utilization (DBAHU) algorithm in order to utilize the unused of service class. DBAHU is based on simple and feasible (SFDBA) algorithm. Like SFDBA, uses common available byte counter down for multiple queues However, class, an can be negative unlike SFDBA. Also, remainder class added counters other classes who require more bandwidth. Using simulations, show that superior SFDBA mean delay frame variance.
Abstract In this paper, a model of cyclic production line is introduced and analyzed. Using Taylor series expansion, we present an approximation method for the analysis average steady-state throughput lines with unreliable machines, finite buffers fixed number pallets circulating in line. method, suggest analytic to calculate optimal that maximizes throughput. Also, show absolute exists which independent machine parameters.
In this paper, we propose a simple and feasible dynamic bandwidth allocation (SFDBA) algorithm in order to utilize the unallocated achieve implementation feasibility. SFDBA is based on an immediate with colorless grant (IACG) but uses only single available byte counter down for multiple queues of same service class. Since share counter, queue can be utilized by another queues. For better fairness, changes starting scheduling round-robin manner. Using simulations, show that superior existing...
Ultra-wideband (UWB) technology has offered a large application bandwidth which been reached up to 1 Gb/s recently. One of various UWB specifications, medium access control (MAC) and physical layer (PHY) technologies WiMedia Alliance can also be used for high speed multimedia applications establish wireless video area network (WVAN). In this paper, we propose transmission system WVANs. order verify the possibilities multiple streams, first built an analysis model simulation model. Then...
In this paper, a model of highly reliable serial production line is introduced and analysed. The high rate machines in isolation quality inspection are the basis lines. Using Taylor series expansion, we decompose lines into two one buffer then utilizing decomposed lines, approximate average steady state closed form. Finally, validate results by simulations.
We propose a two-stage resource allocation algorithm for the high link utilization of an orthogonal frequencydivision multiplexing passive optical network (OFDM-PON).An OFDM-PON is assumed to use synchronous frame structure in supporting service differentiation.In distributing resources, proposed first allocates time window each unit (ONU), and then it arranges subchannel, which group subcarriers.This needs satisfy two constraints.First, computations should be done using unit.Second, ONU has...
This paper proposes a pipelined architecture for dynamic bandwidth allocation (DBA) in 10-gigabit-capable passive optical networks (XG-PONs).In the proposed architecture, DBA operation is performed multiple stages high energy efficiency.Also, forwards requests to next obtain performance.Using simulations, we evaluate performance of architecture.
A fast class-of-service oriented packet scheduling (FCOPS) has a service fairness problem since credit pool for class is initialized at the beginning of transmission cycle whose starting moment fixed specific ONU. To remedy unfairness FCOPS, we suggest an enhanced (ECOPS) that uses new initialization fairly distributed to each Also, ECOPS generates colorless grant utilize resource wastage, when traffic light and total sum grants ONU less than minimum size. Using simulation, validate as...
We suggest a new minimum credit method for the dynamic bandwidth allocation in EPON. In suggested method, to eliminate unused transmission time-slot, each ONU requests no more than predetermined maximum. analyze upstream channel resource wastage when traffic is light. Based on analysis, we derive that wastage. The OLT estimates load and grants request smaller Using simulation, show discipline superior existing methods mean delay frame loss rate.
This paper proposes a new scheme for dynamic bandwidth allocation XG-PON (10-Gbps-capable passive optical network). In the propose scheme, an ONU (optical network unit) does not explicitly report its queue status to OLT line termination). The monitors usage of each estimate ONU. allocates when previous allocated was fully used. Also periodically probe check
Wireless transmission of uncompressed video guarantees higher quality with lower latency than compressed transmission. Although current wireless technologies cannot fully cover required data rates about a few Gb/s for full high definition resolution, some such as ultra-wideband (UWB) provide 1 rate which is adequate in portable devices. In this paper, we propose an system mirroring services potable We firstly simulated the performance using UWB technology. Then implemented hardware-based...
This paper introduces a method of implementing Lenet5, the best-known convolutional neural network, on FPGA (field programmable gate arrays) using HLS (high-level synthesis) tools.HDL (hardware description language) and tools are mainly used in implementation Lenet5.HLS is implemented directly from C code to FPGA, so development period short, but compared HDL, there limits optimization that users can do.In this paper, improve operation speed memory was duplicated, for loop statement unrolled...
With the advent of software-defined networks, network virtualization becomes a key technology to implement softwaredefined networks. Network requires path computation element (PCE) calculate virtual paths connect nodes. The Dijkstra's algorithm has been widely used in PCE shortest between two In this paper, we address that cannot be applicable when non-linear cost metric is evaluation. This paper proposes new genetic (GA) find used.The proposed GA generates immigrants from ordinary...
An XG-PON (10-Gbps-capable passive optical network) system is comprised of an OLT (optical line termination) and multiple ONUs network units). In general, report their requests to the then performs a dynamic bandwidth allocation operation allocate non-overlapping transmission slots ONUs. this paper, we consider that do not to. The estimates ONU status by monitoring upstream usage ONU. We propose method for in which reports. proposed allocates only if has fully used its slot. Also,...