Browsing by Author "Wang, K"

Now showing 1 - 6 of 6

An intelligent on-demand multicast routing protocol in Ad hoc networks
(2001)
Wang, K
;
Chang, CT
;
資訊工程學系
;
Department of Computer Science
In this paper we present an intelligent on-demand multicast routing protocol (IOD-MRP) which is suited in rapidly changing network environments, such as ad hoc networks. This protocol simplifies and enhances the existing core-assisted mesh protocol (CAMP). The main difference between our protocol and CAMP is eliminating the cores from CAMP. We remove the cores from CAMP and apply an on demand receiver-initiated procedure to dynamically build routes and maintain multicast group membership. In addition, flooding is an important issue in ad hoc networks and our protocol is able to reduce flooding messages. Simulation results show that IOD-MRP reduces number of control messages by 3% - 13% compared to CAMP and still has slightly lower packet delay and packet loss rates.
Design and analysis of an ATM switch with priority discarding scheme
(2001-03-01)
Wang, K
;
Wang, HJ
;
資訊工程學系
;
Department of Computer Science
In this paper; we propose an N x N high speed and non-blocking asynchronous transfer mode (ATM) switch with input and output buffers. In this switch, each buffer adopts a priority discarding scheme, which discards incoming cells of low-priority traffic when its queue length is greater than a predefined threshold value. Our switch also supports broadcast/multicast functions without increasing the cost and imposing a significant performance penalty. We use the discrete-time Markov chain model to analyze cell delay and cell loss probability for each traffic class. An example 4 x 4 ATM switch has been described with VHDL. We have verified the functionality of the switch via VHDL simulation, and have synthesized the switch to evaluate its area and timing. Experimental results and synthesis results show that our proposed ATM switch can meet a requirement for high speed and support QOS.
Design and implementation of fault-tolerant and cost effective crossbar switches for multiprocessor systems
(1999-01-01)
Wang, K
;
Wu, CK
;
資訊工程學系
;
Department of Computer Science
Two general crossbar switch models are proposed: the modified one-sided crossbar switch and the ripple K one-sided crossbar switch. They both balance cost and reliability, where cost is expressed in terms of crosspoint count or area. The two-sided crossbar switch and the one-sided crossbar switch are two cases of these structures. These structures provide choices for compromising structures between the two-sided crossbar switch and the one-sided crossbar switch. The effective bandwidths of these four crossbar switches are simulated. Simulation with VHDL has been performed to verify the functionality of each crossbar system. Synthesis has also been conducted to evaluate delay and area for each crossbar design. Experimental results demonstrate that the two general models of crossbar switches are cost-effective in terms of reliability and crosspoint count (or area) without reducing their effective bandwidths. The work promotes the use of the crossbar switch as an interconnection network for multiprocessor systems to enhance system performance and reliability with low cost.
Fault coverage and defect level estimation models for partially testable MCMs
(2000-04-01)
Tseng, WD
;
Wang, K
;
資訊工程學系
;
Department of Computer Science
The authors propose a simple and efficient mathematical model for designers to estimate fault coverage for partially testable multichip modules (MCMs). This model shows a relation between fault covet-age, lest methodology, and the fraction and distribution of design for testability (DFT) dies in MCMs. Experimental results show that the proposed model can efficiently predict the fault coverage of a partially testable MCM with less than 5% deviation. An automatic DFT dies deployment algorithm, based on the genetic algorithm and the model is proposed to help designers to obtain a fault coverage as close to the upper bound of fault coverage as possible. Two defect level estimation models, which relate fault coverage and manufacturing yield for measuring the test quality of MCMs under equiprobable and non-equiprobable faults, respectively, are also formulated and analysed to support the effectiveness of the model.
Intelligent location tracking strategy in PCS
(2000-02-01)
Wang, K
;
Liao, JM
;
Chen, JM
;
資訊工程學系
;
Department of Computer Science
An alternative strategy (AS) for reducing location update signalling load has been proposed to improve the classical strategy (CS) used in the Global System for Mobile Communications (GSM). However, the AS is suitable only for long-term events and fixed movement tracks. Location tracking (paging) costs more when the mobile host (MH) changes movement habits or encounters short-term events. A novel method is proposed that aims to reduce signalling load resulting from location tracking in the above situations. The key idea is to take user recent movement information (called the paging information record) into account to determine which location area to page first. Performance evaluation has been conducted using a city area zone model that can model a realistic city area environment. Experimental results show that the method can reduce location management signalling costs by 14-55% and 6-39% compared to CS and AS, respectively. The overhead of the approach is the additional storage space required (for storing MH profiles and paging information records) and the additional processing time. However, the overhead is insignificant in terms of today's memory capacity and processor speed, and the reduction in signalling load makes more bandwidth available.
Using statistical and neural network methods to explore the relationship between systematic risk and firm's long term investment activities
(2000)
Wang, K
;
Pao, HT
;
Fu, HC
;
管理科學系
;
Department of Management Science
This paper presents a neural network method to explore the relationship between the systematic risk and the long term investing activities for Taiwan's companies in the fiber industry and in the electronics industry. In general, diversification from long-term investment may reduce firm's systematic risk, but the empirical results in some literatures are controversial. For years regression methods have been used to analyze the possible impact of long-term investment activities on systematic risk. Since their relationship may not be linear we thus propose a neural network based sensitivity analysis for the possible non-linear relationships. We adapt 5 years data (1994-1998) from the TEJ financial database to conduct the proposed analysis. The results show that the systematic risk is reduced with investment activities for the fiber industry. But for the electronics industry, the systematic risk is higher as firms increase the long-term investment ratio. The difference is even more significant when we only consider the companies with higher portion of long term investment in assets. This diverse effect between industries may be one of the reasons why the empirical results are inconsistent. Our study can clarify the controversial between the systematic risk and the long-term investment activities, and offer some possible explanations.