2004 ALAR Working Paper Series

Delay Performance Comparison between Queueing Schemes in Packet Switches

Qiang Duan
Computer Science Department
University of Central Arkansas

Abstract
Packet switches are important building elements of computer networks that have significant influence on the overall network performance. The queueing scheme employed by a packet switch is a decisive factor on the performance achieved by the switch. It is well known that the output queueing (OQ) scheme gives the optimal performance. However, output queueing require very complex implementations that are not feasible for high-speed switches. The combined input and crossbar queueing (CICQ) scheme has attracted interests from both academia and industry because of its potential for achieving high performance with fair implementation complexity. In this paper, we compare performance of CICQ and OQ switches. We focus on delay performance, which is one of the most important performance parameters to support quality of service (QoS) in computer networks. Our previous investigation has found that CICQ switches cannot guarantee identical delay for each packet as OQ switches without speeding up the switching fabric. In this paper, we apply the network calculus technique to analyze the delay performance that can be guaranteed by CICQswitches without switching fabric speed up, and the ideal delay performance guaranteed by OQ switches. By comparing the delay performance of CICQ and OQ switches, we find that CICQ switches can achieve almost identical delay performance as OQ switches without switching fabric speedup.

Read paper ALAR-WP-04-01

Interval Computing and Information Technology

Chenyi Hu
Department of Computer Science
University of Central Arkansas

Abstract
Interval computing has become a powerful tool in applied research. In this paper, we briefly introduce interval computing, available software, and resent applications especially in information technology. It is a survey with points to technical details. This work is partially supported by NSF grants 0202042

Read paper ALAR-WP-04-02

Still Image Multi-Layer Watermarking with IDP

Mariofanna Milanova, Department of Computer Sience University of Arkansas at Little Rock
Charles Ford, Departent of Computer Science University of Arkansas at Little Rock
Vladimir Todorov, T&K Engineering
Roumiana Kountcheva, T&K Engineering

Abstract
In this paper, a new method is presented for image watermarking aimed at the archiving of digital images of paper documents. This specific application is very useful when large amounts of paper documents have to be stored for long time periods. The substitution of the paper originals with their electronic image copies offers easy access and storage but the electronic form permits contents editing and un-authorized changes. The insertion of one or more digital watermark in the compressed still image is a secure way to save the image contents unchanged. The new method for image watermarking is based on the Inverse Difference Pyramid (IDP) decomposition. This decomposition is performed in the image spectrum, processing the image in consecutive layers with increasing resolution and using a Discrete Cosine Transform, Walsh-Hadamard Transform, or combinations of both, retaining for the processing only the values of some coefficients selected in advance. The method permits the insertion of different watermark in every consecutive layer. The advantage of the method is that the compressed watermarked document should be used for archiving and every slightest change in the watermark after the decompression will evidence un-authorized editing of the document contents. The watermark is created together with the copy of the original paper document and the owner inserts the watermark using his own password. The receiving side is not able to change or extract the watermark, if not in possession of this password. Such watermarking offers additional encryption capabilities because it is hidden in the compressed image. The embedded watermarks are invisible and could be visualized on request. The method permits the insertion of visible (hiding or authorizing) watermarks as well.

Read paper ALAR-WP-04-03

Data Mining for Analyzing the Impact of Environmental Stress on Plants ? A Case Study Using OSMID

Dr. Richard S. Segall, Arkansas State University, Department of Economics and Decision Sciences, College of Business
Dr. Sarath A. Nonis, Arkansas State University, Department of Marketing and Management, College of Business

Abstract
This paper first provides a brief background on the basic concepts and development of data mining and how it relates to data warehousing. Data mining and data warehousing are relatively new and rapidly expanding areas of information systems for which new courses and curricula are being created. A brief background on the economics of plant biotech is provided. The plant data used in this paper from the Osmotic Stress Microarray Information Database (OSMID) are considered to be representative of those that could be used for biotech application such as the manufacture of plant-made-pharmaceuticals (PMP) and genetically modified (GM) foods. Data mining of selected plant data from the OSMID data warehouse is performed with investigations of both economic and environmental factors. Conclusions and future directions of the research are discussed.

