Implementation Tradeoffs Research Articles

Reducing time delay of ARQ protocols by simple error-correcting codes

Time delay analysis is carried out for type-I hybrid ARQ with BCH codes. It is found that the time delay of ARQ protocols can be significantly reduced by using simple BCH codes. Implementation trade-off can then be made between buffer size and simple decoding facilities to achieve maximum performance and minimum complexity.

Causal Functionally Models For Graphical User Interfaces

When building user interfaces to AI applications for a complex system, there is an often-overlooked knowledge engineering effort to capture an expert's view of the system, and to detennine the abstraction of data that will be presented to users. This knowledge is generally mixed into the interface or graphics code, and is subsequently lost when other applications are developed for the same complex system, or in the event of porting the application to another platform. Few methods exist for representing the underlying structure of an expert’s view of a system that originally indicated the content to be presented by an interface. It is desirable to capture and explicitly represent this knowledge, so that it can be used by a vuriety of interface processes and accessed in different ways to provide appropriate views of the system for new application. We present a technique for using a causal model to represent expert knowledge about system functionality and behavior. Our approach extends previous work in causal reasoning by mapping the behavior of a system at the usual structural level to higher-level functional descriptions of what the system is actually doing. In this paper, we discuss the causal-modeling approach to user interfaces and demonstrate the applicability of the approach with an example system. Practical implementation trade-offs and considerations are discussed.

Hashed and hierarchical timing wheels: data structures for the efficient implementation of a timer facility

Conventional algorithms to implement an Operating System timer module take Ο ( n ) time to start or maintain a timer, where n is the number of outstanding timers: this is expensive for large n . This paper begins by exploring the relationship between timer algorithms, time flow mechanisms used in discrete event simulations, and sorting techniques. Next a timer algorithm for small timer intervals is presented that is similar to the timing wheel technique used in logic simulators. By using a circular buffer or timing wheel, it takes Ο (1) time to start, stop, and maintain timers within the range of the wheel. Two extensions for larger values of the interval are described. In the first, the timer interval is hashed into a slot on the timing wheel. In the second, a hierarchy of timing wheels with different granularities is used to span a greater range of intervals. The performance of these two schemes and various implementation trade-offs are discussed.

Resolution of range and doppler ambiguities in medium PRF radars in multiple-target environment

In medium pulse-repetition frequency (PRF) radars, ambiguities may arise in both range and Doppler measurements. Efficient techniques have been established (Skolnik, 1970; Hovanessian, 1976) to resolve the range ambiguity of a single isolated target using multiple PRFs. In this paper, we propose a simple algorithm to resolve the Doppler ambiguity using the discrete Fourier transform (DFT) output of two PRFs. A condition on the relative values of the two PRFs is derived to account for the errors due to the finite bandwidths of DFT filters. Next, we describe a method to resolve both Doppler and range ambiguities in a multiple-target environment. The method uses three PRFs to resolve Doppler ambiguities and to identify the declarations of a particular target in different PRFs. Range ambiguities are then resolved in a straightforward manner. A fourth PRF is made use of to extract blind-speed targets. The proposed method is computationally efficient and can be used even when ambiguous returns from several targets are received. Implementation trade-offs are also discussed.

Moval: A framework for turning digital signal processing algorithms into custom chips

