ENSE 623 Projects, Fall Semester, 2004

[ Project 1 ]: Using LTSA to Validate Aiport Hangar Operations
[ Project 2 ]: Validation and Verification Plan for Answering Machine Product Line
[ Project 3 ]: Dynamic Requirements and Test Suite Trade-off Tool
[ Project 4 ]: Validation/Verification of Augmented Cognition Systems with Modular or Platform-Based Design
[ Project 5 ]: Validation, Verification, and Behavior Modeling of a Canal System
[ Project 6 ]: Using Higraphs to Map Requirements to Behaviors and Components
[ Project 7 ]: Part 1: Washington Metro; Part 2: Lowjack System
[ Project 8 ]: Automated Generation of High Integrity Test Suites
[ Project 9 ]: Verification of a Fishery Data Collection System with UPPAAL
[ Project 10 ]: Modeling High Integrity Aspects using Role Models

Class Presenatations will be on Nov. 23 and 30. Don't worry if your project isn't complete -- just talk about your main ideas, what you are hoping to accomplish and the method. Plan for 20 minutes + 10 minutes of questions/discussion. The purpose is to share ideas and receive constructive feedback.


PROJECT 1

Title: Using LTSA to Validate Aiport Hangar Operations
Authors: Elizabeth Lee and Alex Lowe

Abstract: In our project, we will study the LTSA to find ways to do verification and validation. We will be studying this on a general level for all systems but the one system we will use as an example will be a system of an airplane hangar. There will be a runway and a plane parking spot. The parking spot is only accessible via the runway and only one plane can be on the runway at a time. Also, a gate for the runway must open for take off and landing signifying all crews on the runway are ready. This simple system will run on LTSA as an animation as a sample system for verification and validation.

Class Presentation: November 23
Final Report: Due December 8, 2004


PROJECT 2

Title: Validation and Verification Plan for the Design of an Answering Machine Product Line
Authors: Elizabeth Koza, Rocio Salas-Lopez, Vanessa Virtudazo

Abstract: During this semester, we will expand on our ENPM 642 project titled "A Systems Engineering Approach to the Design of an Answering Machine Product Line" and develop a complete Validation and Verification (V&V) plan. We will start by grouping the project into layers characterized by levels of abstraction to keep the complexity of design at check, while identifying unintended implied conditions of the system. Our V&V plan will encompass all aspects of the design process and the connection between requirements, verification methods, verification requirements, and levels of application. Relationships will be provided in a matrix format. Lastly, we will provide recommendations for future work that will focus on streamlining and simplifying SE methodologies studied in ENPM 641, ENPM 642 and ENPM643.

Final Report: Answering Machine Plaform ( doc ) ( pdf )


PROJECT 3

Title: Dynamic Requirements and Test Suite Trade-off Tool
Authors: Fred Faber and Julie McNeil

Abstract: Our project is an extension of our 622 project, a software tool that enabled a project team to trade off time and costs in order to select an ideal set of inter-dependent features (requirements) for a system. This semester, we plan on extending the use of the software from the requirements phase to the verification/validation phase of system design.

We will provide a tool that enables testers to select an optimal test suite for a system. A test suite is a set of test cases, with each test case covering at least one requirement. Given a multitude of possible test cases, the test manager must determine which cases will be included in the final test suite. Due to time and money constraints, all test cases cannot be selected.

The tool will use quantified criteria established by the test designers in order to trade-off the costs and benefits of each test case. Examples of these criteria include: The requirements coverage, the cost to run, the time to run, a prediction of fault detection, a demolition factor, and a completeness factor. Each test case will also be scaled by the importance of the requirements it tests. The tool will use a multi-objective model to analyze the trade-offs of minimizing the size of the test suite while maximizing the aggregate error detection rate.

Final Report: Test Suite Analysis: Minimization and Ordering ( pdf )


PROJECT 4

Title: Validation/Verification of Augmented Cognition Systems with Modular or Platform-Based Design
Authors: Latosha Marshall and Colby Raley

Abstract: Modular and platform-based design principles can be used to allow the quick and easy creation of custom Augmented Cognition systems from customer specifications. A modular design enables a higher product flexibility and reduction of development time, parallel development of system components, reduction of production time, reduced capital investment in production, reduced material and purchase costs, improved quality, easier service and upgrading, and easier administration of any system (Ericsson, 1999). Platform-based design takes that one step further, and allows systems to be fully developed, designed, validated, and verified while still at a high level of abstraction in the design process.

Platform-based design is important in almost any high-technology system, but is particularly relevant to augmented cognition systems for two major reasons:


  1. Technologies required to implement augmented cognition systems are still being developed, and will continue to improve dramatically over the coming years - for example, sensing technologies will never catch up with the knowledge of architecture design, so a system should be able to accommodate this.
  2. Augmented cognition systems will continuously be applied to new scenarios and situations; therefore, the systems should be able to be adapted to any necessary environment.

This paper will discuss the migration from modular design to platform-based design in augmented cognition systems, will show the methods of validation and verification of these systems using platform-based design, and will make an argument for the use of platform-based design in all such systmes.

References

Ericsson, A. and G. Erixon. Controlling Design Variables: Modular Product Platforms. ASME Press, New York, NY, 1999.

