https://ir.unisa.ac.za/handle/10500/23875 2026-05-07T12:10:27Z https://ir.unisa.ac.za/handle/10500/23924 Finding regular paths in acyclic graphs Wood, PT Increasing the expressive power of relational query languages by providing some form of recursion is currently a topic of much research. For many recursive queries in relational databases, a relation can be represented as labelled directed graph G, while the query can be viewed as finding simple paths in G that satisfy a given regular expression. Based on such a query language, it has been shown recently that, in this context, the general query evaluation problem on cyclic graphs is intractable. In this paper, we present an efficient algorithm for the case in which the graphs are acyclic. 1990-01-01T00:00:00Z https://ir.unisa.ac.za/handle/10500/23923 The NRDNIX distributed database management system Rennhackkamp, MH A distributed database management system must facilitate integrated processing of data which is physically distributed over a network of node computers, while providing locality transparency. The NRDNIX prototype has locality, logical and physical independence as goals, together with adequate availability and efficient throughput. It is architecturally described in terms of an integration of the ANSI/X3/SPARC database management system and the ISO interconnection reference models. Communication takes place via a broadcast network. The data is horizontally distributed over the horizontally controlled nodes, with duplication according to usage. An adaption of the entity-relationship data model is supported, with extensions to the relationships to include update dependencies. Some supporting operating systems have been considered. The most promising approach seemed to be the embedding of the distributed database management system kernel in the MINIX operating system. This approach accentuated shortcomings of the MINIX operating system, which deemed it inappropriate for the implementation. A more feasible and popular implementation currently being used is the development of device drivers for the XENIX operating system, for example for the network interface. Although the NRDNIX prototype is nearing completion, many related research projects must still be undertaken to complete it into a product. 1990-01-01T00:00:00Z https://ir.unisa.ac.za/handle/10500/23922 The use of a lattice for fast pattern matching Oosthuizen, GD Pattern recognition involves the matching of a given pattern against a collection of patterns in store in order to identify the given pattern as belonging to a particular class already known to the system or not. Similarly, data retrieval from content addressable storage involves the matching of a partial pattern against a collection of patterns in store in order to retrieve the required pattern. This matching procedure can be made very efficient by integrating stored patterns into a graphical structure called a lattice. The lattice forms an optimal indexing mechanism and supports fast parallel matching. The indexing mechanism is optimal because it is both complete and minimal: all possible indices to a particular pattern are generated, and indices are integrated maximally, i.e. paths to similar patterns overlap as much as possible and thereby prevent duplication. Moreover, the indexing mechanism is robust: if faulty or partial information is supplied, the lattice is guaranteed to provide the best approximate answer. 1990-01-01T00:00:00Z https://ir.unisa.ac.za/handle/10500/23921 Image reconstruction via the Hartley transform Murrell, HC; Carson, D The continuous and discrete Hartley transforms are real valued transforms that have similar properties to the continuous and discrete Fourier transforms. In addition, a fast algorithm exists for computing the discrete Hartley transform which is faster than the fast Fourier transform, even when the fast Fourier transform is optimized for dealing with real data. In this paper the authors apply the Hartley transform to the problem of image reconstruction. The authors will show that the projection-slice theorem and the filtered back-projection algorithm can be derived using the Hartley transform and that the filter part of the filtered back-projection algorithm can be implemented using Bracewell's fast Hartley transform. 1990-01-01T00:00:00Z