A framework supported by a set of tools for turning digital signal processing algorithms into custom chips is proposed. The framework, called MOVAL, integrates analysis, layout synthesis and validation and is based on a structured top-down design methodology covering seven succinct abstraction levels: the behavioral, data representation, space/time, hardware, symbolic and geometric (mask) descriptions, and the chip. The top four levels efficiently cope with the implementation trade-offs and are all written in the same high level language, currently “C”. As a result, the design can be modeled using a mixture of components defined at different abstraction levels. This allows an efficient mixed-mode multi-level validation. The hardware description is unambiguous and is the key to (semi-) automatic synthesis of the layout. Furthermore, it generates test vectors for the symbolic layout, the mask and, finally, the chip. The validation is done automatically, based on back substitution. As an example of the proposed design methodology, the crucial steps of the implementation of a fast Fourier cosine transform algorithm are described. Ein Entwicklungskonzept mit Werkzeugen wird vorgeschlagen um Signalverarbeitungsalgorithmen in spezialisierte Chips umzuwandeln. Das Konzept, MOVAL genannt, integriert die Analyse, die Synthese und die Verifikation, und ist eine strukturierte ‘top-down’ Methodologie mit sieben Abstraktionsniveaus: Algorithmus, Datenpräzision, Raum/Zeit, Hardware, symbolisches Layout, Masken, und Chip. Die vier oberen Niveaus erlauben es, die verschiedenen Kompromisse einer Implementation zu erforschen, und sie sind in der gleichen hohen Sprache, “C”, geschrieben. Deshalb kann der Design mit Komponenten modelisiert werden die auf verschiedenen Niveaus geschrieben sind. Dies erlaubt eine effiziente, ‘mixed-mode, multi-level’ Verifikation. Die hardware Beschreibung ist eindeutig und erlaubt eine (semi-) automatische Layout-Synthese. Damit können auch Testvektoren generiert werden, und dies sowohl für das symbolische Layout wie für die Masken und das Chip. Die Verifikation ist automatisch und basiert auf die Rückwärts-Substitution. Als Beispiel für diese Design Methodologie werden die wichtigen Etappen des Designs eines ‘fast cosine transform’ Chips gezeigt. Dans cette contribution, on propose un contexte de développement et certains outils afin de transformer des algorithmes de traitement du signal en circuits intégrés spécialisés. Ce contexte de développement, appelé MOVAL, intègre l'analyse, la synthèse du schéma ainsi que la validation. Il est basé sur une méthodologie ‘top-down’ structurée et faisant appel à sept niveaux d'abstraction: le niveau algorithmique, de précision finie, d'espace/temps, de matériel, de schéma symbolique, de masque, et finalement le niveau physique (circuit). Les quatre niveaux supérieurs permettent d'explorer efficacement les compromis possibles lors de l'implantation et sont tous quatre écrits dans un langag commun de haut niveau, le “C”. Ainsi, on peut simuler un circuit en utilisant des éléments décrits à des niveaux d'abstraction différents. Ceci permet une validation en modes et niveaux mélangés. La description matérielle est non ambiguë et permet une génération (semi-) automatique du schéma symbolique. De surcroît, elle permet la génération de vecteurs de test, tant pour le schéma symbolique que pour le plan de masques ou le circuit physique. La validation est faite automatiquement par voie de substitution. Comme exemple, on montre les étapes principales du développement d'un circuit réalisant une transformation en cosinus rapide.

A system to make visible the structure and execution of student programs

Visual models help to convey programming insights and computer science principles. This paper describes the design criteria for a software system which automatically constructs visual models as a program runs. Line-by-line execution and data structures are dynamically updated on a display screen. The system can be used either by an instructor in class demonstrations, or by the student in running his or her own programs. Motivations for the design criteria are discussed, along with tradeoffs in implementation. A system has been developed for the Pascal programming language, running on a variety of popular microcomputers.

Window Design for a Professional Workstation in the Office Environment

This presentation describes the development of the user interface for a windowing system implemented on a multi-tasking professional workstation used in the office environment. The project was completed under severe schedule restrictions which impacted the implementation tradeoffs for user interface design. The functional and user interface goals for the windowing system were as follows: • The user should be able to subdivide the screen with a different application running in each window. • The user should be able to quickly and easily move between open windows/applications. • Operation of an application within a window should be the same as when the application is full screen. • The user should be able to quickly alternate between a partial view and a full-screen view of an application.

Implementation trade-offs for a short-headway vehicle-follower automated transit system

An important aspect in the design of automated guideway transit (AGT) systems is the communication requirements for the longitudinal control of vehicles, that is, the requirements to measure vehicle states, transmit this information between vehicles, and subsequently control the individual vehicles. This paper examines the trade-offs in the data rates, word lengths, and allowable transmission time lags needed to maintain an acceptable level of vehicle performance, and how these requirements relate to the allocation of control computation between vehicle- and wayside-based computers. A nonlinear control law designed for short-headway vehicle-follower systems provides a baseline to examine these trade-offs. The specification of the parmeters entering the problem is approached through analytically derived results, using simplified linear models with verification and expansion of these results through a detailed simulation.

