User:SOCCERKITTY21
From MariachiWiki
Contents |
About Me
My name is Katherine Cermak. I was born in the San Francisco Bay Area and am now a marine biology major at SUNY Stony Brook. I moved to New York to go school because I wanted a new experience. I got accepted into the Woman in Science and Engineering (WISE) program at SUNY Stony Brook which is what convinced me to attend Stony Brook. I am creating this page for my WSE 187 class. Throughout my life I have always been passionate about marine biology as well as photography. Photography is the one hobby that I continue to do. I own a Nikon N65 and three lens (a 50mm, a zoom, and a fisheye). My camera is one of my favorite things that I own. One of my favorite photographers is Ansel Adams. I used to play drums as well but recently sold my drum set. I played mostly Latin music but also really enjoyed jazz and some rock. I have one older sister (age 21) and two parents. My mother is a therapist and my father is a psychiatrist. We own one house in the San Francisco Bay Area and one in Minnesota. The cabin in Minnesota has been in my family since my mother was about three years old. It a beautiful place on a small lake north of the twin cities. When there I do a lot of water sports like water skiing, wakeboarding, and tubing. I love it in Minnesota because it is very relaxing and a way to get away from normal life. I lead an amazing life that leaves me feeling happy nearly all the time.
Day One
On the first day in the Mariachi lab we found different AM radio stations on the computer receiver and recorded ten seconds of that station. It was hard to search for the station, but once we looked up radio stations and tuned into them, it wasn't difficult anymore. I was surprised at how much static was present when we tuned into the radio stations. I thought it was very easy once you got the hang of it. The recording was as simple as pushing the record button and then the stop button. It automatically saved the recording onto the desktop which we could then drag and drop it into a new folder. Changing the station was as easy as pushing an up or a down button that changed the frequency. Overall, it was not a difficult task.
Day Two
On day two we learned about the program MATLAB. We put different matrices into the program and learned how to plot them in different ways. The most interesting thing that we did was making sin graphs and changing the number of periods that were shown. It was fun to play around with these number and change the color of the graph to make different patterns on the graph. We also learned how to make matrices with only 0s and 1s. There were a lot of things that we learned in day two and I thought that it all was more fun to be plotting the graphs and seeing results than to be tuning into the stations. Overall, this day was more interesting and fun because we actually got to see things that made sense instead of listening to a radio station mixed with a lot of static.
Day Three
This day we learned about sound on MatLab. We started with generating a sound signal with MatLab by putting in the sinusoid signal y(t)=sin(2*pi*f*t) with f=2. T was equal to an array that we had generated from 0 to 1 with a step of 0.01. We ploted the signal and changed f to get different numbers of periods in the one second. When we tried an f value of 100, the plot got very distorted because it had too many periods and not enough information because the step was too large. When we generated a new array with a step of 1/8000 we could increase the f value to 500, but the plot was impossible to read becuase there were so many periods in the one second that it filled the graph with a solid color. When we decreased the stop time to 0.01, the graph was easily readable. Next we listened to the signal using the sound function of MatLab. We generated an array and created the signal y(t)=sin(2*pi*440*t) where t=the array. then we listened to the signal using the command sound(0.5*y(t), 8000) with y(t) being the signal and 8000 the frequency. When we changed the 0.5 to 2, the signal was louder. Dividing the frequency by 2 made the wavelength longer so the signal lasted longer and was a lower tone because it had a smaller frequency. Multiplying the frequency by 2 did the opposite. The wavelength decreased, making the signal last a shorter time, and the frequency increased, making the tone higher pitched.
