IRAQI JOURNAL OF COMPUTERS, COMMUNICATIONS, CONTROL AND SYSTEMS ENGINEERING

A nonlinear PI controller for a system with a mismatched and unknown external
disturbance is proposed in the present paper. A Sigmoidal function is proposed to be used
in the nonlinear integral element to attenuate the disturbance effect. Formally the controller
forces the state to a positively invariant set around the origin. As an application, the
nonlinear PI controller is used as a virtual controller for the upper subsystem in the frame of
Backstepping control approach for the DC motor system. In this Backstepping controller
design the restriction about the disturbance form in the upper system is relaxed. The results
show the effectiveness of the nonlinear PI controller to attenuate the effect of unknown and
varying disturbance torque and force the angular velocity to follow the desired reference.
Finally a reduced order observer is designed to estimate the armature current used in the
designed controller for the DC motor system.

This paper produces a developed method based on steganography techniques to prevent
intruders from obtaining the transmitted information. This work is based on a combination of
steganography and cryptography techniques to increase the level of security and to make the
system more complex to be defeated by attackers. The algorithm used for encryption is the
RC6 algorithm.
Two methods of hiding are used in this work: the first method is the Least Significant Bit
(LSB) and the second is the proposed and modified method used to hide bits in LSB of
iterated loop in brightness, red, green and blue of hiding image. The proposed method was
tested using standard objective measures such as Mean Square Error (MSE) and Peak Signal
to Noise Ratio (PSNR). A comparison between the two methods is performed.

This paper presents an application of a robust deadbeat controller for permanent magnet
stepper motor. This approach has been considered in order to assure robust stability and
performance (disturbance rejection, reference tracking) with the presence of system
parameters uncertainty. The Particle Swarm Optimization (PSO) is used to tune the
controller parameters by minimizing the cost function subject to H-infinity constraints.
It is shown that the designed deadbeat controller presents simple, low order, and robust
position control for a permanent magnet stepper motor. A two-phase motor is
considered in this paper.

The purpose of this paper is to present the modeling and simulation of a servo
hydraulic system. Hydraulic systems are broadly used in the industry due to their
ability to adapt to a wide range of purposes and their proven robustness. First we
develop a mathematical model to obtain the system responses. These responses
represent the response of the servo hydraulic system with a sinusoidal input.
Secondly, we design a PID controller in order to improve the position transient
response and reach minimum steady state error in the output displacement. Then, we
use Genetic Algorithm technique to find the best (Kp, K₁ and KD) gains for PID
controller to enhance the output of the servo hydraulic system.
The results have showed a higher improvement of the servo hydraulic system
response with minimizing the steady state error after using the PID controller gains
obtained from the Genetic Algorithm technique.

Security and data confidentiality has become a critical issue for companies and
individuals. In a classical client-server network, the access control management is
performed on the server, relying on the assumption that the server is a trusted party. In
this paper, we focus on how to strengthen the data confidentiality in client-server
networks. We address the improving data confidentiality protocol by using symmetric
encryption based on different secret keys for each session derived from timestamp. This
key is generated in a local machine and without needing to exchange it between the client
and the server for encryption decryption process. We have to use symmetric encryption
without exchanging the secret key to ensure the security of information from attacks for
eavesdropping or modification.

Speech recognition based on visual information such as the lip shape and its
movement is referred to as lip reading. The visual features are derived according to the
frame rate of the video sequence. The proposed work adopted in this paper is based
upon the lower part of the human face to extract the speaker sound relevant features
accurately and robustly from the inner edge of the lips and trace it acoustically to prove
its unique features and the possibility of merging it with sound features by measuring
their physiological or behavioral characteristics curves. The results were promising and
offered a good reaction: 94% - 100%.

Learning Classifier Systems (LCS), are a machine learning technique which
combines reinforcement learning, evolutionary computing and other heuristics to
produce adaptive systems. The system HRC (Human Rat Cheese) focuses in
creating artificial creature (Rat) using computer simulation, and learning it how to
choose between two different basic behaviors, (approach / escape) combining them to
perform complex behavior, which represents the final response in changing
environment.
The HRC is built of two-classifier subsystems working together, each
classifier system learns a simple behavior, and the system as a whole has as its learning
goal the control of activities. Flat architecture was used. The flat organization allows
distinguishing between two different learning activities: the learning of basic behavior
and the learning of switch behavior. One classifier system learns basic behavior,
(approach/escape), i.e., it is used to learn the simulated robot single step movement in
every direction in the environment. Whereas the other classifier system learns to control
the activities of basic classifier systems, i.e., it is used to learn to choose between basic
behaviors using suppression as a composition mechanism to chose between two basic
behaviors which represent complex behavior.
Simple experiments were executed for HRC: comparing and contrasting the
effect of the reinforcement learning using reward & punishment with learning using
reward only. Experiment results show that the run using reinforcement learning with
reward only is unable to perform as well as the run with reinforcement learning with
reward and punishment.

This paper proposes two important mathematical models related to network topology
which helps in computing some of the efficiency or reliability factors of communication
network as well as design purposes. Each of these models represents a topology
property. The first (second) of these models is used to compute the number of
appearances of any link (node) in the geodesics between nodes in a given network
topology, and so can be used to help in uniformly distributing the data flow through
links (nodes), as well as helping in measuring the degree of survivability of the network
in case of failure of some of its links (nodes). The two models have been developed
using “Graph Theory”, and so, giving the advantage of using the very wide range of
ideas, tools, and theorems of this field in case of developing other network topology
formulas based on the two models proposed in this paper.

In this paper, a new method for finding global optimal path planning is
proposed using a Genetic Algorithm (GA). A map of known static environment as well
as a start node and a target node connecting an optimal path which is required to be
found are given beforehand. The chosen nodes in a known static environment are
connected by sub-paths among each other. Each path is represented by a series of sub-
paths which connect the sequential nodes to form this path. Each sub-path radiating
from each node is labeled by an integer. The chromosome code of a path is a string of
series integers that represent the labels of sub-paths which are passed through traveling
from start node to target node. Two factors are integrated into a fitness function of the
proposed genetic algorithm: the feasibility of collision avoidance path and the shortest
distance of path. Two examples of known static environment maps are taken in this
study with different numbers of obstacles and nodes. Simulation results show the
effectiveness and feasibility of the proposed GA using sub-path codes to find optimum
path planning for mobile robot.

In the quest for digitizing the synchro, this paper proposes a method for processing
the synchro format voltages to extract a single reliable shaft angle reading. This
eliminates sophisticated and expensive electronic parts and replaces them by
software algorithms. Kalman estimation techniques are applied to improve sensor
dynamic response, precision and efficiency. Fuzzy logic techniques are used to speed
up the estimation process so that the time taken to produce a result is within the time
of half a cycle of the excitation frequency (less than 1.25 ms). Wavelet techniques
are also used to improve the accuracy much further. The synchro digitizer was
simulated using Matlab, and the random noise was taken into effect. Theoretical
analysis and experimental data ascertained the above technique.