Yao-Nan Lien

Transformation of the generalized traveling-salesman problem into the standard traveling-salesman problem

Abstract In the generalized traveling-salesman problem (GTSP), we are given a set of cities that are grouped into possibly intersecting clusters. The objective is to find a closed path of minimum cost that visits at least one city in each cluster. Given an instance G of the GTSP, we first transform G into another instance G ′ of the GTSP in which all the clusters are nonintersecting, and then transform G ′ into an instance G ″ of the standard traveling-salesman problem (TSP). We show that any feasible solution of the TSP instance G ″ can be transformed into a feasible solution of the GTSP instance G of no greater cost, and that any optimal solution of the TSP instance G ″ can be transformed into an optimal solution of the GTSP instance G .

Design of Distributed Databases on Local Computer Systems with a Multiaccess Network

Concurrency control, distribution design, and query processing are some of the important issues in the design of distributed databases. In this paper, we have studied these issues with respect to a relational database on a local computer system connected by a multiaccess broadcast bus. A broadcast bus allows information to be distributed efficiently, and hence simplifies the solutions to some of these issues. A transaction model that integrates the control strategies in concurrency control and query processing is proposed. In concurrency control, the lock, unlock, and update of data are achieved by a few broadcasts. A dynamic strategy is used in query processing, as less data are transferred when compared to a static strategy. The status information needed in dynamic query processing can be conveniently obtained by broadcasting. Lastly, some NP-hard file placement problems are found to be solvable in polynomial time when updates are broadcast.

A novel mobile agent search algorithm

Intelligent agents have been shown to be a good approach to addressing the issues of limited capacity and unreliable wireless links in mobile computing. However, before the approach can be commercially viable, a set of management capabilities that support the controls of intelligent agents in a mobile environment need to be in place. Since controls can only be applied after the target agent is located, an effective agent search algorithm is an indispensable part of the management functions. In this paper, we propose a new algorithm, the highest probability first algorithm, for locating the target agent. The approach makes use of the execution time information to reduce cost and network traffic. The execution time of the agent on a server is assumed to be binomial distributed and therefore is more realistic.

A New TCP Congestion Control Mechanism over Wireless Ad Hoc Networks by Router-Assisted Approach

Due to unawareness of network conditions, regular transmission control protocol (TCP) is not able to fully control the limited resources and distinguish packet loss from congestion loss and random loss. Because of such characteristics, the performance of TCP will be downgraded severely if it runs on wireless networks, such as multi-hop ad hoc networks. In an ad hoc network, nodes play both roles of end system as well as router. We take this advantage to propose a new TCP congestion control mechanism by router-assisted approach. Based on the information feedbacked from routers, a TCP sender is able to adjust its sending speed dynamically in order to avoid overshooting problem. Our proposed protocol has 5-10% higher throughput than TCP NewReno and much less number of retransmissions. The fairness is also achieved while our proposed protocol coexists with other major TCP variants.

A Walkie-Talkie-Like Emergency Communication System for Catastrophic Natural Disasters

When a catastrophic natural disaster strikes, an organized and effective rescue operation is essential to rescue those victims trapped under collapsed buildings or landslides as well as to relief massive survivals who lose their life support. However, communication systems were usually paralyzed by many causes. The loss of communication systems made rescue and relief operations extremely difficult costing many lives unnecessarily. We had proposed a MANET based emergency communication and information system, P2Pnet, which can support a large number of voluntary workers under catastrophic natural disasters. This paper demonstrates the design of a subsystem of P2Pnet, a Walkie-Talkie-like communication system, which can be used in the early hours or days after a natural disaster strikes. We wish to stimulate the research on the emergency communication systems that is inexpensive and easy to deploy for future catastrophic natural disasters.

