Self-timed logic circuits

Self-timed and asynchronous design techniques are currently proposed as a vehicle for pushing digital integrated circuits to higher levels of density and performance. The arguments for and against the adoption of these techniques are presented with illustrations from practical development projects. Some of the key principles behind self-timed operation are reviewed. Design tools to enable complex practical applications to be engineered are considered. For engineers who wish to find out more a selection of key references is provided     

Digital logic circuits

Recent progress in Josephson digital logic circuits is described. It is noted that changing the junction material from a lead alloy to niobium has dramatically improved process reliability, and that high-speed, low-power operations have been demonstrated at large-scale integrated-circuit levels. The first Josephson microprocessor, operated at 770 MHz, verified the potential of Josephson devices for future digital elements. The possibilities of the ultrafast Josephson computer, previously shelved because of a number of problems, are being actively reconsidered. The performance anticipated for Josephson digital circuits using high-temperature superconducting materials is also discussed     

Regenerative logic circuits with CMOS transistors

Logic circuits with transfer characteristics in the form of hysteresis, proposed in the paper, consist of two stages. The input stage is a standard CMOS logic circuit (inverter, NAND or NOR); the output stage is a simple flip-flop. The flip-flop consists of two inverters and one pair of CMOS transistors functioning as a resistor. The positive feedback loop is closed through these transistors. In the paper the most important static parameters and conditions of normal operation are analysed in detail.     

Dynamic logic CMOS circuits

A structure of dynamic CMOS logic based on the direct interconnection of p-channel logic and n-channel logic dynamic gates is reported. Prevention of glitches and other circuit problems are discussed. Application to a 16-bit parallel-adder design resulted in improved speed as well as important savings in layout area when compared to standard static design.     

Josephson nonlatching logic circuits

The authors describe nonlatching logic circuits that can be designed using Josephson junctions as the switching elements. The circuits require no current resetting and can be switched between their two logic states with a subnanosecond delay time. The switching behavior has been simulated numerically. The choice of parameters and junction types is analyzed. The distinctive features which make these circuits attractive are discussed.     

Self-biasing Josephson logic circuits

A self-biasing network for Josephson logic circuits that permits wide variations in junction critical currents, resistors, and power supply voltage is presented. The self-biasing network automatically switches resistors in or out to make the gate currents track with the critical currents of the logic gates. Results of Monte Carlo statistical analyses of the tolerances of this scheme are presented as a function of amount of correlation between the critical currents of the logic device and the biasing network, amount of systematic variation on a chip, and number of junctions used in the biasing network. Results indicate that almost a factor of two larger variations in the critical currents of the Josephson junctions can be tolerated when the self-biasing network is used, without adverse impact on the gate delays and the power dissipation.     

All-refractory Josephson logic circuits

A process for the fabrication of Josephson integrated circuits is described which uses only refractory materials. The Josephson devices are Nb-Si-Nb tunnel junctions which are formed in the initial phase of the process. After depositing a Nb-Si-Nb `trilayer' over the entire substrate, the individual devices are isolated by the selective niobium anodization process (SNAP). Other materials used are molybdenum for the normal resistors and bias-sputtered SiO/SUB 2/ for additional insulator layers. The process uses only five photolithographic steps to produce circuits of the direct-coupled isolation type. This simplicity is achieved by using some layers for multiple purposes and by fabricating components with different functional purposes in a single step. For example, the lower electrode of the Josephson devices also functions as the ground plane and the contacts to the ground plane are actually large-area Josephson junctions formed simultaneously with the active devices. Low capacitance junctions (~0.025 pF//spl mu/m/SUP 2/) are produced with good uniformity.     

Multifault and delay-fault testability of multilevel circuits

This paper investigates the relationship between test sets for multiple stuck-at faults and robust path-delay-fault tests in multilevel combinational circuits. It is shown that, in multilevel circuits, a complete robust path-delay-fault test set may not detect all multiple stuck-at faults. We also show that the detectability of the former does not imply the detectability of the latter, as suggested in a recent paper. The presence of undetectable or untested multiple stuck-at faults may invalidate some path delay tests.Supported in part by NSF Grant MIP-9320886.     

Opening the terahertz window with integrated diode circuits

The terahertz region of the electromagnetic spectrum, spanning from 100 GHz through 10 THz, is of increasing importance for a wide range of scientific, military and commercial applications. This interest is spurred by the unique properties of this spectral band and the very recent development of convenient terahertz sources and detectors. However, the terahertz band is also extremely challenging, in large part because it spans the transition from traditional electronics to photonics. This paper reviews the importance of this frequency band and summarizes the efforts of scientists and engineers to span the "terahertz technology gap." The emphasis is on solid-state circuits that use nonlinear diodes to translate the functionality of microwave technology to much higher frequencies.     

Switching nets and electronic logic circuits

It is suggested that methods of improving the reliability of switching nets are not necessarily applicable to electronic logic circuits.     

