Lecture 2 Signal Characteristics

Goals: Introduce analog and digital signals Use basic descriptors of time dependent data Understand the relationship between time and frequency domains Determine the Fourier series of periodic time dependent measurands

Input-Output Signal Concepts

Tasks that face the engineer in the measurement of physical variable Selecting a measurement system Interpreting the output from the measurement system

Digital World
What is a digital signal and what differentiates it from other signals? All electronic signals (and thus circuits) are either analog or digital. Both are voltage (or current) signals. An analog signal is continuous; i.e., infinitely divisible into smaller and smaller parts. A digital signal is quantized; i.e., the information is divided into discrete quantities of a finite size. Up to c. 1950, prior to the advent of the transistor (and thereafter the computer) all signals were essentially analog.

Classification of Waveforms
Any signal is either a digital or analog signal the difference is obvious in their appearance.

Analog Signals
Analog signals are always continuous (there are no time gaps). The signal is of infinite resolution.

Discrete Time Signals

Digital signals Information about the signal magnitude is available only at discrete points in time

The magnitude of the signal is continuous and thus can have any value within the operating range

Are particularly useful when data acqusition and processing performed by using a digital computer

Analog versus Digital Examples

Thermocouple The thermocouple provides a continuous (analog) signal into the meter, which is digitized and displayed on a LCD panel.

Sampling: The process of obtaining discretized information from a continuous variable at finite time intervals.

Classification of Waveforms (cont.)

Static: steady in time

Dynamic: changing in time Simple periodic Complex periodic Step Ramp Pulse

Step Simple periodic Ramp

Pulse
Complex periodic

Classification of Waveforms (cont.)

Non-deterministic: random or chaotic

Signal Characteristics: Definitions

Magnitude - generally refers to the maximum value of a signal. Range - difference between maximum and minimum values of a signal. Amplitude - indicative of signal fluctuations relative to the mean. Frequency - describes the time variation of a signal. Dynamic - signal is time varying. Static - signal does not change over the time period of interest. Deterministic - signal can be described by an equation (other than a Fourier series or integral approximation). Non-deterministic - describes a signal which has no discernible pattern of repetition and cannot be described by a simple equation. Mean - average or static portion of a signal over the time of interest. Sometimes call the dc component or the dc offset of the signal [Excel tip: Mean = AVERAGE(numbers...)]. RMS - root-mean-square - average value of the square of the signal over the time of interest. [Excel tip: RMS = SQRT(SUMSQ(numbers1 to n)/n)]

Signal Analysis
There are essentially two ways to analyze a time depending signal (considering that a time independent signal, i.e., constant is trivial) Fourier Analysis The signal is reconstructed using trigonometric terms; this allows one to reproduce the signal and say something about its spectral (frequency) content. Statistics The signal is characterized by its statistics (max, min, mean, median, deviation, etc.) which allows one to say something about its typical behavior

Average or Mean Value Provides a measure of the static portion of a signal over a period of time. This is often referred to as the dc component or dc offset.

Signal Analysis (cont.)

The average value or mean value of Analog (continuous) is found by. The mean value of discrete time signal is found by.

The rms value of any continuous analog signal y(t) over period of time.

The rms value of any discrete time signal y(t) over period of time.

The fluctuating portion alone is often characterized by the a term called the variance, , or the square root of the variance the standard deviation, .

Where, y', is the true mean value of the signal.

Signal Analysis (cont.)

Simple Harmonic (SH): data which vary as a sine or cosine function of time

Works well for a simple function. But what about something more complex, such as a waveform composed of two or more different sine waves?

Complex Signals can be represented by the addition of a number of simpler periodic functions, as an example a white light is combined of colors in the spectrum and of simple Periodic functions

Any complex signal can be thought of as being made up of sines and cosines of differing periods and amplitudes, which are added together in an infinite trignometric series.

Signal Analysis (cont.)

Complex Signals: Composed of SH signals of different frequency, amplitude, or both

This dynamic signal can be represented in the frequency domain by a frequency spectrum

Fourier Analysis
The representation of a signal as a series of sines and cosines is called Fourier Series It is used to determine the frequency spectrum of dynamic signals spectrum analyzer hardware
Fourier Series

T = 2/f is the period of the signal, f is the fundamental frequency (first harmonic), 2f is the second harmonic, etc.

Even and Odd Functions

Even functions are symmetric about the origin, odd functions are anti-symmetric Even: g(-t) = g(t) Odd: h(-t) = -h(t) i.e. cos(nt) i.e. sin(nt)

Therefore if y(t) is even the Fourier series will contain only cosine functions. If it is odd it will contain only sine functions. Functions that are neither even nor odd result in Fourier series that contain both sine and cosine terms.

The Fourier Transform

The most familiar is that which transforms the time function x(t) into the frequency function X(f) through the use of the following relationship:

Sometimes this is written as:

The original independent variable is t is for time, usually in seconds, and has the range of (to ). The new independent variable f is usually in units of hertz, abbreviated Hz. The range of f is also (- to ). Some times is used rather than f. Remember the variable is in radians per unit time. The Fourier Transform is reversible

or

Discrete Fourier Analysis

While the derivation for the FT assumed a continuous signal, in reality all signals we measure will be discrete Finite time with a finite number of points, thus, there will be a t between each sample, in practicality Sampling period = sampling record length = length of time you sampled the input signal = period of the lowest resolvable frequency = number of samples/sampling rate (SR)

Highest frequency = Nyquist frequency

Frequency resolution = lowest resolvable frequency = 1/Sampling Period

Example: 1000 point data set sampled at 100 samples/sec T = 1000/100 sec = 10 seconds t = 1/1000 sec fNyquist = fs/2 = 500 Hz. f = 1/T = 0.1 Hz . Example: To resolve a spectra with 0.5 Hz resolution up to 1 KHz what sampling rate, period and number of samples are required? f = 1/T = 0.5 Hz therefore the sampling period must be 2 seconds. fNyquist = 1 KHz = fs/2 therefore the sampling rate must be 2 KHz. f = 1/T = 0.5 Hz = fs/N = 2000 samples/sec/N therefore N = 2000/0.5 = 4000 samples Frequency amplitude ambiguity or spectral leakage is caused by this discrete frequency step size. If an input signal frequency is not a multiple of the frequency resolution it will be represented by several data set frequencies of varying amplitudes.

Discrete Fourier Analysis