Quarter-micrometer gate ion-implanted GaAs MESFET's with an f1 of 126 GHz

The fabrication and characterization of a 0.25-μm-gate, ion-implanted GaAs MESFET with a maximum current-gain cutoff frequency f_t of 126 GHz is reported. Extrapolation of current gains from bias-dependent S-parameters at 70-100% of I _dss yields f₁'s of 108-126 GHz. It is projected that an f₁ of 320 GHz is achievable with 0.1-μm-gate GaAs MESFETs. This demonstration of f₁'s over 100 GHz with practical 0.25-μm gate length substantially advances the high-frequency operation limits of short-gate GaAs MESFETs     

Binary images of cyclic codes over Z4

We determine all linear cyclic codes over Z₄ of odd length whose Gray images are linear codes (or, equivalently, whose Nechaev-Gray (1989) image are linear cyclic codes or are linear cyclic codes)     

Delay analysis of a packet voice multiplexer by the ΣDi/D/1 queue

The performance of a statistical multiplexer whose inputs consist of a superposition of voice packet streams is studied. The delay for such a system is analyzed by solving the ΣD_i/ D/1 queue. The analytic method can be used to find the approximate mean delay for an arbitrarily large number of trunks and the approximate delay distribution when the number of trunks is less than 100. An efficient hybrid simulation of the packet voice multiplexer which can be used to find the delay distribution for a large number of trunks is presented. In addition, easily computable error bounds for the present approximation are provided, and the accuracy of the M/ D/1 approximation is investigated     

Measurement of the characteristics of the optical-microwavedouble-resonance effect of the 87Rb D1line for application as an atomic frequency standard

The OMDR (optical-microwave double resonance) spectrum of ⁸⁷ Rb with the aim of using a frequency-stabilized GaAs semiconductor laser instead of an Rb lamp as a pumping source in a gas-cell-type Rb frequency standard. Natural isotope ⁸⁷Rb was sealed in a glass cell with buffer gases (Ar/N₂=1.2, total pressure=39 torr). The double resonance signal in the 5P_1/2(F=2)←5S_1/2( F=1) transition appearing at the resonance to the F=2←1 hyperfine transition of the 5S_1/2 state was detected. The optimum operational cell temperature was 56°C. The peak-to-peak frequency width of the atomic hyperfine resonance discriminator used to stabilize the microwave frequency shifts induced by detuning of the laser frequency, changes in the laser and microwave powers, and temperature drift of the cell were investigated     

Quasi-cyclic codes over Z4 and some new binarycodes

Previously, (linear) codes over Z₄ and quasi-cyclic (QC) codes (over fields) have been shown to yield useful results in coding theory. Combining these two ideas we study Z ₄-QC codes and obtain new binary codes using the usual Gray map. Among the new codes, the lift of the famous Golay code to Z₄ produces a new binary code, a (92, 2²⁴, 28)-code, which is the best among all binary codes (linear or nonlinear). Moreover, we characterize cyclic codes corresponding to free modules in terms of their generator polynomials     

Image restoration by convex projections using adaptive constraintsand the L1 norm

A new projection method based on the row action projection (RAP) algorithm for image restoration is presented. The new implementation is computationally attractive and facilitates local adaption of projection operators. The local mean and the least L₁ norm set of solutions are used as constraints. Computer simulations illustrate the new methods to be very competitive in restoring missing spectral components of a degraded image     

On algebraic decoding of the Z4-linearCalderbank-McGuire code

The quaternary Calderbank-McGuire (see Des., Codes Cryptogr., vol.10, no.2, 1997) code is a Z₄-linear code of length 32 which has 2³⁷ codewords and a minimum Lee distance of 12. The Gray map of this code is known to be a nonlinear binary (64, 2³⁷,12) code. The Z₄-linear Calderbank-McGuire code can correct all errors with Lee weight ⩽5. An algebraic decoding algorithm for the code is presented in this paper. Furthermore, we discuss an alternative decoding method which takes advantage of the efficient BCH decoding algorithm     

