Microsystems for Bioelectronics

3.2 INFORMATION AND INFORMATION PROCESSING

Information can be defined as a tec hnically quantitative measure of the distinguishabi lity of a physical

subsystem from its environment [1,2]. One way to create distinquishable states is by the presence or

absence of material particles (information carrier) in a given location. For example, information is

encoded in DNA through specific locations of certain molecular fragments, information of a printed

English text is created by positioning die particles on paper. Several examples of distinguishable states

used to create information are shown in Table 3.1. A more detailed quantitative discussion of the

concept of information is provided in Chapter 6. Also, comprehensive discussion of the topic in [3] can

be recommended for additional reading.

Information of arbitrary kind and amount (such as letters, numbers, colors, or graphics specific

sequences and patterns) can be represented by combination of just two distinguishable states, 0 and 1.

The maximum amount of information, which can be conveyed by a system with just two states is used

as a unit of information known as a 1 bit (abbreviated from ‘binary digit’).

A system with two distinguishable and controllable states forms a basis for the binary switch, the

fundamental computational element in information-processing systems (Fig. 3.1).

Three essential properties of a binary switch are Distinguishability, Controllability and Commu-

nicativity. We say that a bina ry switch is Distinguishable if and only if the binary state (0 or 1) can be

FIGURE 3.1

Constituents of an abstract binary switch

Table 3.1 Examples of distinguishable states used to create information

Information coding

system Number of Distinguishable states

English alphabet 27: a, b, c, .z, ‘space’

Morse code 3:  (‘dot’), — (‘dash’), ‘space’

Genetic code (DNA) 4: A (adenine), C (cytosine), G (guanine), T (thymine)

Binary code 2: 1 and 0

3.2 Information and information processing 53

determined with an acce ptable degree of certainty by a measurement (READ operation). The binary

switch is Controllable if an external stimulus can reliably change the state of the system from 0 to 1 or

from 1 to 0 (WRITE operation). The binary switch is communicative if it is capable of transferring its

state to other binary switches (TALK operation).

An arbitrary binary inf ormation-processing system consists of N binary switches connected in

a certain fashion to implement a specific function (e.g. logic, arithmetic etc.). Each binary switch is

characterized by a dimension L and switching time t

sw

(or switching frequency f ¼ 1/t

sw

). A related

dimensional characteristic is the number of binary switches, N (or the number of binary switches per

unit area, n). If area is fixed, to increase N, the char acteristic dimension, L, of the binary switch must

decrease:

N w

1

L

2

(3.1)

One indicator of the ultimate performance of an information processor, realized as an interconnected

system of binary switches, is the maximum binary throughput (BIT); that is the maximum number of

binary transitions per unit time:

BIT ¼

N

t

sw

¼ N,f (3.2)

One can increase the binary throughput by increasing the number of binary switches N, and/or by

decreasing the switching time, i.e., the time to transition from one state to the other, t

sw.

Increased

binary throughput has historically resulted in an increased information-processing system capability.

Table 3.2 shows several examples of Intel microprocessors characterized by the number of switches

(transistors), switching (clock) frequency, and their maximum binary throughput.

Another fundamental characteristic of a binary switch is the switching energy E

sw

and the related

power dissipation by a system of N binary switches is:

P ¼

N

t

sw

,E

sw

¼ BIT,E

sw

(3.3)

In the next sections the fundamental relations for n

bit

, t

sw

, E

sw

and the corresponding implications for

the computing systems are investigated.

Table 3.2 Examples of Intel microprocessors with respect of the number of switches, switching

frequency, maximum binary throughput and their computational capability*

Processor

# Switches

(transistors)

Switching (clock)

frequency

Max. binary

throughput

Capability/

application

8008 3500 200 kHz 7  10

8

General calculators

8080 6000 2 MHz 1.2 10

10

1

st

PC

Pentium

Extreme 965

376 000 000 3.73 GHz 1.4  10

18

High-performance

desktop

*

Data from the Intel Microprocessor Quick Reference Guide (http://www.intel.com/pressroom/kits/quickreffam.htm).

54 CHAPTER 3 Nanomorphic electronics