Proportional Bandwidth Engineering for Service Differentiation
P. Pereira, A. Casaca
Abstract
This paper describes two bandwidth engineering techniques for implementing proportional service differentiation based on Multiprotocol Label Switching (MPLS) traffic engineering. Both use dynamic bandwidth allocation schemes to modify the bandwidth reserved by each traffic class according to the current network load. The first scheme uses an adaptive algorithm that qualitatively determines the required average throughput per flow for each class and moves bandwidth between classes for each path as necessary. The second scheme mathematically divides the existing bandwidth through the traffic classes for each path. The quality of service that users get with both techniques is assessed by simulation and compared with a fixed bandwidth allocation scheme.
Super Light Virtual Machine (SLVM) – Virtual machine for Microcontrollers with Highly Reduced Resources
S. Rusinov, R. Ilarionov
Abstract
This article presents a concept of a virtual machine able to run on microcontrollers with very small resources, such as PIC18 family of Microchip. The main features of the VM are real-time processing, multi-tasking and memory management which meets the needs of small embedded projects.
A Comparative Study of the TCP Control Algorithms
G. Kirov
Abstract
The publication presents a comparative study of the different congestion control mechanisms implemented by the Transmission Control Protocol (TCP): Slow-Start and Congestion Avoidance without Fast Retransmit, Tahoe, Reno, and New-Reno. The state of the art of the TCP control approaches is described. The advantages and drawbacks of the above-mentioned algorithms are investigated through the simulation investigations. The TCP performance analysis is based on two scenarios of the network simulation with different percentages of the packet loss.
Robust Decoupling Control of Induction Motors by Exact Feedback Linearization
S. Enev
Abstract
This paper presents an induction motor control system based on the exact feedback linearization approach. The underlying model used for control design is the fifth-order stator-flux model. Linearizing and decoupling transformation and control law are derived. Effects of parameter variations on the control system structure are derived in the frame of this approach. It is shown that the proposed control law does not lose its linearizing and decoupling properties, though some additional feedback connections enter the system structure. PID and PI controllers are designed in the outer loops taking into account a second degree of freedom, introduced by specified prefilters in both loops. Speed control simulation results are presented, confirming the feasibility of the overall control system.
Concept Learning and Classification with Prime Implicants Applied to Numerical Domains
Z. Shevked, L. Dakovski
Abstract
In this paper we discuss an algorithm for logical function minimization and its application to the problem of concept learning from examples. The algorithm is based on complementing of the available negative examples. The goal is to find a more compact representation of classification function and use it for further prediction of unknown cases. This is accomplished by an innovatory strategy for logical function minimization. The method can be applied in every domain where observations might be described by attributes with nominal values (valued in a finite set). Here we extend this approach to handle numerical data (valued in a linear interval) as well.
Interval Algorithms for Solving Minimal Spanning Tree and Shortest-Route Models
G. Gatev, A. Hossain
Abstract
Simple algorithms are presented for solving network models under parametric uncertainty. The new algorithms are applicable to the case when the generalized distance or the probability associated with each arc is nonnegative, interval, or real. The first three interval algorithms are developed on the base of midpoint and half-width representation of intervals, and these are more efficient than the interval algorithms that could be obtained by using traditional interval description. The fourth algorithm utilises the concept of interval possibility, and the interval product operator. The applicability of the result is demonstrated by considering several examples.
An Improvement of Functional State Local Models of Escherichia Coli MC4110 Fed-batch Cultivation
O. Roeva, S. Tzonkov
Abstract
This paper presents an improvement of functional state local models structures of Escherichia coli fed-batch cultivation. In the previous results it has been already shown how the cultivation process can be divided into functional states and how the model parameters can be obtained using genetic algorithms. The aim of this work is to find better local models structures of E. coli cultivation model based on Zhang investigations. The proposed modification of local models predicts very well the dynamics of the process variables biomass, substrate, acetate, dissolved oxygen as well as carbon dioxide. Moreover, the modified local models in general are simpler than the previous ones.