Midterm Study

Question 1

What is the DSP top level block diagram?

Answer 1

Transducer connected to the body > A/D converter > DSP> DAC converter > informative display

Question 2

Why do we rarely need a FPGA in biomedical applications?

Answer 2

The human body is very bandwidth limited.

Question 3

Most signals in the real world are what type of signals?

Answer 3

Analog signals

Question 4

What do transducers do?

Answer 4

They provide continuous electrical signals

Question 5

What do processors require?

Answer 5

Bits of data that represent logic ones and zeros

Question 6

What is the quantization error?

Answer 6

The difference between he quantized level and the analog signal

Question 7

Equation for quantization error?

Answer 7

(q^2)/12

Question 8

What does the quantization noise term tell us?

Answer 8

The error term is a function of the smallest step size possible

Question 9

What is dynamic range?

Answer 9

The range over which we can recover a signal

Question 10

Largest signal we can detect?

Answer 10

Given: 2^n quantization bands each at q wide, then maximum sine wave that can be encoded is (2^n)*q

Question 11

What is the smallest signal we can detect?

Answer 11

q, anything less ends up in lowest band

Question 12

How would you convert the resultant dynamic range into dB?

Answer 12

20 log (2^n)

Question 13

ADC resolution limitations result in what?

Answer 13

quantization errors

Question 14

increasing the sample rate also increases what?

Answer 14

bandwidth able to use

Question 15

What if the sampling rate is not sufficiently high?

Answer 15

Signals alias into digital systems at a false frequency, end up having an entirely differnt frequency present

Question 16

The minimum sampling rate is called what?

Answer 16

Nyquist Rate

Question 17

How to get the Nyquist rate?

Answer 17

Take sampling rate and divide by two. The positive and negative of this number is the Nyquist frequency range.

Question 18

Fast fourier transforms put the frequency from the

Answer 18

time domain to the frequency domain

Question 19

Aperiodic-Continuous

Answer 19

Analog signals that extend from negative infinity to positive infinity without repeating. (decaying exponentials would meet this criteria) The Fourier Transform is applied to these type of signals

Question 20

Periodic-Continuous

Answer 20

Sine waves or any waveform that repeats periodically from negative to positive infinity. The Fourier Series is applied to enable the Fourier transform

Question 21

Aperiodic-Discrete

Answer 21

Discrete digital signals that do not repeat between positive and negative infinity, such as a digitized decaying exponential. The type of Fourier Transform that is applied to this type of signal is the Discrete Time Fourier Transform.

Question 22

Periodic-Discrete

Answer 22

Discrete signals that repeat themselves between positive and negative infinity. The class of Fourier Transform that is applied is the Discrete Fourier Transform

Question 23

What is the equation of the Fourier Transform

Answer 23

f(t) = av + Σ an cos n w0 t + bn sin n w0 t

Question 24

What are the two properties of Fourier Transforms

Answer 24

homogeneity: change in amplitude in one domain = change in amplitude in other domain
Additivity: addition in one domain = addition in other domain

Question 25

What is the digital frequency?

Answer 25

how many samples we have per the fundamental frequency

Question 26

What is the equation for digital frequency?

Answer 26

F = f/ fs = f*ts fs = sampling frequency ts = sampling period

Question 27

if you had a frequency of 10 Hz and it was sampled 5 times per cycle, then you would need a sampling frequency of...

Answer 27

F= ts*f = 5 times per sample * 10 Hz = 50 Hz

Question 28

If a 50 Hz signal is sampled at 250 Hz, what is the digital frequency?

Answer 28

F = 50 / 250 = 0.2 cycles per sample or (0.2*2pi) = 1.26 rad per sample

Question 29

What does the DIscrete Fourier Transform do?

Answer 29

Transforms data between the time domain and frequency domain

Question 30

What does the P wave correspond to?

Answer 30

Atrial depolarization

Question 31

What does the sinoatrial node initiate?

Answer 31

electrical impulse

Question 32

What does the QRS complex represent?

Answer 32

Ventricular depolarization (electrical signal that causes ventricles to contract)

Question 33

Where does the elecrical impulse travel from?

Answer 33

atrioventricular node

Question 34

Q wave is initial _______, R wave is the __________ and the S wave is the __________

Answer 34

downward deflection, upward deflection, downward deflection

Question 35

What does the T wave represent?

Answer 35

ventricular reolarization (recovering electrically after contraction)

Question 36

What happens to the heart muscle cells during repolarization?

Answer 36

return to their resting electrical state

Question 37

What is a filter's purpose?

Answer 37

provide frequency selectivity characteristics

Question 38

What is the pass band?

Answer 38

frequencies where minimal input signal attenuation is provided

Question 39

What is the stop band?

Answer 39

The frequency at which the filter attenuates

Question 40

What is the gain equation?

Answer 40

Gain (dB) = 20 *log(gain) = 20*log(output/input)

Question 41

What does a low pass filter do?

Answer 41

passes low frequenices and attenuates high frequencies

Question 42

What happens when there is linear phase shift?

Answer 42

The filter's time delay is constant with respect to frequency

Question 43

What happens when there is no linear phase shift?

Answer 43

The input waveform will be distorted when it emerges from the filter

Question 44

What can filtering to tighter bandwidths do?

Answer 44

Reduce noise levels and make signals easier to detect.

Question 45

The basic ECG filter is what?

Answer 45

A low pass filter to pass the P,T and QRS waves while rejecting higher 60 Hz noise frequency

Question 46

What is convolution?

Answer 46

A mathematical approach to combine two signals, considered the tecnique upon which DSP is based.

Question 47

What do we do when we convolve two functions?

Answer 47

We multiply one by a flipped and shifted version of the other and sum

Question 48

What is the equation for convolution?

Answer 48

Y(n) = Σ Bk * x(n -k)