[LabVIEW]The Analysis of Amplitude Modulation

Introduction:

In this lab, I have learnt the basic concepts and steps of AM. At the same time, I understand deeply the difference between envelope detection and coherent detection by construct the LabVIEW program. After complete the tasks which were required by Lab, I start to try to output the shape of the envelope and use Hilbert transform to compute the envelope directly. 

Lab results & Analysis:

1. The principle of amplitude modulation and demodulation:

Modulation is actually the shift of the spectrum (so the simplest implementation is based on signal multiplication), while demodulation is the opposite (so the simplest is still signal multiplication). However, in the shift process, because of the characteristics of signal multiplication, the intermodulation spectrum is generated, so the unfavorable frequency need to be filtered out.

1. Amplitude modulation:

Let the modulating signal be,

m(t)=Amcos(2πfmt)

and the carrier signal be,

c(t)=Accos(2πfct)

Where,

Am and Ac are the amplitude of the modulating signal and the carrier signal respectively.

fm and fc are the frequency of the modulating signal and the carrier signal respectively.

Then, the equation of Amplitude Modulated wave will be

s(t)=[Ac+Amcos(2πfmt)]cos(2πfct)s(t)

 

1. Amplitude demodulation:

1. Envelope detector is used to detect (demodulate) high level AM wave. Following is the block diagram of the envelope detector.

1. Coherent demodulation

P(t) = s(t) cos(2πfct)

   Then we can let the signal p(t) pass a LPF, the we can get the low frequency terms of the signal.

 

1. The Lab result:

This part I will show results of my program.

First of all, let’s check the parameters of the signals:

Figure 1 The parameters of the signals

And there is my program diagram, the function of each part has been written on the diagram (I had tried my best to make the elements and lines layout beautiful and neat.):

Figure 2 LabVIEW program diagram

1. The time domain of baseband signal, carrier signal and modulated signal:

Figure 3 Time domain of baseband signal, carrier signal

 

This is time domain of baseband signal, carrier signal, because our sampling frequency is not large enough, so that the figure looks jagged. And We need to adjust the size of the interval of the GUI to show the figure clearly.

 

Then, we would pay attention to the figure of modulated signal in time domain. Because this figure affects the performance of the envelope detection.

Figure 4 Modulated signal in time domain when ka=0.5 

We can see the waveform of the modulated signal; the waveform also shows us the envelope of the signal.

But how the modulation index (ka) influences the modulation performance?

Figure 5 Modulated signal in time domain when ka=1

Figure 6 Modulated signal in time domain when ka=1.5

 

In order to show the whole figure, I adjust the size of the interval of y-axis.

As the value of the modulation index increases, the carrier experiences a 180 degrees phase reversal, which causes additional sidebands and hence, the wave gets distorted. Such an over-modulated wave causes interference, which cannot be eliminated. So that we can see that at the third picture, the distortion of envelope of the signal occurs. This is what we don’t want to see.

1. The frequency domain of baseband signal, carrier signal and modulated signal:

Complete the time domain, we will continue to research the signals in the frequency domain.

Figure 7 Frequency domain of baseband signal, carrier signal

 

This is frequency domain of baseband signal, carrier signal, we can see the

the pulse signals show on the GUI. The baseband signal’s pulse is almost on the frequency is about 2000HZ. And the carrier signal’s pulse is almost on the frequency is about 10000HZ. This result is matched the parameters we set.

Figure 8 The frequency domain of modulated signal

 

We can see that on both sides of the carrier frequency are the upper frequency and the lower frequency. And if take a close look to the amplitude of the pulse.

Figure 9 The close look to the amplitude of the pulse

 

We will find that the amplitude of the carrier frequency is larger than both sides of itself. And we will know that the it is easier to analyze the power from the frequency domain.

1. Test and analysis the envelope detection and coherent detection:

Let’s pay attention to the time domain of signal is demodulated by the envelope detection and coherent detection.

Figure 10 The time domain of demodulated signal (order=3)

Let’s pay attention to the time domain of signal is demodulated by the envelope detection and coherent detection.

.  

Then we can check the effect of the program:

Figure 11 Baseband signal and modulated signal on the same figure

Then we look at the frequency domain of the modulated signal. And we can se that the waveform of two signals are same. So that our demodulation is almost successes.

Then we look at the frequency domain of the modulated signal.

Figure 12 The frequency domain of demodulated signal

1. The solution to the problems on the slides:

1. The bandwidth of the baseband signal is 4000HZ.

1. What is the cutoff?

The cutoff is just larger than the baseband frequency to ensure that the baseband signal terms can pass the filter.

1. Explain the frequency result in the coherent detection.

Because the multipuly of the signal, there must be some high-frequency terms occur.so that the figure will show some high-frequency terms. And the filter maybe not ideal.

1. Explain the frequency result in the envelope detection

I an not very clear to this question, I think maybe the filter maybe not ideal.

1. How the orders of the filters influence the waveform of the demodulated signal?

Figure 13 The time domain of demodulated signal (order=1)

Figure 14 The time domain of demodulated signal (order=2)

Figure 15 The time domain of demodulated signal (order=4)

Figure 16 The time domain of demodulated signal (order=5)

By obversion the waveform with different order of filter in the figure, we can find that when n is small, the order higher, the waveform better. But when order is large enough, like when the order of filter is larger than 3, the influence is not clear.

1. The topology of LPF will influence the bandwidth of the modulation. And different topology kinds will influence the demodulation efficiency of the system.

1. Feedback:

1. How to show the envelope of the demodulation, this word means that how to output the envelope line directly?

I think this problem for a long time, and I get a solution that is maybe right:

Just get the mean value of each small interval.

The mean filter over one carrier freq. period is done by a convolution with a rectangular window (or array with a constant value) where the length of the window (size of the array) is about the sample point length of one carrier frequency period.

(This week my mentor asks me to determine the project that I want to research in my team, so that I really not have time to write this solution. If I can upload this report on the weekend, I am sure that I can write the program for this method)

1. Why we don’t use the band-pass filter to make the AM to DSB?

In my view, maybe this kind of BPF is too hard to make.

Experience:

I understand deeply the difference between envelope detection and coherent detection by construct the LabVIEW program. And I I felt every step of modulation and demodulation carefully. This lab tells me that it’s not enough for me only know the knowledge about the theorem.