Challenges of emergency communication network for disaster response

When stricken by a catastrophic natural disaster, the efficiency of disaster response operation is very critical to life saving. However, communication systems, including cellular networks, were usually crashed due to various causes making the coordination among a large number of disorganized disaster response workers extremely difficult. Survival rate can be greatly increased by breaking the communication bottleneck. Based on our firsthand experience in 921 Chi-Chi Earthquake and extensive research, we synthesize a set of requirements for deploying an emergency communication network for disaster response (ECN-DR), which can be deployed and operated under extreme environment and unusually communication behaviors in disaster areas. There are seven challenges, called 7-ability, that have to be addressed when designing an ECN-DR. 7-ability includes popularity, usability, practicability, capacity, reliability, operability and adaptability. Popularity and usability are used to fulfill end-user requirements. Practicability, capacity, reliability, operability and adaptability are used to fulfill operator requirements. Finally, we propose high level system architecture of ECN-DR. Several research issues about how 7-ability can be achieved will be presented.

On the search of mobile agents

In a mobile computing environment that supports mobile agents, a client can send an agent to visit a sequence of servers in the network. To track the location of agents becomes a critical problem in managing a mobile agent service network. This paper studies the agent search problem based on an open architecture proposed by Lien (see Proc. of the First Workshop on Mobile Computing, Hsing-Chu, Taiwan, p.2-9, 1995) and is currently being implemented in the National Chengchi University.

A Multi-hop Walkie-Talkie-Like Emergency Communication System for Catastrophic Natural Disasters

When stricken by a catastrophic natural disaster, the efficiency of disaster response operation is very critical to life saving. However, communication systems, including cellular networks, were usually crashed due to various causes making the coordination among a large number of disorganized disaster response workers extremely difficult. Unfortunately, rapid deployment of many existing emergency communication systems relies on a good transportation system, which is usually not available in a catastrophic natural disaster. We designed a multi-hop walkie-talkie-like communication system based on P2Pnet platform, which is a MANET P2P network constructed using volunteers’ laptops. This system can support a large number of voluntary workers in the early hours of a catastrophic natural disaster when external assistance is blocked by the paralyzed transportation system. The multi-hop version can help to bypass the obstacles that block face-to-face communication as well as direct WiFi communication links. We wish to stimulate the research on the emergency communication systems that is inexpensive and easy to deploy for future catastrophic natural disasters.

Hop-by-Hop TCP over MANET

In a MANET environment, communication links are unstable due to various reasons. Error rate is higher and bandwidth is smaller than fixed networks. Running conventional TCP protocol over MANET will suffer from severe performance degradation. To handle a packet loss, conventional TCP retransmits the lost packet from its source. However, when error rate is high, it may have to take several retransmissions to deliver a packet to its destination successfully. Considering that most applications on a MANET prefer shorter packet delivery time to higher bandwidth, this paper proposes a hop-by-hop TCP protocol for MANET aiming to accelerate packet delivery. Hop-by-hop TCP makes every intermediate node in the transmission path execute a local TCP to guarantee the transmission of each packet on each link. The retransmission of a lost packet is right at the transmitting end of the link where the packet is lost. It takes less time in average to deliver a packet in a high error rate environment. The performance of our approach is evaluated by simulation using NS-2 simulator. Our experiments show that our proposed protocol outperforms TCP Reno by 25.7% in throughput and 25% reduction in average packet delivery time. The fairness requirement is also achieved while our proposed protocol coexists with other major TCP variants.

Performance issues of P2P file sharing over asymmetric and wireless networks

Some peer-to-peer (P2P) file sharing operation models over asymmetric networks have several shortcomings that may affect system and network performance: data transmission paths are highly redundant wasting a lot of backbone bandwidth, the download throughput in a node may be limited by the upward bandwidth of other nodes; TCP performance is deteriorated due to the blocking of acknowledge packets on the upward channel. These shortcomings severely impair the efficiency of P2P file sharing as well as network performance. These problems are further complicated by the voluntary nature of P2P: peer nodes are fairly unstable. For similar reasons, when such P2P file sharing paradigm is moving to wireless networks, it may suffer from even worse performance degradation due to many reasons such as lower link reliability, lower bandwidth, and the impairment of TCP protocol. This paper analyzes these problems mainly from topology and protocol viewpoints and proposes some solution approaches to alleviate some of these problems.