Browsing by Author "Leeds, J. Venn"
Now showing 1 - 10 of 10
Results Per Page
Sort Options
Item A linear group and state variable formulation of the electrical network analysis problem(1966) Barksdale, Gerald Leonard; Leeds, J. VennThis paper reports on the development of a network analysis technique employing both a linear graph theory formulation of the topological relations and a state variable analysis of these relations. As a consequence of these investigations, a network analysis program, GENA, was developed. Special points of interest include the development of an algorithm for efficient determination of the fundamental cut-set equations of the network and investigation of restrictions on the state variable formulation for electrical networks.Item A study of the effect on dynamics of a nuclear boiler modification(1965) Lessem, Allan Stuart; Leeds, J. VennThis report discusses the study of a modification in the structure of a boiler representative of the type used aboard nuclear-powered ships. The modification involves the relocation of the feed water injection point from the steam drum to the boiling region. On the basis of a simple mathematical model, the study is intended to be a first step in deciding the feasibility of this modification. The mathematical model is developed, and solutions of the system equations are compared with earlier work describing the original boiler configuration. Within the limitations of the model, the modification in boiler structure is seen to produce no essential alteration of system response to changes in steam flow. However, a substantial alteration is produced in system response to changes in feed water flows This is in such a direction as to suggest that the modification may make possible an improvement in boiler dynamics for control purposes, or at the very least, that no degradation of dynamics is produced.Item An analog/hybrid computer for the solution of partial differential equations(1969) Torgrimson, Peter Allen; Leeds, J. VennThe design of a specialized analog/hybrid computer for the solution of partial differential equations was investigated. The solution is computed with an analog computer controlled by a digital computer. The system designed has low cost and low accuracy. The interconnection of analog elements required for a general class of PDE's was derived. Several types of digitally set potentiometers were compared and the one selected was designed into a basic module. If the pot is used as a multiplier (with one digital input and one analog input) the speed of the integrator is severely limited by the time required for the digital computer to generate the necessary multiplication.Item An optimized model of the human cochlea(1967) Huff, Talbot Shelton; Leeds, J. VennIn this thesis, an electrical analogue of the human cochlea was constructed from the equations of cochlear motion and optimized to produce a response which agreed with the experimental response measured from the human cochlea. The response of the electrical analogue and the perturbation of the response for small changes in the parameters describing the physical structure of the cochlea were first calculated on a digital computer. This data was then used in a linear program to minimize the difference between the model response and that measured from the human cochlea. The procedure was then repeated until no further reduction occurred. The result was a new model of the cochlea, the response of which agreed more closely with that of the human cochlea than previous models.Item Applications of circuit analysis computer programs to estuarine pollution problems(1967) Bybee, Halbert Homer; Leeds, J. VennThis research establishes a method of analyzing the equations that describe pollution problems employing a digital computer program previously in existence, the IBM Electronic Circuit Analysis Program (ECAP). In this analysis, a waterway is approximated by an electrical network which adapts the partial differential equation normally used as a mathematical model for fluid flow, to solution by ECAP. This modeling method is investigated by analyses made in the frequency and time domains for specific estuarine pollution problems and by checking the results against the analytic solution of the partial differential equation. The application of ECAP is further expanded by calculations of the sensitivities of the various solutions to changes in the parameters of the river, using Leeds' method. This research provides an accurate method of investigating pollution problems on a digital computer without requiring any special programming for this problem.Item Automatic design of single stage common emitter transistor amplifiers(1968) Naden, Rex Alan; Leeds, J. VennThe creative process of the design of a single stage transistor amplifier is investigated. The minimum set of specifications is first considered. A procedure for selecting the best topology, element values, and transistor for the given inputs is described. A defense of the search method vs. the method of classical constrained optimization is presented. Resulting from this paper is the computer program CEDESP which takes the minimum data, plus additional optional inputs and sets forth the best design. Since the program is meant for a batch-processing-type computer with no operator interaction, internal decision loops are included which will correct inconsistent data and proceed to the solution.Item Error analysis of optical matched filter(1966) Lin, Chia-Hsiung; Leeds, J. VennThe degradation of performance of optical matched filters due to translation and rotation is examined. Included is a similar discussion due to variation in signal sizes. A rectangular signal is employed in carrying out the analysis. As a result of which the estimation of the allowed position errors in distinguishing signals of different sizes can be made.Item Frequency dependence of the sensitivity in continuously equivalent networks(1966) Ugron, Gabor Imre; Leeds, J. VennThe purpose of this paper is to examine sensitivity and its frequency dependence in continuously equivalent networks. First, continuously equivalent network theory and different methods of sensitivity calculations are presented. Then, recent results on minimum sensitivity networks are discussed and, as an extension, the frequency dependence of the sensitivity in equivalent networks is studied. As a method of calculation a new development using linear graph theory is applied. This allows efficient, repeated sensitivity calculation for different frequencies using a digital computer.Item Speech communication by means of infinitely clipped signals(1965) Hoffman, Eric Jay; Leeds, J. VennThe transmission of speech through difficult channels is more easily accomplished if the complicated speech waveform can be simplified. One method of simplification is to "infinitely clip" the speech to a rectangular signal which preserves only the original zero-crossings. Early research with clipped speech and clipped, differentiated speech is extended to clipped, twice differentiated speech and an "enriched" mixture of two clipped derivatives of different order. The two new systems produce a higher quality signal than the earlier methods, the "enriched" speech approaching typical communications quality. Data is presented comparing the intelligibility of these four systems. A scheme for preserving simultaneously the exact times of both zeros and extremals is discussed in detail and experimentally investigated. The intelligibility of time-quantized, second-order clipped. speech is studied, and certain statistical parameters of the clipped signals measured. These data indicate: that a three-to-one reduction in bit rate over telephonic PCM is possible while still preserving tolerable intelligibility. .Item The three-state-polarity-coincidence correlation detector(1964) Chang, Chieng-Yi; Leeds, J. VennThe three-state-polarity-coincidence correlator (PCC'), a two-input detection device with dead zone clippers, has been investigated in detail. A criterion is devised to compare two detectors. Comparisons are made between the PCC' and several other types of detectors. The superiority of the PCC' in the detection of a weak signal embedded in an additive noise over the PCC and other detectors is demonstrated. It is shown, particularly for certain Gaussian inputs, that there exists a maximum relative efficiency with respect to the dead zone width of clippers or the clipping level "b". It is concluded that the PCC' can always be used to replace the PCC in a weak signal detection.