0.33-μm gate-length millimeter-wave InP-channel HEMTs with highf1 and fmax

The fabrication of 0.33-μm gate-length AlInAs/InP high electron mobility transistors (HEMTs) is reported. These InP-channel devices have f_t values as high as 76 GHz, f_max values of 146 GHz, and maximum stable gains of 16.8, 14, and 12 dB at 10, 18, and 30 GHz, respectively. The extrinsic DC transconductances are as high as 610 mS/mm; with drain-source breakdown voltages exceeding 10 V. The effective electron velocity in the InP channel is estimated to be at least 1.8×10⁷ cm/s, while the f_tL_g product is 29 GHz-μm. These results are comparable to the best reported results for similar InGaAs-channel devices     

Analysis of erbium-doped fiber amplifiers pumped in the4I15/2-4I13/2 band

A theoretical model for the amplifiers predicts gain coefficients of around 2 dB per milliwatt of launched pump power, in close agreement with previously reported measurements. The model is also used to determine the optimal pump wavelength λ_opt that maximizes the amplifier gain. It is shown that the latter is insensitive to pump detuning near λ_opt=1.48 μm, within a 20 nm range, which indicates that broadband, multimode laser diode pump sources and pumps with narrow linewidths should yield identical gain performance     

Order N log2 N WHT/DHT algorithm

An efficient algorithm is proposed which computes the coefficients of the higher order discrete Hartley transform (DHT) directly from the coefficients of lower-order DHTs. With this new development, the two-stage Walsh-Hadamard transform/discrete Hartley transform (WHT/DHT) is comparable to the existing fast algorithms. The same approach can also be used for the computation of DCT coefficients     

Thermal shifts of the spectral lines in the 4F3/2 to 4I11/2 manifold of anNd:YAG Laser

A report is presented of the thermal shifts of eleven of the twelve lines from the ⁴F_3/2 Stark energy levels to the ⁴I_11/2 energy levels in an Nd:YAG laser for a temperature change from 20-200°C. The thermal shift difference between the Stark sublevels R₁, R₂ in ⁴F_3/2 is found to be about -0.6±0.6 cm^-1/100°C. Within experimental uncertainty, all of the lasing lines either moved to longer wavelength or remained unchanged with increasing temperature     

A single-hop F2 propagation model forfrequencies above 30 MHz and path distances greater than 4000 km

Standard textbook explanations of ionosphere propagation suggest that 30 MHz and 4000 km are the limiting frequency and distance for single-hop communications. During the peak of sunspot cycle 22, however, there have been hundreds of reports of F₂ communications exceeding great-circle distance of 4200 km at 50 MHz. It is shown that parabolic F₂ ionospheric models explain this propagation irregularity, although textbooks and reference books fail to show that such long distances at VHF are possible     

A proposed chemically pumped laser based on theb→X transition in O2

Simple kinetics calculations demonstrate that the well-known electronic energy pooling reaction involving O₂(a ¹Δ) is capable of producing an effective population inversion in O₂(b¹Σ). The densities of O₂(a¹Δ) which are potentially achievable suggest that the extractable energy storage density of an O ₂(b→X) chemical laser might exceed 0.5 MJ/m³. The b→X emission lifetime measurements conducted under conditions of high relative O₂( b¹Σ) density reveal no evidence of rapid self-quenching effects which would be potentially detrimental to laser performance. The relatively long energy storage times predicted, together with freedom from reagent mixing requirements, make such a laser an attractive alternative to other existing and proposed short wavelength chemical lasers     

A linear dependence of Fmin on frequency forTET's

The minimum noise figure of a FET, expressed in decibels, is shown to increase approximately linearly with frequency to frequencies approaching the FET's f_max. This observation is useful for extrapolating noise figures and provides a simple frequency-of-merit     

Optimal pump wavelength in the4I15/2-4I13/2 absorption band for efficient Er3+-doped fiberamplifiers

