Jin Hee Choi

One-way delay estimation and its application

Delay estimation is a difficult problem in computer networks. Accurate one-way delay estimation is crucial because it serves a very important role in network performance and thus application design. RTT (Round Trip Time) is often used as an approximation of the delay, but because it is a sum of the forward and reverse delays, the actual one-way delay cannot be estimated accurately from RTT. To estimate one-way delay accurately, this paper proposes a new scheme that analytically derives one-way delay, forward and reverse delay, respectively. We show that the performance of TCP can improve dramatically in asymmetric networks using our scheme. A key contribution of this paper is that our one-way delay estimation is much more accurate than RTT estimation so that TCP can quickly find the network capacity in the slow start phase. Since RTT is the sum of the forward and reverse delays, our scheme can be applied to any protocol that is based on RTT.

Eliminating the performance anomaly of 802.11b

In this paper, we propose a mechanism to eliminate the performance anomaly of IEEE 802.11b. Performance anomaly happens when nodes that have different transmission rates are in the same wireless cell. All the nodes in the cell might experience the same throughput even though their transmission rates are different because DCF of WLAN provides equal probability of channel access, but it does not guarantee the equal utilization of the wireless channel among the nodes. To reduce such a performance anomaly, we adjust the frame size proportionally depending on the bit rate. Additionally, our scheme eliminates the performance anomaly in multi-hop case. Simulation study shows that our scheme achieves an improvement in the aggregate throughput and the fairness.

Analytic end-to-end estimation for the one-way delay and its variation

Delay estimation is a difficult problem in computer networks. RTT (round trip time) is often used as an approximation of the delay, but because it is a sum of the forward and reverse delays, the actual one-way delay cannot he estimated accurately from RTT. Accurate one-way delay estimation becomes crucial because it serves a very important role in network and application design. This paper proposes a new scheme to estimate one-way delay and its variation. The scheme calibrates estimated one-way delay so in a brief duration as to be used in many protocols that adjust their behavior depending on the network condition. We analytically derive one-way delay, forward and reverse delay respectively, and show that our one-way delay estimation is much more accurate than RTT estimation by simulation.

A flow control scheme based on buffer state for wireless TCP

Reliable transport protocols such as TCP (transmission control protocol) are tuned to perform well in traditional networks where packet losses occur mostly because of congestion. However, on the wireless links where packet losses frequently happen, TCP shows a serious drop in performance. We propose a scheme, called BSF (buffer state based flow control), which can solve this problem. BSF is implemented at the base station of a wireless network. So, it does not need any modification of the TCP stacks in the end system. BSF uses the buffer state of the base station as the criterion to detect wireless link error and reduces unnecessary contraction of congestion window in TCP. As a result, the buffer of the base station maintains a non-empty state to maximize the link utilization. In our simulation study, this approach achieved up to 250 % improvement in performance.

TCP-aware source routing in mobile ad hoc networks

Temporary link failures and route changes occur frequently in mobile ad hoc networks. Since TCP assumes that all packet looses are due to network congestion, TCP does not show satisfactory performance in ad hoc networks. In this paper, we propose a simple and new mechanism called TSR, TCP-aware source routing, which can improve TCP performance in mobile ad hoc networks. By reducing the number of invalid routes, TSR minimizes TCPs consecutive timeouts. In our simulation study, TSR achieves up to 50% TCP performance improvement without requiring any modification of TCP stack in end systems.

Adapting TCP segment size in cellular networks

In cellular networks, a frame size is generally made small to reduce the impact of errors. Thus, a segment of transport layer is splitted into multiple frames before transmission. A problem is that the whole segment is lost when a frame of a segment is lost. So, the segment error rate tends to be high even though the cellular network provides relatively the low frame error rate, which drops TCP performance. However, the relation between the frame size, the segment size and the error rate has not been closely investigated. In this paper, we analyze the impact of the segment size on TCP performance in cellular networks and propose a scheme alleviating the performance drop of TCP. Our result shows that our scheme reduces the drop by 82%.

An enhancement scheme for TCP over mobile ad hoc networks

In a mobile ad hoc network, temporary link failures and route changes occur frequently. With the assumption that all packet losses are due to congestion, TCP performs poorly in such an environment. This paper proposes a new mechanism called TSR, TCP-aware source routing, which can improve TCP performance in wireless ad hoc networks. TSR adds a hold state to an existing routing protocol to reduce consecutive timeouts, retransmission, and out-of-ordered packets in TCP. In our simulation study, TSR achieves up to a 60% improvement in performance, without requiring any TCP stacks in end systems to be modified.

DR-TCP: Downloadable and reconfigurable TCP

Advances in communication technology allow a variety of new network environments and services available very rapidly. Appearance of various network environments tends to enable a user with a mobile terminal to access among different network simultaneously. However, since new network environment affects performance of communication protocols, terminal systems should provide adaptation schemes for the protocols in order to keep the quality of network performance high. A possible solution is to make the protocol reconfigurable to be adapted to current network environment. Unfortunately, because most existing network systems are implemented monolithically, they cannot support protocol reconfiguration dynamically at runtime. This paper proposes a new reconfigurable model that enables TCP functions to be adapted whenever network environment is changed. The proposed scheme also supports binary-level protocol upgrade for extensibility by downloading new TCP variants which the terminal does not have for new network environment, and it is more suitable for mobile hand-held devices than existing source-level solution. Our model is based on a recursive state machine. We re-implement TCP Reno from scratch using our proposed model. The new implementation of TCP Reno is named DR-TCP. To demonstrate the effectiveness of DR-TCP, dynamic reconfiguration is performed over Internet, which successfully converts DR-TCP to TCP Westwood at runtime.

Dynamic segment size adjustment for TCP performance in cellular networks

We analyze the impact of the maximum segment size (MSS) on TCP performance in cellular networks and propose a scheme alleviating the performance drop of TCP, which adjusts the segment size dynamically.

A decision maker for transport protocol configuration

This paper proposes an approach, called Protocol Configuration Decision Maker, for TCP to dynamically adapt to a network environment. The proposed mechanism monitors the network condition with parameters like loss rate. Then it consults a knowledge database to see whether a better performance can be achieved and replaces a relevant TCP module with the one instructed by the database. Through simulation studies, we show that our mechanism helps TCP achieve better throughput than normal TCP Reno, up to 80~194%.