Fault detection for multiple-valued logic circuits with fanout-free

The single fault and multiple fault detections for multiple-valued logic circuits are studied in this paper. Firstly, it is shown that the cardinality of optimal single fault test set for fanout-free m-valued circuits with n primary inputs is not more than n + 1, for linear tree circuits is two, and for multiplication modulo circuits is two if n is an odd number or if n is an even number and m > 3, where the optimal test set of a circuit has minimal number of test vectors. Secondly, it is indicated that the cardinality of optimal multiple fault test set for linear tree circuits with n primary inputs is 1 + [n/(m - 1)], for multiplication modulo circuits is n + 1, for fanout-free circuits that consist of 2-input linear tree circuits and 2-input multiplication modulo circuits is not greater than n+ 1, where [x] denotes the smallest integer greater than or equal to x. Finally, the single fault location approaches of linear tree circuits and multiplication modulo circuits are presented, and all faults in th     

Full-swing BiCMOS logic circuits with complementaryemitter-follower driver configuration

Various full-swing BiCMOS logic circuits with complementary emitter-follower driver configurations are described. The performance of the circuits is demonstrated in a 1.2 μm complementary BiCMOS technology with a 6 GHz n-p-n and a 2 GHz p-n-p transistor. For the basic circuit, gate delay (fan-in=2, fan-out=1) is 366 ps and driving capability is 288 ps/pF at 4 V. Delay-power tradeoffs that depend on characteristics of the clamping diode between two base nodes of the complementary emitter-follower driver, parasitic capacitances at the two base nodes, and a technique that can be used to achieve full swing have been identified for these circuits. These circuits show leverage over the conventional BiCMOS circuit for reduced power-supply voltages     

Reliability evaluation of logic circuits

An important guide in the project of logic circuits is the ability to estimate rightly its reliability. In this paper a new type of analysis is presented, by following the references [1], [2], [3] and the method implemented by [4]. This is applicable to all logic circuits, combinational or sequential and by this new structure a reliability matrix is obtained.     

Characterization of integrated logic circuits

The characterization of integrated logic circuits must be accomplished in a manner which fully accounts for the circuit's nonlinear behavior and is amenable to experimental verification. The approach taken in this paper is to describe both the dc and the transient performance of the circuit by developing nonlinear equivalents of the 2-port "black box" parameters used in specifying linear networks. Such terminal parameter characterization has the obvious advantage of eliminating the need to probe the integrated circuit for testing purposes. In addition, knowledge of terminal performance is a necessity when the circuit is studied from a system point of view. In this paper an emitter-coupled logic circuit is used as an example to illustrate the analysis techniques. After accomplishing the terminal parameter characterization of this circuit, attention is directed towards using these results to establish a design procedure. To this end the relationship that exists between power consumption and the circuit safety margins is explored, and the minimum power-delay time product is derived. The analysis accounts for the parasitics which are present in a monolithic integrated circuit and illustrates the use of the nonlinear transistor model.     

Current-mode CMOS multiple-valued logic circuits

Current-mode CMOS circuits are receiving increasing attention. Current-mode CMOS multiple-valued logic circuits are interesting and may have applications in digital signal processing and computing. In this paper we review several of the current-mode CMOS multiple-valued logic (MVL) circuits that we have studied over the past decade. These circuits include a simple current threshold comparator, current-mode MVL encoders and decoders, current-mode quaternary threshold logic full adders (QFAs), current-mode MVL latches, current-mode latched QFA circuits, and current-mode analog-to-quaternary converter circuits. Each of these circuits is presented and its performance described     

Comparison of MOSFET logic circuits

Linear load, depletion-mode load, four-phase dynamic, and complementary MOSFET logic circuits are compared on the basis of power, delay, and density for two specific master slice layouts. The circuits were designed in a common technology base, and normalized power and delay characteristics were calculated by simulation. Chips of 1280 circuits were designed in images having one and two layers of metal, and power versus delay curves were calculated. The effect of an insulating substrate was also considered.     

Boolean decomposition in multilevel logic optimization

Algorithms are presented for Boolean decomposition, which can be used to decompose a programmable logic array (PLA) into a set of smaller interconnected PLAs such that the overall area of the resulting logic network, deemed to be the sum of the areas of the constituent PLAs, is minimized. These algorithms can also be used to identify good Boolean factors which can be used as strong divisors during the logic optimization to reduce the literal counts/area of general multilevel logic networks. Excellent results have been obtained     

Statistical power estimation of CMOS logic circuits with variableerrors

A statistical power estimation method is proposed where estimation time and accuracy can be balanced by assigning smaller (higher) errors to the nodes with higher (lower) power dissipation. To determine the errors, a quadratic programming based problem is formulated. Experimental results show a drastic reduction in the number of simulation patterns compared to previous methods