John A. Copeland

Short-cavity InGaAsP injection lasers: Dependence of mode spectra and single-longitudinal-mode power on cavity length

Simple expressions are given to describe the lower and upper limits of the single-mode (single-frequency) power as a function of the cavity length for InGaAsP injection lasers. It has been found that the lower limit of the single-mode power is proportional to the cavity length, while the upper limit is inversely proportional to the cavity length. Thus, a short-cavity laser provides a favorable geometry for obtaining single mode output over a wide range of power levels and currents above threshold. The mode stability versus temperature is also improved by a short-cavity design. The theoretical results agree with our recent experiments on very-short-cavity ( 50-75 \mu m) stripe-geometry InGaAsP lasers, which have shown consistent single-mode output over wide current ranges.

Visualizing network data for intrusion detection

As the trend of successful network attacks continue to rise, better forms of intrusion detection and prevention are needed. This paper addresses network traffic visualization techniques that aid an administrator in recognizing attacks in real time. Our approach improves upon current techniques that lack effectiveness due to an overemphasis on flow, nodes, or assumed familiarity with the attack tool, causing either late reaction or missed detection. A port-based overview of network activity produces a improved representation for detecting and responding to malicious activity. We have found that presenting an overview using stacked histograms of aggregate port activity, combined with the ability to drill-down for finer details allows small, yet important details to be noticed and investigated without being obscured by large, usual traffic. Due to the amount of traffic as well as the range of possible port numbers and IP addresses, scaling techniques are necessary to help provide this overview. We provide graphs with examples of forensic findings. Finally, we describe our future plans for using live traffic in addition to our forensic visualization techniques.

LSA Oscillator‐Diode Theory

The efficiency, negative resistance, and conditions for space‐charge control relevant to operation of a bulk semiconductor diode in the limited space‐charge accumulation (LSA) mode of oscillation are discussed. Numerical results for n‐GaAs indicate a maximum dc to rf‐conversion efficiency of 18.5% for sine‐wave excitation. The usable range of doping to frequency for n‐GaAs is found to be 2×104 to 2×105 sec/cm3 with an optimum value of about 6×104 sec/cm3. Although a reduction in efficiency at frequencies above 100 GHz is expected due to the finite response time of GaAs, worthwhile efficiency may be obtained at several hundred GHz.

Balanced media access methods for wireless networks

Fairness algorithms and access methods enable non-zero channel access for wireless communication systems operating in a random access channel environment. Fair access to a random access channel for each station in a wireless network is assured by each station calculating a priority or probability for accessing the channel based on logical connections among certain stations, based on other stations perception of the channel and based on each calculating stations own perception of the channel properties.

A technique for directly plotting the inverse doping profile of semiconductor wafers

A new technique for plotting doping profiles of semiconductor wafers is described. This technique involves driving a Schottky diode deposited on the surface with a small constant RF current (a few hundred microamperes at 5 MHz). The depth of the depletion layer is varied by changing the dc bias, but this is the only role of the dc voltage. The inverse doping profile n-1(x) is obtained by monitoring the voltage across the diode at the fundamental frequency, which is proportional to the depth x, and the second harmonic voltage, which is proportional to n-1. This type of plotter has the advantages of simplicity, high resolution (limited only by the Debye length in most cases), immediate results, and economy.

IDS rainStorm: visualizing IDS alarms

The massive amount of alarm data generated from intrusion detection systems is cumbersome for network system administrators to analyze. Often, important details are overlooked and it is difficult to get an overall picture of what is occurring in the network by manually traversing textual alarm logs. We have designed a novel visualization to address this problem by showing alarm activity within a network. Alarm data is presented in an overview where system administrators can get a general sense of network activity and easily detect anomalies. They then have the option of zooming and drilling down for details. The information is presented with local network IP (Internet Protocol) addresses plotted over multiple yaxes to represent the location of alarms. Time on the x-axis is used to show the pattern of the alarms and variations in color encode the severity and amount of alarms. Based on our system administrator requirements study, this graphical layout addresses what system administrators need to see, is faster and easier than analyzing text logs, and uses visualization techniques to effectively scale and display the data. With this design, we have built a tool that effectively uses operational alarm log data generated on the Georgia Tech campus network. The motivation and background of our design is presented along with examples that illustrate its usefulness.

Rogue access point detection using temporal traffic characteristics

As the cost of IEEE 802.11 hardware continues to fall, the appeal of inserting unauthorized wireless access into enterprise networks grows. These rogue access points (AP) expose the enterprise network to a barrage of security vulnerabilities in that they are typically connected to a network port behind the firewall. Most of the current approaches to detecting rogue AP are rudimentary and are easily evaded by hackers. We propose the use of temporal traffic characteristics to detect rogue AP at a central location. This detection is independent of the wireless technology (IEEE 802.11a, 802.11b, or 802.11g), is scalable, does not possess the inefficiencies of the current solutions, and is independent of the signal range of the rogue AP.

Semiconductor impurity analysis from low-frequency noise spectra

The usefulness of material for semiconductor devices can be degraded by the presence of deep impurity levels or traps which are often difficult to detect. The equations necessary for using the noise emitted by a uniform semiconductor when biased with a small dc current to find the energy, degeneracy factor, density, and time constants of such a deep level are derived. The theory was verified by an experimental study of a level in n-GaAs which was found to have an energy 0.175 eV below the conduction band and an acceptor-like degeneracy factor of 4. This technique appears to be useful for fundamental studies and as a means of monitoring material used for devices.

Theoretical study of a Gunn diode in a resonant circuit

A computer simulation of a GaAs Gunn diode in a parallel resonant circuit has been made to determine the optimum device and circuit parameters. The maximum dc to RF efficiency, 5 to 8 percent, is obtained when the product of doping and length is between 1012and 2 × 1012cm-2, the product of frequency and length is 107cm/s, and the bias voltage divided by length is 8000 V/cm for a load resistance of 30 R_{0} where R 0 is the low-voltage resistance of the diode. The product of output power and load resistance varies with frequency f as C f^{2} where C is 12,000 watt-ohm-GHz2for a load resistance of 50 R_{0} . The frequency can be varied over an octave tuning range by the resonant circuit.

Diode edge effect on doping-profile measurements

Diode edge effects can be a source of error in doping profile measurements made by the capacity versus voltage technique or CIP technique. A numerical calculation indicates that for uniform doping the apparent doping will be higher than the actual doping by a factor of (1 +bx/r)^{3} where b=1.5 and x/r is the ratio of depth x to diode radius r . Comparison with other sources of error shows that edge-effect error will be the largest error when the diode diameter is greater than 10 mils (250 µm) because this error decreases at the slowest rate as the diameter increases (as the reciprocal of the diameter rather than the reciprocal of the diameter squared).