Signals and systems - Biomedical Engineering Desk Reference - page 64

Figure 2.3-3 An early electrocardiogram machine.

an imaging system based on ultrasonic reflection, because
CT uses ionizing radiation that may have an associated
risk. (The actual risk of ionizing radiation is still an open
question, and imaging systems based on x-ray absorption
are considered minimally invasive. ) Ultrasound and ra-
diographic imaging would be considered less invasive
than, for example, monitoring internal cardiac pressures
through cardiac catheterization in which a small catheter
is threaded into the heart chamber. Indeed, many of the
outstanding problems in biomedical measurement, such
as noninvasive measurement of internal cardiac pressures
or intracranial pressure, await an appropriate (and un-
doubtedly clever) transducer mechanism.

based technology, because such signals are easier to
manipulate electronically. Discrete signals are usually
transmitted as a series of pulses at even (synchronous
transmission) or uneven (asynchronous transmission)
intervals. These pulses may be of equal duration, or the
information can be encoded into the pulse length. Within
the digital domain, many different encoding schemes can
be used. For encoding alphanumeric characters, those
featured on the keys of a computer keyboard, the ASCII
(American Standard Code for Information Exchange)
code is used. Here each letter, the numbers 0 through 9,
and many characters are encoded into an 8-bit binary
number. For example, the letters a though z are encoded
as 97 (for a) through 122 (for z) whereas the capital letters
A through Z are encoded by numbers 65 ( A ) through 90
(Z). The complete ASCII code can be found in some
computer texts or on the Internet.
In the continuous domain, information is encoded in
terms of signal amplitude, usually the intensity of the
signal at any given time. For an electronic signal, this
could be the value of the voltage or current at a given
time. Note that all signals are by nature time varying,
because a single constant value contains no information.
(Modern information theory makes explicit the differ-
ence between information and meaning. The latter de-
pends upon the receiver; that is, the device or person for
which the information is intended. Many students have
attended lectures with a considerable amount of in-
formation that, for them, had little meaning. This text
strives valiantly for both information and meaning.) If the
information is linearly encoded into signal amplitude, the

2.3.2.1 Signal encoding

Given the importance of electrical signals in biomedical
engineering, much of the discussion in this text is based on
electrical or electronic signals. Nonetheless, many of the
principles described are general and could be applied to
signals carried by any energy form. Regardless of the
energy form or specific variable used to carry information,
some type of encoding scheme is necessary. Encoding
schemes vary in complexity: human speech is so complex
that automated decoding is still a challenge for voice-
recognition computer programs. Yet, the exact same in-
formation could be encoded into the relatively simple
series of long and short pulses known as Morse code.
Most encoding strategies can be divided into two
broad categories or domains: continuous and discrete. The
discrete domain is used almost exclusively in computer-