Class Presentation: Platform-Based Design of Augmented Cognition Systems ( pdf )
Final Report: Platform-Based Design of Augmented Cognition Systems ( pdf )


PROJECT 5

Title: Validation, Verification, and Behavior Modeling of a Canal System
Authors: Noosha Haghani and Nazanin Alborni

Abstract: This project involves developing a model of a canal system very similar to the Panama Canal. A detailed set of requirements shall be generated. We shall then use the Unified Modeling Language (UML) to design the initial model of the system. Afterwards, the LTSA Tool shall be used to verify the UML model using its sequence diagram and state diagram capabilities. In the end, using the Extreme Markup Language (XML) and the LTSA Tool, and animation of the canal shall be created to verify and validate the model. An attempt shall be made to use the UPPAAL tool for modeling behavior, validation, and verification.

Final Report: Validation, Verification, and Behavior Modeling of a Canal System ( doc ) ( pdf )


PROJECT 6

Title: Using Higraphs to Map Requirements to Behaviors and Components
Authors: Chad Rivera and Kevin Fogarty

Abstract: As the complexity of today's systems continues to increase (with individual components having mechanical, thermal, electrical, and logical requirements) there is a need to clearly represent the multiple "domains" that must be considered when designing, testing, and very importantly, upgrading systems. Standard visual representations such as UML can fairly accurately define system behavior and structure, and can be used to map the two together.

However, there is little consensus on a visual formalism that will map requirements (of multiple domains) to behaviors and components. We propose using higraphs, a general kind of diagramming object that forms a visual a visual formalism of topological nature" to accomplish this objective. A higraph uses features of hypergraphs and Euler-Venn diagrams to represent system components, and the relationships (mappings, interactions, etc.) between them.

One practical application of an accurate system higraph would be its use in determining the best way to upgrade a system to meet new requirements. The higraph would provide a visual way to map new requirements to existing components, and would indicate what component specifications would be impacted. The higraph would also allow for the conclusion that no existing component could satisfy the new requirement(s), and a new system component would have to be introduced.

Our project for this semester will focus on creating higraph(s) for existing, defined, systems, and documenting the procedure and logic of how the higraph(s) could be used to allocate one or more requirement(s) to the existing system.

Class Presentation: November 30
Final Report: Due December 7, 2004


PROJECT 7

Title 1: Washington Metro System Safety Improvement
Author: Albert Anoubon Momo

Abstract 1: The purpose of this project is to plan, analyze, and design an automated system to provide "real time" emergency management capabilities for the Washington DC Metro System. The case study will focus specifically on developing an information system to assist in Metro evacuation, emergency response coordination in the event of an emergency affecting the Metro system. This system will monitor train traffic, train locations, train speeds, train directions and estimated numbers of passengers and continuously compare train information to predetermined evacuation routes and provide real-time updates to the evacuation plans. In scope this semester is the use of UPPAAL for validation and verification of the real-time system.


PROJECT 8

Title: Automated Generation of High Integrity Test Suites
Author: Ron Henry

Abstract: I will investigate some of the automated V&V concepts and techniques described in Simon Burton's thesis "Automated Generation of High Integrity Test Suites from Graphical Specifications." This will be done by developing graphical specifications using finite state machines (FSMs) or Statecharts. The chosen domain is flight software control of COS, a scientific instrument on the Hubble Space Telescope. Required capabilities such as automated target acquisition and bright object protection are reactive in nature and well suited to state-based modeling. Using available tools, this graphical model will be automatically validated for properties such as completeness, determinism, and satisfaction of safety constraints. If suitable tools can be obtained, an attempt will also be made to generate test cases automatically from the specification.

Class Presentation: ( ppt )
Final Report: System Validation and Verification using SDL ( doc ) ( pdf )


PROJECT 9

Title: Verification of a Fishery Data Collection System with UPPAAL
Authors: Jonathan Eser and Noriaki Suzuki

Abstract: We aim to analyze our fishery data collection system with UPPAAL. System requirements generated during earlier stages of the system design will be provided to UPPAAL along with a finite state automata representation of our system. We will determine by using logical queries to the system whether our requirements are indeed met by our system. Temporal logic will be required to properly demonstrate that time varying requirements are supported by the system design. Previous work relied upon the LTSA tool to identify problems associated with concurrent behavior in the system and deadlocked behavior. Data collection systems are based on timed automata and UPPAAL has been developed to represent timed automata, finite state-machines with clocks. In other words, the tool is essentially LTSA with the incorporation of timed events. We believe that UPPAAL will model our system well. Verification through use of UPPAAL and temporal logic will be a suitable conclusion to our system analysis. UPPAAL has been developed by the Basic Research in Computer Science at Aalborg University (AAL) in Denmark and the Department of Computer Systems (DoCS) at Uppsala University (UPP) in Sweden

Class Presentation: November 23
Final Report: ( doc )


PROJECT 10

Title: Modeling High Integrity Aspects using Role Models
Authors: Gokul Rathinasamy and Pampa Mondal

Abstract: Different integrity mechanisms can be modeled as aspects (e.g., authentication aspect, data sensitivity aspect). In our project, we focus on specifying an authentication aspect. Authentication is the process of verifying that the identity claimed by a user is his true identity.

HCMS - Engineering aspect oriented system
       Requirements Engineering
       Specification
       Design
       Evolution
       Aspect Interaction
       Aspect Oriented Design with UML
       Extending UML with Aspects Profile
       Stereotypes
       Constraints
       Tagged values
       Aspects at architecture level (Architecture Design Language)
       Expressing Aspects Using UML Behavioral and structural Diagrams
       Concern Modeling
       Aspect-Oriented Dependency Manangement

Also we are trying to use UPPAAL to do the verification and validation of the identification module.

Class Presentation: November 30
Final Report: Due December 7, 2004


Developed in October 2004 by Mark Austin
Copyright © 2004, Institute for Systems Research, University of Maryland