• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2014.05a
• /
• pp.693-694
• /
• 2014

This paper proposes a new pipelined dynamic bandwidth allocation algorithm for XG-PON (10-Gbps-capable passive optical network) passive optical networks) system. The pipelined algorithm is used when a dynamic bandwidth allocation algorithm cannot finished in an unit time. In the proposed mechanism, the request is immediately transferred to each pipeline block to improve performance.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2014.05a
• /
• pp.705-706
• /
• 2014

This paper proposes a new dynamic bandwidth allocation algorithm for an XG-PON (10-Gbps-capable passive optical network) system. In the XG-PON system, an ONU (optical network unit) does not explicitly send it request to an OLT (optical line termination). The OLT monitors the upstream bandwidth usage of the ONU to estimate the request of the ONU. Then the OLT performs a dynamic bandwidth allocation process based on the estimated requests.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.28 no.7B
• /
• pp.620-627
• /
• 2003

Ethernet passive optical networks (EPONs) are an emerging access network technology that provide a low-cost method of deploying optical access lines between a carrier's central office and customer sites. Dynamic bandwidth allocation (DBA) provides statistical multiplexing between the optical network units for efficient upstream channel utilization. To support dynamic bandwidth distribution, 1 propose an cyclic polling-based DBA algorithm for differentiated classes of service in EPONs. And, I show that an interleaved polling scheme severely decreases downstream channel capacity for user traffics when the upstream network load is low. To obtain realistic simulation results, I used synthetic traffic that exhibits the properties of self-similarity and long-range dependence I then analyzed the network performance under various loads, specifically focusing on packet delays for different classes of traffic.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.32 no.8B
• /
• pp.547-552
• /
• 2007

IEEE ratified IEEE802.3ah as the standard of E-PON, while it leaved the specific method of upstream bandwidth allocation as a role of implementation vendors. Many experts have researched the method of enhancing upstream bandwidth throughput and released related papers. This paper presents another novel mechanism to enhance upstream throughput. This mechanism performs the management of upstream queues by giving the minimum bandwidth of different level to each queue. In order to process packets on each queue we adopted a modified weighted DRR technology. By doing so, the transmission throughput of upstream packets can be largely enhanced. The experimental simulation of this mechanism showed an enhancement of bandwidth utilization more than 10% in comparison to legacy method.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.29 no.11B
• /
• pp.891-899
• /
• 2004

Ethernet passive optical network (EPON) has drawn many attention as a promising access network technology because it can provide a high bandwidth with a low cost. Since the uplink in the EPON system is shared by many users, it is necessary for an EPON system to have an efficient bandwidth allocation mechanism. In addition, as the users demand more QoS-oriented applications, it is necessary for an EPON system to have an efficient mechanism supporting QoS. In this paper, we propose a variable window dynamic bandwidth allocation (DBA) algorithm for the EPON system. Unlike the previously proposed DBA algorithms, the variable window algorithm guarantees the QoS for the highest priority class, and at the same time provides more enhanced QoS for the lower priority classes by dynamically allocating bandwidth if necessary. It is verified through the simulations that the variable window algorithm can provide more enhanced QoS performance than other DBA algorithms.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2015.05a
• /
• pp.449-450
• /
• 2015

This paper describes performance evaluation results of a new dynamic bandwidth allocation algorithm for an XG-PON (10-Gbps-capable passive optical network) system without using an explicit request. An ONU (optical network unit) does not send its request to an OLT (optical line termination). The OLT monitors the upstream bandwidth usage of the ONU to estimate the request of the ONU.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.26 no.9A
• /
• pp.1574-1582
• /
• 2001

본 논문에서는 무선 ATM 망에서 VBR 트래픽에 대한 동적인 대역할당 기법을 제안하였다. 제안된 기법에서는 동적 파라미터를 전송하기 위하여 인 밴드 방식을 사용하였으며, 버퍼의 길이와 변이를 고려하여 이동국에서 2 state bit를 기지국에서 전송하도록 하였다. 기지국에서 수신된 비트의 상태에 따라서 이동국에 데이터 슬롯을 할당한다. 제안한 알고리즘은 VBR 트래픽의 QoS를 보장하고, 시스템의 효율을 높일 수 있도록 하였고, 컴퓨터 시뮬레이션을 통하여 제안된 알고리즘의 성능을 평가하였다.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.43 no.5 s.347
• /
• pp.94-103
• /
• 2006

Ethernet passive optical network (EPON) has drawn many attention as a promising access network technology for FTTH because it can provide a high bandwidth with a low cost. Since the uplink bandwidth in the EPON system is shared by many users, it is necessary for an EPON system to have an efficient bandwidth allocation mechanism. To support QoS in EPON, the previous bandwidth allocation schemes employ strict priority queueing (SPQ). Since SPQ gives unlimited priority to higher service class, the QoS of lower service classes gets worse. In this paper, we propose Class Gated DBA (Dynamic Bandwidth Allocation) algorithm in which the bandwidth is requested / granted in a service class basis. To avoid the monopoly in bandwidth usage by higher classes the maximum bandwidth that is allocate to each service class is limited (fairness between services classes). In addition, to avoid the monopoly in bandwidth usage by some particular users, each ONU runs fairness bandwidth allocation algorithm within each service classes. Through computer simulations, it is verified that the proposed algorithm achieves a good level of QoS, and at the same time maintains a good level of fairness between both service classes and users.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.24 no.7B
• /
• pp.1229-1238
• /
• 1999

To maximize the statistical multiplexing gain in wireless ATM MAC, a centralized scheduling function must be provided for the dynamic slot assignment in the wireless ATM access point. Exploiting the dynamic parameters received in the wireless ATM access point, the centralized scheduler will execute a dynamic slot assignment to determine bandwidth required for individual wireless terminal. In this paper, we propose a new scheduling algorithm, which can effectively reduce a burden of the explicit signaling requirement of transmitting the dynamic parameters in a timely manner. We have demonstrated that the proposed approach outperforms the existing schemes in terms of the timeliness of transmitting the dynamic parameters, which is a critical factor to the performance of delay-sensitive service classes, and works as an effective means of dynamic slot assignment without heavily relying on the dynamic parameters in the wireless ATM networks.

• The KIPS Transactions:PartC
• /
• v.9C no.1
• /
• pp.73-78
• /
• 2002

In WATM networks, in order to perform dynamic slot allocation required slots of mobile terminals are estimated based on DP (Dynamic Parameter) reflecting characteristics of traffic. In VBR (Variable Bit Rate) traffic, slot allocation is done at MT considering both time-dependent characteristics and QoS (Quality of Service) requirements. In this paper, DPs-buffer state information and buffer state change-are transmitted through in-band signaling. BS (Base Station) performs dynamic slot allocation considering traffic characteristics of each MT (Mobile Terminal), in other words, buffer state information informs the potentiality of 'buffer full state'to BS if MT buffer is over the specific threshold value and buffer state change notifies change in buffer state of incoming cells to MT. If buffer state information is equal to 'low (more than threshold)' and 'abrupt increase' it generates 'buffer full' state cell transmission delay or cell loss might occur. At this time BS should assign additional slots to MT, and then MT consumes cells in its buffer. In simulation, the proposed scheme shows better performance in cell delay and loss than EPSA (Estimation-Prorated Slot Assignment) in-band scheme.