Рефераты. Physical Methods of Speed-Independent Module Design

Physical Methods of Speed-Independent Module Design

Physical Methods of Speed-Independent Module Design

Oleg Izosimov

INTEC Ltd, Room 321, 7a Myagi Street, Samara 443093, Russia

1. Introduction

Any method of logic circuit design is based on using formal models

of gates and wires. The simplest model of a gate is determined by only two

"parameters": (a) Boolean function is to be calculated, (b) fixed

propagation delay. The simplest model of a wire is an ideal medium with

zero resistance and consequently, with zero delay. Such simple models allow

circuit design procedures which are a sequence of elementary steps easily

realized by a computer.

When logic circuits designed by using the simplest models expose

unreliable operation as in the case of gate delay variations, designers

introduce less convenient but more realistic models with arbitrary but

finite delay. Using more complicated models may produce logic circuits that

are called speed-independent [1].

In speed-independent circuits transition duration can be arbitrary.

So a centralized clock cannot be used. Instead special circuitry to detect

output validity is applied. Besides, additional interface circuitry is

needed to communicate with the environment in a handshaking manner. A speed-

independent circuit can be seen as a module consisting of combinational

logic (CL) proper, CL output validity detector (OVD) and interface

circuitry (Fig.1). To enable OVD to distinguish valid output data from

invalid ones, the redundant coding scheme was proposed [2]. The main idea

of the scheme is to enumerate all possible input and output data, both

valid and invalid. The OVD must be provided with appropriate information on

data validity. To realize the idea of redundant coding some constraints on

CL design are imposed [3]:

[pic]

(i) CL must be free of delay hazards, i.e. CL output data word must not

be dependent on the relative delay of signal paths through CL.

(ii) In changing between input states, any intermediate or transient

states that are passed through must not be mapped by CL onto valid output

states.

When these constraints were formulated, the circuit designers

realised that not every Boolean description could be implemented in a speed-

independent style. Other approaches to speed-independent module design were

needed.

SIM design as a science has two branches: logical and physical. For

a long time physical branch was overshadowed in spite of its

competitiveness. The main properties of physical approach to SIM design

are:

(a) Arbitrary coding scheme.

(b) Conventional procedure of operational unit design.

(c) Races of signals in SIM do not affect on its proper operation.

In this paper we propose an approach based on the physical nature of

transitions in CL. We believe that each transition is actually a transfer

of energy which can be naturally detected by physical methods.

From the viewpoint of a radio engineer CL behaves like a radio

transmitter. It emits radio frequencies in the 108-1010Hz band modulated by

signals of 106-108Hz. Obviously, the carrier wave is produced by gate

switchings during transitions in CL. The modulating wave is produced by

control schemes (OVD and interface circuitry) that detect transition

completion and inform the environment about the readiness of CL. OVD is a

kind of radio receiver that extracts the modulation envelope and enhances

the received signal. The main properties that OVD circuit must expose from

a radio engineer's point of view are selectivity and high gain. Since the

useful signals can propagate through non-conducting medium, OVD circuits

can be coupled with CL indirectly.

Advances in semiconductor technology gave birth to two methods of

transition detecting based on two kinds of the information carrying signal,

namely electromagnetic radiation and current consumption. Frequency of the

signal produced by switching logic gates is determined by gate delay.

For instance, CMOS network of 1-ns gates produces 1-GHz signal, ECL

array of 100-ps gates gives 10-GHz radiation. Logic circuits consisting of

10-ps gates will emit infra-red radiation. That signal could be easily

detected by photosensitive devices.

2. Background

Let us have a closer look at the structure of speed-independent

modules (SIM) as presented in Fig.1. All input data are processed in CL,

all output data are obtained from CL, too. So, CL is the only unit in SIM

which is involved in proper data processing. The result of that processing

is specified by Boolean functions. Algorithms for calculating the Boolean

function are realised by the internal structure of CL. Generally, its

structure is series-parallel as well as algorithm implemented.