Read paper ALAR-WP-04-04




	2004 ALAR Working Paper Series Delay Performance Comparison between Queueing Schemes in Packet Switches Qiang Duan Computer Science Department University of Central Arkansas Abstract Packet switches are important building elements of computer networks that have significant influence on the overall network performance. The queueing scheme employed by a packet switch is a decisive factor on the performance achieved by the switch. It is well known that the output queueing (OQ) scheme gives the optimal performance. However, output queueing require very complex implementations that are not feasible for high-speed switches. The combined input and crossbar queueing (CICQ) scheme has attracted interests from both academia and industry because of its potential for achieving high performance with fair implementation complexity. In this paper, we compare performance of CICQ and OQ switches. We focus on delay performance, which is one of the most important performance parameters to support quality of service (QoS) in computer networks. Our previous investigation has found that CICQ switches cannot guarantee identical delay for each packet as OQ switches without speeding up the switching fabric. In this paper, we apply the network calculus technique to analyze the delay performance that can be guaranteed by CICQswitches without switching fabric speed up, and the ideal delay performance guaranteed by OQ switches. By comparing the delay performance of CICQ and OQ switches, we find that CICQ switches can achieve almost identical delay performance as OQ switches without switching fabric speedup. Read paper ALAR-WP-04-01 Interval Computing and Information Technology Chenyi Hu Department of Computer Science University of Central Arkansas Abstract Interval computing has become a powerful tool in applied research. In this paper, we briefly introduce interval computing, available software, and resent applications especially in information technology. It is a survey with points to technical details. This work is partially supported by NSF grants 0202042 Read paper ALAR-WP-04-02 Still Image Multi-Layer Watermarking with IDP Mariofanna Milanova, Department of Computer Sience University of Arkansas at Little Rock Charles Ford, Departent of Computer Science University of Arkansas at Little Rock Vladimir Todorov, T&K Engineering Roumiana Kountcheva, T&K Engineering Abstract In this paper, a new method is presented for image watermarking aimed at the archiving of digital images of paper documents. This specific application is very useful when large amounts of paper documents have to be stored for long time periods. The substitution of the paper originals with their electronic image copies offers easy access and storage but the electronic form permits contents editing and un-authorized changes. The insertion of one or more digital watermark in the compressed still image is a secure way to save the image contents unchanged. The new method for image watermarking is based on the Inverse Difference Pyramid (IDP) decomposition. This decomposition is performed in the image spectrum, processing the image in consecutive layers with increasing resolution and using a Discrete Cosine Transform, Walsh-Hadamard Transform, or combinations of both, retaining for the processing only the values of some coefficients selected in advance. The method permits the insertion of different watermark in every consecutive layer. The advantage of the method is that the compressed watermarked document should be used for archiving and every slightest change in the watermark after the decompression will evidence un-authorized editing of the document contents. The watermark is created together with the copy of the original paper document and the owner inserts the watermark using his own password. The receiving side is not able to change or extract the watermark, if not in possession of this password. Such watermarking offers additional encryption capabilities because it is hidden in the compressed image. The embedded watermarks are invisible and could be visualized on request. The method permits the insertion of visible (hiding or authorizing) watermarks as well. Read paper ALAR-WP-04-03 Data Mining for Analyzing the Impact of Environmental Stress on Plants ? A Case Study Using OSMID Dr. Richard S. Segall, Arkansas State University, Department of Economics and Decision Sciences, College of Business Dr. Sarath A. Nonis, Arkansas State University, Department of Marketing and Management, College of Business Abstract This paper first provides a brief background on the basic concepts and development of data mining and how it relates to data warehousing. Data mining and data warehousing are relatively new and rapidly expanding areas of information systems for which new courses and curricula are being created. A brief background on the economics of plant biotech is provided. The plant data used in this paper from the Osmotic Stress Microarray Information Database (OSMID) are considered to be representative of those that could be used for biotech application such as the manufacture of plant-made-pharmaceuticals (PMP) and genetically modified (GM) foods. Data mining of selected plant data from the OSMID data warehouse is performed with investigations of both economic and environmental factors. Conclusions and future directions of the research are discussed. Read paper ALAR-WP-04-04 © Copyright 2005 Acxiom Corporation. All rights reserved. Please read our Detailed Online Privacy Statement.