The Effects of Interconnections on High-Speed Logic Circuits
TL;DR: It is shown that high-speed circuitry must be miniaturized and the implications are discussed.
Abstract: By way of worked examples in typical but somewhat idealized cases the effect on circuit speed of circuit interconnections is studied. The source, calculation and minimization of interconnection crosstalk is also discussed. It is shown that high-speed circuitry must be miniaturized and the implications are discussed.
Citations
More filters
IBM1
TL;DR: In this paper, an effective low-power CMOS terminating circuit for coupled transmission lines is described, which substitutes for a terminating resistor at the receiving end of the line in order to provide substantial benefits in minimizing power dissipation, crosstalk, overshoot, and ringing.
Abstract: An effective low-power CMOS terminating circuit for coupled transmission lines is described. This CMOS terminating circuit substitutes for a terminating resistor at the receiving end of the line in order to provide substantial benefits in minimizing power dissipation, crosstalk, overshoot, and ringing. In notebook-type personal computers, it is important to minimize the power dissipation required for termination with improvement of crosstalk, overshoot, and ringing so as to extend battery life. Power dissipation is reduced to a quarter of that of shunt resistor termination and crosstalk is reduced to one half of that of series resistor termination. Furthermore, the CMOS terminating circuit can be effectively used as a damping resistor to eliminate ringing on a signal line which has a large inductance. >
2 citations
TL;DR: The most frequently-used approximate model for predicting crosstalk is referred to as the Jarvis model after its original publisher in 1963, which uses three approximations that simplify the result.
Abstract: The coupled transmission-line equations for a three-conductor, lossless transmission line are solved in symbolic form for its crosstalk voltages using various approximations [1, 2]. The most frequently-used approximate model for predicting crosstalk is referred to as the Jarvis model after its original publisher in 1963 [3]. The Jarvis model uses three approximations that simplify the result. One of the approximations requires that all four line terminations be matched which is unrealistic for virtually all crosstalk problems.
2 citations
Cites background or methods from "The Effects of Interconnections on ..."
...In this case the exact solution obtained in (12) simplifies considerably in the time domain as [3-7]...
[...]
...The most frequently-used approximate model for predicting crosstalk is referred to as the Jarvis model after its original publisher in 1963 [3]....
[...]
...10 Jarvis [3] and ill be referred to as the Jarvis odel....
[...]
...This solution was apparently first published by Jarvis [3] and will be referred to as the Jarvis Model....
[...]
...In this case the exact solution obtained in (12) si plifies considerably in the ti e do ain as [3-7] ) 2 ( ) ( ) ( t S t S b t (19a)...
[...]
TL;DR: In this article, the authors present theories permettant la modelisation and l'analyse temporelle d'un ensemble of lignes couplees, and examine les ameliorations apportees sur le decouplage par l'utilisation de ligneecran ou dǫun second plan de masse rapporte audessus.
Abstract: L’auteur presente les theories permettant la modelisation et l’analyse temporelle d’un ensemble de lignes couplees. Pour un bus d’interconnexion en technologie GaAs de 3 mm de long, active par un signal de 30 ps de temps de montee, il met en evidence le role preponderant du couplage: degradation des signaux sur les lignes activees et parasitage des lignes non activees. Il examine les ameliorations apportees sur le decouplage par l’utilisation de lignesecran ou d’un second plan de masse rapporte audessus des lignes.
2 citations
01 Jan 2003
TL;DR: In this article, a technique known as swizzling has been proposed to make use of this fact to control inductive coupling in wide global signal busses without negatively impacting the capacitive coupling.
Abstract: Recent advances in Deep Submicron Design (DSM) have considerably increased the importance of inductive coupling effects for long interconnects. This is especially significant for global interconnect with its wide busses that may run from one corner of a chip to the other. Inductive coupling has been shown to depend on the distance wires run in parallel to each other and the activity on these wires. It has also been shown that the presence of signal wires separating the attacker and the victim leads to a reduction in the inductive coupling between the two. A technique known as swizzling has been proposed to make use of this fact to control inductive coupling in wide global signal busses. In this paper we show that this technique reduces the inductive coupling for the most vulnerable wires neighboring the attacker in significantly long busses by around 20% without negatively impacting the capacitive coupling.
2 citations
13 May 2007
TL;DR: A simple and efficient model to predict the timing impact of crosstalk in high-speed interconnects with impedance discontinuities based on the reflection and superposition principles is proposed.