Day Four
Today we learned to add noise to the sounds that we generated before. We used the randn function of MatLab for the first time which gnerates a Gaussian Noise, which sounds just like static. After generating the sound signal y(t)=sin(2*pi*830*t) where t=an array from 0-2 with a step of 1/8000, we used the randn function to generate a noise signal with the same length, randn(1,length(y))*0.1. Then we listened to sound(randn(1,length(y))*0.1, 8000) because the frequency was equal to 8000. This sounded like a quiet static sound. Then we listened to the signal that we had generated before, sound(y(t)=sin(2*pi*830*t), 8000) which sounded just like a tone. Then we added the two together and listned to them both together, sound(y(t)=sin(2*pi*830*t) + randn(1,length(y))*0.1, 8000). This sounded just like it should, like the static with an underlying tone. When we changed the 0.1 that the randn was multiplied by to 0.5, the static got louder. Changing the 0.1 to 0.01 made it so that we could not hear the static at all. Next, we learned about WAVE files in MatLab. We used the wavwrite function to write the file and the wavread function to make it into a readable and playable sound on MatLab. We generated a WAVE file of the y(t) signal we had generated before and saved it as 1.wav. Then we read it and listened to it. It sounded the same as the original sound. We continued with this with changing the frequency so that it would be lower and higher and then added the two to make a complex signal. That was all.
Day Five
Day five consisted of taking notes and experimenting with the time and frequency representations of sound signals. We launched GUI (graphical user interface) and learned to generate, load, play, and plot different signals in time and in frequency representations. Increasing the frequency of my previously recorded radio clip would change the pitch and the length of the clip. A larger frequency shortened the clip and made it a higher pitch. A smaller frequency lengthened the clip and made it a lower pitch. The periods in the sinusoid graph were equal to the number of F1. A smaller F1 made the peak in the frequency move to the left and a larger F1 made the peak move to the right. When you made there be two components and add a F2, then there are two peaks on the frequency domain and the sinusoid becomes a bit more jagged on the time domain. The noise only signal with a noise level of 0.5 that we generated next was a complex, very jagged graph on both the time and the frequency domain. The different noise levels would change where the peaks and dips are as well as changing the amplitude, a larger noise level would create a larger amplitude. When we created a 1-component sinusoid with a noise signal, four different graphs were created, two in the time domain (a sinusoid and a noise signal) and two in the frequency domain (a sinusoid and a noise signal). In the time domain, the sinusoid was uniform whereas the noise had no uniformity and instead was jagged and random. In the frequency domain, the sinusoid had one peak whereas the noise was jagged and random with a small dip where the peak in the sinusoid graph was. Next we created graphs of different windows system wave files. With the normal frequency, the noise signal graph was very dense and jagged in both the time and the frequency domains. The sinusoid graph was uniform in the time domain, but in the frequency domain, you have to zoom in to see the peak that is located around the 20 mark. When we changed the frequency to a smaller frequency, the noise signal became less dense, but still just as random. The sinusoid graph was still uniform in the time domain and in the frequency domain, you can actually see the peak now. And that was day five.
Small Project
The following two plots are of a 10 second recording of a radio signal that I recorded using the WinRadio G313i receiver on the computer. I found out that the frequency was 16000 Hz by using the wavread function on MatLab. After reading the signal, i was able to play it on MatLab and also plot it on MatLab.
Original signal for full 10 seconds
Original signal for 0.1 seconds
Data Processing Project
In this project we learned about how to look at real data from a radio signal in different ways. First, we opened a file that was pre-recorded and graphed the signal power and the estimated noise power in dB and not in dB. There was a distinct difference between the two types of graphs. When the graph was not in dB, there was less amplitude variation, but a lot of peaks. The graph in dB, had less peaks, but more variation in the amplitudes. Then we converted all of the data out of dB and made the signal power 10^.6 times larger. Then we could change the two arrays into one that showed which one was greater. If the signal power was greater than the threshold, it was a 1. If it was not bigger, then it was a 0. The following is a graph of the detection indicators vs. time.
Next we found out how many times there was a group of 1s by finding the difference between the numbers. If there was two 0s or two 1s next to each other, than the difference was 0, but if there was a 0 then a 1, the difference was 1 and if there was a 1 and then a 0, the difference was -1. We could then find all of the times that the difference was 1 and count them up. This was the number of total events. We did this for all of the recordings that were in the file, a total of 24 for every hour of the day. This next graph is of the total number of events vs. time.
Next we detected where there was any patterns that were 101 and changed them to 111 so that there were no two events within 0.1 seconds of each other. We then graphed this modified number of events vs. time.