Partial differential equations and finite difference methods in image processing--Part II: Image restoration

Application of Partial Differential Equation (PDE) models for restoration of noisy images is considered. The hyperbolic, parabolic, and elliptic classes of PDE's yield recursive, semirecursive, and nonrecursive filtering algorithms. The two-dimensional recursive filter is equivalent to solving two sets of filtering equations, one along the horizontal direction and other along the vertical direction. The semirecursive filter can be implemented by first transforming the image data along one of its dimensions, say Column, and then recursive filtering along each row independently. The nonrecursive filter leads to Fourier domain Wiener filtering type transform domain algorithm. Comparisons of the different PDE model filters are made by implementing them on actual image data. Performances of these filters are also compared with Fourier Wiener filtering and spatial averaging methods. Performance bounds based on PDE model theory are calculated and implementation tradeoffs of different algorithms are discussed.

The GPS Navigation Message

The gps users continuously receive navigation information from the GPS Space Vehicles in the form of data bits modulated on the received signals. This information, which is computed and controlled by the GPS Control Segment, includes the vehicle's time, its clock correction and ephemeris parameters, almanacs and health for all GPS Space Vehicles, and text messages. From this information the Users compute the Space Vehicle's precise position and clock offset and less precise positions and clock offsets of Space Vehicles yet to be acquired. The GPS Navigation Message design process included numerous trade studies which weighed various representations and algorithms against variables such as message size, accuracy, update frequency, User computational requirements, and legacy for the operational GPS. Other factors such as graceful degradation and future User requirements were also considered. Finally, upon selecting the appropriate design structure, the design was fine tuned to its final form and User algorithm implementation trade–offs were performed. The representation algorithms and User algorithms were jointly tested using a simulated Space Vehicle ephemeris trajectory and Space Vehicle clock. The results of these tests demonstrate that the User models represent the simulated ephemeris and clock to within 0.01 meters with precise parameters, and to within 0.1 meters with truncated parameters.

Partial differential equations and finite-difference methods in image processing, part 1: Image representation

Application of Partial Differential Equation (PDE) models for restoration of noisy images is considered. The hyperbolic, parabolic, and elliptic classes of PDE's yield recursive, semirecursive, and nonrecursive filtering algorithms. The two-dimensional recursive filter is equivalent to solving two sets of filtering equations, one along the horizontal direction and other along the vertical direction. The semirecursive filter can be implemented by first transforming the image data along one of its dimensions, say Column, and then recursive filtering along each row independently. The nonrecursive filter leads to Fourier domain Wiener filtering type transform domain algorithm. Comparisons of the different PDE model filters are made by implementing them on actual image data. Performances of these filters are also compared with Fourier Wiener filtering and spatial averaging methods. Performance bounds based on PDE model theory are calculated and implementation tradeoffs of different algorithms are discussed.

A Diagnostic Emulator for HEAO software development

Diagnostic Emulation is the application of microprogramming to the emulation of an operational computer to support software development and verification for that computer. A conventional technique, Interpretive Computer Simulation (ICS), has been used for many years in support of such software development and verification efforts. The ICS method is becoming less cost effective. For the development of attitude control software for NASA's High Energy Astronomical Observatory (HEAO) diagnostic emulation was considered as an alternative. This paper describes the motivation for using an ICS or diagnostic emulator to validate command and control software, the potential advantages of microprogramming, and the analysis which led to the development of the HEAO Diagnostic Emulator rather than the use of conventional techniques. Also discussed are the emulator implementation tradeoff analysis, the resulting program design, and the interface between the microprogram and the assembly language modules. The results to date are described, including the capability, speed, and reliability of the program and initial user response.

Standardizable low-pass voice band hybrid circuit filters for pulse code modulation (PCM) communication systems

A standardizable design approach to realize the voice band low-pass filters for a variety of pulse code modulation (PCM) communications systems is described. This design concept incorporates cascaded RC active filter sections, thick and thin film hybrid technologies, and functional tuning to allow the use of the same hybrid design for both transmit and receive filters, for a range of input and output requirements. Tuning, sensitivity analyses, and implementation tradeoffs are described. Thick film implementations, using NPO chip capacitors, with the new high stability resistor inks as well as thin film implementation, using the usual tantalum integrated thin film technology are presented. Very stringent requirements have been successfully met using state-of-the-art hybrid technologies.