Abstract: In this paper, we propose a simple and efficient model to predict the timing impact of crosstalk in high-speed interconnects with impedance discontinuities. This model is based on the reflection and superposition principles. The results of the simple model are accurate to picoseconds or less, when compared to HSPICE simulations with 1~40mm interconnect length. The model can also be extended to deal with multiple parallel interconnects. In addition, this model can predict timing results relatively quickly, compared to simulation by HSPICE.
2 citations
Cites methods from "The Effects of Interconnections on ..."
...In our previous work [9], we proposed a simple and efficient model based on the earlier equations [4,5] to predict the timing effects of the edge transition due to crosstalk on multiple parallel interconnects....
[...]
...Jarvis presented equations derived from Maxwell’s equations to calculate the voltage amplitude of crosstalk [4]....
[...]
References
More filters
01 Nov 1954
TL;DR: In this article, a simple configuration of four wires (or two wires and ground) can serve simultaneously as a directional coupler, filter, and transformer, and the coupled lines may be of equal or different impedance.
Abstract: The natural coupling between parallel transmission lines is inherently directional. Very simple and cheap directional couplers can be made which utilize this effect. By introducing appropriate variation of coupling with distance a wide variety of transmission characteristics may be realized, including high-pass (ideally, infinite bandwidth) characteristics. The coupled lines may be of equal or different impedance. Thus, a simple configuration of four wires (or two wires and ground) can serve simultaneously as a directional coupler, filter, and transformer.
215 citations
IBM1
TL;DR: This computer, like the 704, is aimed at scientific problems such as reactor design, hydrodynamics problems, partial differential equations etc., its instruction set and organization are such that it can handle with ease data-processing problems normally associated with commercial applications, such as processing of alphanumeric fields, sorting, and decimal arithmetic.
Abstract: The Stretch Computer project was started in order to achieve two orders of magnitude of improvement in performance over the then existing 704. Although this computer, like the 704, is aimed at scientific problems such as reactor design, hydrodynamics problems, partial differential equations etc., its instruction set and organization are such that it can handle with ease data-processing problems normally associated with commercial applications, such as processing of alphanumeric fields, sorting, and decimal arithmetic.
76 citations
[...]
TL;DR: This paper gives a brief description of work originating in the Computer Group at Manchester University, the name given to a large computing system which can include a variety of peripheral equipments, and an extensive store.
Abstract: This paper gives a brief description of work originating in the Computer Group at Manchester University. Atlas is the name given to a large computing system which can include a variety of peripheral equipments, and an extensive store. All the activities of the system are controlled by a program called the supervisor. Several types of store are used, and the addressing system enables a virtually unlimited amount of each to be included. The primary store consists of magnetic cores with a cycle time of under two microseconds, which is effectively reduced by multiple selection mechanisms. The core store is divided into 512 word "pages"; this is also the size of the fixed blocks on drums and magnetic tapes. The core store and drum store are addressed identically, and drum transfers are performed automatically as described in Section 3. There is a fixed store which consists of a wire mesh into which ferrite slugs are inserted; it has a fast read-out time, and is used to hold common routines including routines of the supervisor. A subsidiary core store is used as working space for the supervisor. The V-store is a collective name given to various flip-flops throughout the computer, which can be read, set, and re-set by reading from or writing to particular store addresses.
28 citations
Book•
[...]
01 Nov 2001
TL;DR: In this article, a brief description of work originating in the Computer Group at Manchester University is given, where the core store is divided into 512 word "pages"; this is also the size of the fixed blocks on drums and magnetic tapes.
Abstract: This paper gives a brief description of work originating in the Computer Group at Manchester University. Atlas is the name given to a large computing system which can include a variety of peripheral equipments, and an extensive store. All the activities of the system are controlled by a program called the supervisor. Several types of store are used, and the addressing system enables a virtually unlimited amount of each to be included. The primary store consists of magnetic cores with a cycle time of under two microseconds, which is effectively reduced by multiple selection mechanisms. The core store is divided into 512 word "pages"; this is also the size of the fixed blocks on drums and magnetic tapes. The core store and drum store are addressed identically, and drum transfers are performed automatically as described in Section 3. There is a fixed store which consists of a wire mesh into which ferrite slugs are inserted; it has a fast read-out time, and is used to hold common routines including routines of the supervisor. A subsidiary core store is used as working space for the supervisor. The V-store is a collective name given to various flip-flops throughout the computer, which can be read, set, and re-set by reading from or writing to particular store addresses.
27 citations
01 Jan 1960
18 citations