week 4.1 Flashcards
biosignals
variations in energy as a function of space and time
what is a measured signal composed of?
biosignal of interest and noise
(biosignal of interest is the device recording the signal, like ECG, EKG, etc)
noise
any variation to quantity being measured not related to process of interest
encoding
part of processing measured signal, can be divided into coninuous (analog) and discrete (digitial)
continous/analog signal
information is coded in terms of amplitude
signal has defined amplitude at every point in domain
dumbed down: the y is amplitude and signal is always being measured
what is the domain of a continuous/analog signal?
it’s uncountable and may or may not be finite. signal will have some value at every point
-> function doesn’t need to be continous
examples of continuous/analog signal
temperature, sine wave, etc
discrete/digital signal
defined over a discrete time domain, therefore has a countable domain
signal measurement system
analog to digital
sampling is done by slicing up to sequence digital numbers at time intervals Ts
dumbed down: chop up analog into time intervals (Ts) for digital
most biosignals are digitized by the data collection system:
data acquisition and analog to digital conversion
digital signals is a series of…
numbers
x[k] = x0, x1, x2, x3, x4… xN
what does x[k] represent?
the value of the analog signal at discrete points in time
how is the sample interval (Ts) determined
t = n * Ts = n / fs
what is fs
sampling frequency
fs = 1/ Ts
what is n
position of the number in the sequence
digitizing of analog signal
requires slicing of signal in time and amplitude
what is the functions of analog and digital signals?
analog: x(t), x is a function of time
digital: x[n], x is a function of sampled numbers n
analog to digital (A/D board)
16 bit, multifunction I/O device
LSB
least significant bit, minimum change in voltage required to guarantee a change in output
(measure of accuracy)
V LSB equation
V LSB = V max / 2^Nbits - 1
systematic error
each measurement deviates from true value by a fixed amount
random error
noise: unpredictable variations in measured signal from moment to moment
examples of noise
building vibrations, air currents, electric power fluctuations, stray radiation from nearby electricals
physiological variability noise
what you’re measuring can vary from other mechanisms
solution: modify approach
environmental noise or interference
caused by other source of similar energy effecting signal of interest
ex: fetal ECG signal is corrupted by mother’s ecg signal
arifiact
caused by transducer response to other energy variation
ex: motion artifact, stimulus artifact
solution: design better transducers
electronic noise
caused by known sources and characteristics
ex: white noise
solution: transducer/electronic design
two sources of electronic noise
thermal or johnson noise (resistors
shot noise (current noise, like voltage barriers of semiconductors)
white noise
has similar energy levels over all frequencies range
error can be calculated by
subtracting mean from the original data set
variance
signal variability irrespective of its average
matlab command var(x)
standard deviation
square root of variance
sinusoidal
a sign graph
periodic signal = x(t) = x(t + T)
t = time
T =period in seconds
what information about the variability of a signal does the average provide?
None. it does not give us any info
root mean squared (RMS) acronym
S quare the signal
M ean (calculate)
R oot (sqrt)
effective value (RMS) of AC voltage is ___ to the DC voltage
equivalent
does a lamp with 6V rms AC supply have the same brightness as a lamp with 6V DC supply
yes
