Ahmad Hasnah

Non-linear simulation of controller for longitudinal control augmentation system of F-16 using numerical approach

The non-linear simulation of controller for automatic flight control system is an essential part of the complete control design process. After using the linear approach in the design and simulation of the controller (control laws), it is important to simulate the controller on the non-linear real aircraft model. This simulation allows us to evaluate its functional performance during the real-time flight on the computer. The results of that process are used to determine whether the designed controller meets the design requirement and objective or not. The paper deals with application of numerical method fourth-order Runge Kutta to simulate the function of the longitudinal-CAS controller that is embedded into the non-linear, six degree of freedom model of fighter F-16. A computer program is made to implement the numerical approach and to illustrate the result of that simulation on the computer.

An approach for constructing complex discriminating surfaces based on Bayesian interference of the maximum entropy

In this paper we present a comprehensive Maximum Entropy (MaxEnt) procedure for the classification tasks. This MaxEnt is applied successfully to the problem of estimating the probability distribution function (pdf) of a class with a specific pattern. which is viewed as a probabilistic model handling the classification task. We propose an efficient algorithm allowing to construct a non-linear discriminating surfaces using the MaxEnt procedure. The experiments that we carried out shows the performance and the various advantages of our approach.

A neural network algorithm for hardware-software verification

Formal verification is the task of proving that a property holds for a model of a design. This paper examines the idea of a Neural Network-based algorithm used to find the set of states that makes a specification valid. The paper addresses a singular approach for those doing theoretical research for the verification of soft programs, and, for hardware designers. The approach of the application of the Artificial Neural Network is not new, but it becomes interesting if one can improve the truth-building efficiency by using some known artifices. Topics described include Integer Linear Programming, Propositional Logic, Model Checking, Satisfiability problems (SAT) and Artificial Neural Networks (ANN).

An efficient model checker based on the axiomatization of propositional temporal logic in rewriting logic

In this paper, we propose an efficient Model Checker for the Propositional Temporal Logic denoted by PTL. This logic is known to be well suited to verify electronic circuits and reactive systems. A typical verification problem consists of establishing formally a relationship between the specification of a software/hardware system and its implementation. In the sequel we show how a hardware designer should proceed to specify his design and prove its correctness using a PTL module under Maude System. A series of experiments have been conducted successfully on a well-known benchmark to prove the effectiveness of mixing temporal logic and rewriting logic techniques.