Underwater acoustic transients from man-made structures and biologics are very rich in structure and detail, very diverse in terms of duration, highly nonstationary, and often mixed with multipath. Orincon has an ongo...
详细信息
It is proposed that existing neurobiological data on the properties of single nerve cells and of the systems formed by them will allow, when reproduced in models, a new paradigm of the functioning of neuronal nets, ba...
详细信息
The authors explain the architecture, modes, and applications of a parallel processor based on a neural net structure. This processor is designed as a coprocessor for the IBM PC/XT. LOCON 9B51 is intended to accelerat...
详细信息
The authors explain the architecture, modes, and applications of a parallel processor based on a neural net structure. This processor is designed as a coprocessor for the IBM PC/XT. LOCON 9B51 is intended to accelerate sorting, large-scaledataanalysis, solution of equations of mathematical physics, computational geometry, text processing, discrete optimization, computer-aided design, and complex system simulations. The implementation of some parallel algorithms for NP-complete problems on this coprocessor is shown.< >
Two systems are introduced that increase the information density of textual presentation by reconsidering text as pictures, expanding the range of written expression. Both schemes indicate nested associativity and bot...
详细信息
Two systems are introduced that increase the information density of textual presentation by reconsidering text as pictures, expanding the range of written expression. Both schemes indicate nested associativity and both employ stripes, but in different styles: Blush uses large-scale vertical gutters, superimposed as reverse-fielding on indented outlines of computer programs; Zebrackets uses small-scale horizontal striations, superimposed on parenthetical delimiters. These systems are implemented as computer programs, active filters that represent textual information graphically.< >
A time-domain architecture for real-time processing of SAR (synthetic aperture radar) data using lower data rates and dedicated hardware is discussed. The architecture uses a novel correlator for processing range and ...
详细信息
A time-domain architecture for real-time processing of SAR (synthetic aperture radar) data using lower data rates and dedicated hardware is discussed. The architecture uses a novel correlator for processing range and azimuth data. In compressing the range correlated data in the azimuth direction, no explicit range migration correction is employed. A data reduction rate of 8 is achieved in the presummer circuit, which enables real-time processing with moderately high-speed multipliers. With high speed multipliers and lower data rates, images like those from Seasat can be processed in real time. For the data rates proposed, multipliers operating at 6.25 MFLOPS are sufficient. Since the architecture uses CMOS VLSI chips and a minimum amount of memory, the size and power consumption are expected to be low and hence ideally suited for spaceborne applications.< >
It is proposed that existing neurobiological data on the properties of single nerve cells and of the systems formed by them will allow, when reproduced in models, a new paradigm of the functioning of neuronal nets, ba...
详细信息
It is proposed that existing neurobiological data on the properties of single nerve cells and of the systems formed by them will allow, when reproduced in models, a new paradigm of the functioning of neuronal nets, based on the inner time scale concept to be introduced. The concept of inner (physiological) time provides a new approach to modeling processes in neuronal networks. Different inner time scales may be associated with different coordinate systems, and decomposition into coordinates may allow one to extract various invariant measures from the input information. It is noted that in the case of the description of neuronal network behavior in terms of phase packages the time variable is not explicitly present at all, whereas in the case of using stability theory techniques, it is interesting to analyze the process dynamics on a large time scale.< >
A small-scale study to evaluate the principal modes of propagation for an experimental, beyond-the-horizon UHF link is presented. The propagation channel was characterized statistically in several ways. The data colle...
详细信息
A small-scale study to evaluate the principal modes of propagation for an experimental, beyond-the-horizon UHF link is presented. The propagation channel was characterized statistically in several ways. The data collected exhibited characteristics of both troposcatter and diffraction modes of propagation, with heavy influence from anomalies such as aircraft reflections and ducting. The movement of weather fronts through the region influences the runways used by aircraft, and thus the degree of fading due to aircraft. Weather fronts also influenced the principal mode and characteristics of propagation. The correlation between meteorological conditions and the degree of fading observed was biased by the heavy air traffic and the large beamwidth receiver antenna used.< >
Addresses the design of fixed, low-order infinite impulse response (IIR) filters for modeling the perceptually significant features of the spectra of string instrument bodies. The problem is stated mathematically, and...
详细信息
Addresses the design of fixed, low-order infinite impulse response (IIR) filters for modeling the perceptually significant features of the spectra of string instrument bodies. The problem is stated mathematically, and the design methodologies compared here are reviewed. The experimental results are presented. The experimental set-up, data acquisition and data preprocessing are described. The spectra of the IIR filters designed using the least squares modified Yule-Walker equation, the Steiglitz-Mcbride method, and Hankel norm methods are compared. These spectra are also compared to the spectrum obtained by using an equivalent-order autoregressive model computed using the autocorrelation method. Results are summarized and ongoing work is discussed.< >
Automatic graphic object layout methods have long been studied in many application areas in which graphic objects should be laid out to satisfy the constraints specific to each application. In those areas, carefully d...
详细信息
Automatic graphic object layout methods have long been studied in many application areas in which graphic objects should be laid out to satisfy the constraints specific to each application. In those areas, carefully designed layout algorithms should be used to satisfy each application's constraints. However, those algorithms tend to be complicated and not reusable for other applications. Moreover, it is difficult to add each user's preferences to the layout scheme of the algorithm. To overcome these difficulties, the author developed a general-purpose interactive graphic layout system GALAPAGOS based on genetic algorithms. GALAPAGOS is general-purpose because graphic objects are laid out not by specifying how to lay them out, but just by specifying the preferences for the layout. GALAPAGOS can not only lay out complicated graphs automatically, but also allow users to modify the constraints at run time so that users can tell the system their own preferences.< >
The problem of pipelined data path synthesis in VLSI design is investigated. Given the behavioral description of the circuit to be synthesized, in which one or more conditional branches may be nested, the authors prop...
详细信息
The problem of pipelined data path synthesis in VLSI design is investigated. Given the behavioral description of the circuit to be synthesized, in which one or more conditional branches may be nested, the authors propose a heuristic algorithm, which takes the mutual exclusiveness of conditional branches into account so that more than one operation shares a hardware module, and hence a chip with small size will be produced. Experimental results show that the proposed heuristic algorithm efficiently produces a nearly optimum solution. The authors present a conditional resource sharing algorithm which simultaneously determines resource sharing and pipeline scheduling. The algorithm is based on a module sharing graph, in which each vertex corresponds to a pair of operators that may share a hardware module. With the module sharing graph, the conditional resource sharing problem is reduced to the problem of finding a set of paths on the module sharing graph. The proposed algorithm does not need user interaction, and hence no careful adjustment of parameter values is needed. Experimental results show that the proposed heuristic algorithm efficiently produces a nearly optimum solution.< >
暂无评论