The signal and data processor in the Pave Pace architecture, the Pace Pace Integrated Core Processor (ICP), is described. Technology and architectural innovations were primarily used to improve reliability, maintainab...
详细信息
The signal and data processor in the Pave Pace architecture, the Pace Pace Integrated Core Processor (ICP), is described. Technology and architectural innovations were primarily used to improve reliability, maintainability, survivability, affordability, weight, and volume. In addition, turn-of-the-century avionics applications require that the ICP must be able to provide >20 GLOPS and >750 MIPS. The ICP is a modular, logically integrated multiprocessor where components may be physically distributed throughout a platform. To achieve this, large Pave Pace electrical backplanes are replaced by small electrical/photonic backplanes and a photonic exchange network, interconnecting the smaller backplanes via fiber optics. The ICP operating system and architecture provide a uniform logical view to the application programmer, improving programmability and survivability. The Principles of Operation (POPS) is a critical component of the ICP design. The ICP POPS provides the open software architecture required for an advanced avionics processor. It synchronizes and controls the large number of autonomous processors to gain the maximum benefit from a highly concurrent architecture.< >
The authors describe the development of an artificial intelligence (AI) model. XTEL 2.0 is an architecture solution system that recommends specific node hardening and protection, specific link protection and transitio...
详细信息
ISBN:
(纸本)0879426918
The authors describe the development of an artificial intelligence (AI) model. XTEL 2.0 is an architecture solution system that recommends specific node hardening and protection, specific link protection and transition, and specific leasing with allied host nation systems. Where users enter desired architecture preferences, XTEL 2.0 generates feasible transition states for network components using a rule base, and expresses them as variables in mathematical equations that represent preference boundaries and architecture constraints. Because these mathematics equations are simple linear equations, they are solved using a linear programming framework so that a quick architecture resource allocation can be performed.
Adaptive k-means clustering algorithms have been used in several artificial neural networkarchitectures, such as radial basis function networks or feature-map classifiers, for a competitive partitioning of the input ...
详细信息
ISBN:
(纸本)0780301641
Adaptive k-means clustering algorithms have been used in several artificial neural networkarchitectures, such as radial basis function networks or feature-map classifiers, for a competitive partitioning of the input domain. The authors present a modification of the traditional k-means algorithm. This approach approximates an optimal clustering solution with an adaptive learning rate, which renders it usable even in situations where the statistics of the problem task slowly vary with time. Simulations comparing this improved adaptive k-means algorithm with other k-means variants are presented.
The author examines the recent efforts of the ANSI T1M1.5 working group and T1M1 committee to apply object-oriented (O-O) data modeling techniques to the development of standards for the operation, administration, mai...
详细信息
ISBN:
(纸本)0879426977
The author examines the recent efforts of the ANSI T1M1.5 working group and T1M1 committee to apply object-oriented (O-O) data modeling techniques to the development of standards for the operation, administration, maintenance, and provisioning (OAM&P) of telecommunications networks (TNs). The use of these state-of-the-art data modeling techniques is launching the North American Telecommunications network into an open, standard, multi-vendor communications environment. The author presents a description of the organization of the T1M1 standards body, a brief overview of the Telecommunications Management network (TMN) model, the basic principles of O-O data modeling, a description of recent network management standardization efforts in T1M1 that use these techniques, and a brief description of current T1M1 work.
A novel design technique for asynchronous binary neural networks is proposed. This design uses linear programming to design two architectures: i) a fully connected network that reads a N-digit cue and classifies it in...
详细信息
A novel design technique for asynchronous binary neural networks is proposed. This design uses linear programming to design two architectures: i) a fully connected network that reads a N-digit cue and classifies it into a category represented by a N-digit pattern;and ii) a two-layer network (with lateral connections) that has M neurons in the first layer and L neurons in the second layer;the network reads an M-digit cue to the first layer and associates it with a second-layer L-digit pattern. In both cases, the objective function is a weighted sum of the number of errors that can be corrected by the network. A cue with this number of errors (or fewer) is guaranteed to converge to the correct pattern. An economical VLSI realization of the designed networks can be easily accomplished.
A series of simulations comparing different hypercube architectures for application in a network configuration of personal computers (PCs) is described. The study was done using network II.5, a simulation package for ...
详细信息
A series of simulations comparing different hypercube architectures for application in a network configuration of personal computers (PCs) is described. The study was done using network II.5, a simulation package for networking and processing, to simulate various levels of hypercube architectures. The data are used to analyze the various levels of hypercube programming (N = 1 to 4) for comparison with a single- and multiple-user microcomputer. Implementation includes an application communications program running on a PC that uses message passing as a form of control between other computers in the hypercube network.
The authors propose a programming system that combines pattern matching of Prolog with a novel approach to logic and the control of resolution. A network of nodes and arcs together with a three-valued logic is used to...
详细信息
ISBN:
(纸本)0818620846
The authors propose a programming system that combines pattern matching of Prolog with a novel approach to logic and the control of resolution. A network of nodes and arcs together with a three-valued logic is used to indicate the connections between predicates and their consequents, and to express the flow from facts and propositions of a theory to its theorems. In this way, one can handle uncertainty and negation properly in this 'neural logic network.' A neural logic program consists of a specification of network fragments, labeled with predicates and arc weights, and they can be joined dynamically to form a tree of reasoning chains. The architecture of the neural logic computational model is left open and the authors do not intend the model to be interpreted literally as a physical architecture.
An NOS implemented on Texas Instrument's Mission Computer is described. The NOS is targeted to advanced common module architectures, and supports distributed Ada applications by implementing an intermodule messagi...
详细信息
An NOS implemented on Texas Instrument's Mission Computer is described. The NOS is targeted to advanced common module architectures, and supports distributed Ada applications by implementing an intermodule messaging scheme. In addition, the NOS contains provisions which support the concept of run-time reconfiguration in order to more easily allow full use of available hardware resources in the event of mission mode changes or degraded mode operation. The NOS has evolved from a VHSIC Phase I IBUS implementation to the current VHSIC Phase II PI-Bus target implementation. Improvements were made in this evolution due to redesign based on the first implementation and due to superior PI-Bus hardware support.
The Entity System, a model for large-scale storage and manipulation of data using object-based programming techniques, is described. The model uses persistence to store arbitrarily complex objects with minimal program...
详细信息
The Entity System, a model for large-scale storage and manipulation of data using object-based programming techniques, is described. The model uses persistence to store arbitrarily complex objects with minimal programming effort and supports smooth evolution of object types and representations. The design and implementation of a distributed Entity System that allows location-independent access to objects by client programs on a network are described. The architecture of the system is designed to scale to networks and object bases of global proportions.
FLAME is an experimental language for distributed programming. It is intended for applications that run on a collection of computers connected by a network. It is part of a project that deals with the building of open...
详细信息
ISBN:
(纸本)0818620366
FLAME is an experimental language for distributed programming. It is intended for applications that run on a collection of computers connected by a network. It is part of a project that deals with the building of open software systems. open systems place stringent requirements on both the type of language one uses and the style of programming. FLAME tries to address the difficulties of writing such open, distributed applications with current technology. FLAME is based on C++, and its main contribution is to integrate the notions of objects and process groups at the programming language level. The result is a powerful language which, nevertheless, raises several new questions. The current state of FLAME is discussed, and potential areas for future development are outlined.
暂无评论