The authors report the measured gain of a highly efficient erbium-doped fiber amplifier pumped at wavelengths between 1.46 and 1.51 μm. The optimal pump wavelength, λ_opt, was determined to be 1.475 μm. At this wavelength, the maximum gain coefficients for signals at 1.531 and 1.544 μm were 2.3 and 2.6 dB/mW, respectively. At λ_opt, high gains ranging from 32 dB at pump power P_p=20 mW up to 40 dB at P _p=80 mW were obtained. These modest pump powers are within the capabilities of currently available 1.48-μm diode lasers. The width about λ_opt for 3-dB gain variation exceeded 27 nm for P_p=10 mW and 40 nm for P_p >20 mW. With this weak dependence on pump wavelength, single-longitudinal-mode lasers do not have a significant advantage over practical Fabry-Perot multimode pump lasers     

A self-injection locked, Q-switched, line-narrowed Ti:Al2O3 laser

Line-narrowing, Q-switched, and self-injection locking are studied independently and as a system. Line narrowing is shown both theoretically and experimentally to depend on the inverse square root of the pulse evolution time interval. Q switching of the Ti:Al₂O₃ laser is demonstrated and the laser output energy as a function of the Q-switch delay is investigated. Self-injection is demonstrated and the operation of the laser is explored as a function of loss and the Q-switch delay. Self-injection locking is demonstrated and the performance as a function of the Q-switch delay is determined     

Measurement of the stimulated emission cross section for the 4F3/2-4I13/2transition of Nd3+ in YAlO3 crystal

A method for the simultaneous measurement of the stimulated emission cross section and fluorescence lifetime by studying the relation between laser parameters and the laser relaxation oscillation frequency is discussed. The stimulated emission cross section for the ⁴F_3/2-⁴I_13/2 transition of Nd³⁺ ion in YAP crystal was measured to be (22±1)×10^-20 cm²     

Optical parametric oscillation in MgO:LiNbO3 driven by adiode pumped single frequency Q-switched laser

Optical parametric oscillation tunable over the 940-1220-nm spectral region has been obtained from a monolithic magnesium-oxide-doped lithium niobate ring resonator. The monolithic optical parametric oscillator (OPO) was pumped by the 532-nm second harmonic of a diode-laser-pumped single-frequency Q-switched Nd:YAG laser. The exceptional frequency and amplitude stability of the single-frequency pump source provided stable OPO output pulses     

Frequency stabilization of a GaAs semiconductor laser by use of theoptical-optical double-resonance effect of the Doppler-free spectrum ofRb D1 line

The frequency of a GaAs semiconductor laser was locked to the hyperfine spectrum of the Rb D₁ line by saturation spectroscopy. A first GaAs laser was locked to one of the Doppler-free hyperfine lines using a source modulation method. The burnt hole modulated by the stabilized laser beam was detected by a second laser beam irradicated from the same or the opposite direction and was used as a frequency discriminator to stabilize the second laser. The characteristics of the double-resonance spectrum were measured. The frequency stability of the second laser, which was free from frequency modulation, was estimated to be on the order of 10^-9     

Power-efficient layout of a fiber-optic multistar that permits log2 N concurrent baseband transmissions among N stations

A passive, single-hop, fiber-optic interconnection among N stations, each with two transmitters and one receiver, and a round-robin transmission schedule for it, which jointly permit log ₂ N concurrent noninterfering transmissions on a single wavelength, has recently been described. This is a substantial improvement over the previously known limit of two concurrent transmissions, but the layout of this interconnection poses a challenge in terms of both wiring complexity and path loss. A power-efficient implementation of this interconnection using several stages of balanced fiber-optic star couplers is presented here. With lossless components, path loss is N, the same as that of a single-star interconnection that permits only a single transmission at a time. Consequently, the high degree of parallelism translates into higher capacity. The required number of (2×2) star couplers is also very similar to that required for implementing a single N×X star