When n-bit data word is put into the CL, n or more signal

propagation paths (SPPs) can be activated concurrently. So, one can say

that the calculation of a Boolean function by CL is of parallel nature. On

the other hand, each SPP is a gate chain which processes data in a serial

manner. So, calculation in CL is also of sequential nature.

The OVD circuit is intended for detecting transient and steady

"states" of CL. If any SPP in CL is still "active", CL is in transient

state, otherwise it is in steady state. Each gate switching results in both

logical and electromagnetic effects on its surrounding medium. The logical

effects of switching has been heavily investigated; we consider physical

one.

To provide speed-independence of the module the OVD and interface

circuitry must also work in a speed-independent mode. This means that any

arbitrary but finite transistor or wire delay cannot impair proper

operation of OVD and interface circuitry.

The interface circuitry is a mediator between OVD and environment of

SIM. It implements any kind of signalling convention, commonly a two- or

four-cycle one [4] based on request Req and acknowledgement Ack signal

using. The interface circuitry receives the output validity (OV) signal

from the OVD circuit, a Req signal from the environment and transmits an

Ack signal to the environment (Fig.1).

Consider an algorithm of operation for interface circuitry realizing

speed-independent four-cycle signalling convention (FCSC). In accordance

with FCSC the control signals must go in the following sequence: Req+OV-

Ack+Req-Ack- where "+" corresponds to rising the signal and "-" corresponds

to falling the signal. All signals are assumed to adhere to positive logic.

Initially the signals Req and Ack are low, the signal OV is high. If the

environment state changes, the Req signal rises and transient state of CL

occurs (OV-). Upon completion of the transitions in CL, signal OV rises and

the interface circuitry generates the Ack signal rising. After that the

environment produces a falling Req signal and then the interface circuitry

transmits the falling Ack signal to the environment. All the signals have

to be reset into the initial state.

To develop the interface circuitry a circuit designer must take into

account that any OVD circuit has finite (non-zero) turn-on delay ton. This

means that OVD cannot respond on transitions of short duration t tr< ton .

An example of interface circuitry is shown in Fig.2. It contains a

flip-flop, a NOR-gate, an asymmetrical delay and an inverter as an output

stage [5].

[pic]

The asymmetrical delay is intended for delaying Req rising signal

for + period where + > ton . Delaying Req falling signal noted - is to be

as short as possible. Note that speed-independent operation of interface

circuitry is vulnerable to delay + variation. If + becomes less than ton ,

proper operation of SIM can not be guaranteed. Otherwise, if + is much

more than ton , performance of SIM will be significantly reduced. To

provide exact accordance of + and ton a circuit emulator can be used.

Such an emulator is either an exact copy of OVD or its functional

copy, i.e. resistive-capacitive model of OVD's critical path. In the chip

the emulator must be placed next to active OVD circuit in order to ensure

identical conditions of fabrication and operation.

In this example we use a simplified asymmetrical delay implemented

as an asymmetrical CMOS inverter chain (Fig.3). Contrary to the common

inverter an asymmetrical one has non-equal rise and fall times of output

signal.

[pic]

A time diagram for interface circuitry is presented in Fig.4 for two

cases: (a) ttr < ton and (b) ttr ton. In case (a) the signal sequence

Req+Ack+ is formed for (++tNOR) period where tNOR is a NOR-gate delay. In

case (b) the above sequence is formed for (ttr +toff+tNOR) duration where

toff is a turn-off delay of OVD circuit. When the SIM returns to the

initial steady state, the signal sequence Req-Ack- is formed for (-+tNOR)

interval.

[pic]

After considering the SIM in operation it is obvious that the main

problems of the module design are in the area of CL and OVD interaction.

This includes (a) kind of signal used as a carrier of information about CL

output validity, and (b) method of OVD circuit design.

4. Current consumption detection

Using current consumption of CMOS CL for output validity detection

Страницы: 1, 2, 3



2012 © Все права защищены
При использовании материалов активная ссылка на источник обязательна.