User:GNLEWIS89
From MariachiWiki
Contents |
Gabrielle's World
HEY WORLD, You have entered Gabrielle N. Lewis’ page. My hobbies include watching television esp. reality shows, reading books, hanging out with my friends, surfing the web, listening to music and dancing…the basic teen activities. Words that I would use to describe myself are driven, hard-working, and entertaining. I was born on November 6, 1989 in Brooklyn, New York. My essence comes from my parents and my siblings, five sisters and two brothers, whom are all older than me. My family emigrated from Jamaica in the 1970’s in search of a better means of living. They are the reason why I work so hard in everything that I do. As a 1st generation American, I have the best of both worlds because I have the advantage of opportunity and the constant reminder that anyone can overcome life’s struggles.
I graduated from one of New York City’s Specialized High Schools, Brooklyn Technical High School. As a freshman at the State University of New York at Stony Brook, I am in the WISE (Women In Science and Engineering) Program and in the Clinical Laboratory Sciences major. My goal is to obtain my Ph.D in Forensic Science. Forensics has always fascinated me. When my grandmother passed away in 1997, I converted my sorrow into an interest of how people died. While watching shows such as The First 48, Forensic Files, Dr. G: Medical Examiner, my love of Forensics grew even greater. I joined the Clinical Laboratory Sciences major because it has a Forensic Medical Diagnostics field, which I feel will further me in my career. I believe the WSE 187 MARIACHI project will broaden my learning experience outside of the Forensics field, giving me an opportunity to learn a fraction of Physics, a major change from my required subjects, consisting of mainly of heavy Chemistry, Biology, and Calculus.
My interests
This is a picture of the Jamaican Flag. I am a very cultured person. My past makes me the person I am today. “The past is a source of knowledge, and the future is a source of hope. Love of the past implies faith in the future.” - Stephen Ambrose
FORENSICS, my passion...
I grew around to the hip hop movement. I listen to many genres of music including Reggae, R & B, Pop, etc., but I love hip hop because the MC's have an unlimited amount of talent to create metaphors, similes, personification and other parts of speech at the drop of a dime...that's real talent...
These are some of the websites I visit on a daily basis...
Day One
Jay, our instructor for the MARIACHI project, introduced us to the dynamics and essentials of the experiment. He explained in depth terminology and methods that we will be using in the lab to obtain our general objective, which is to detect and classify events based on radio signals. The ultimate goal of the MARIACHI project is being able to detect and identify signals due to UHECRS, or Ultra-High Energy Cosmic Rays. (A cosmic ray is a high energy particle that comes from “outside of the atmosphere.”) We also learned the steps, in which to acquire data acquisition and signal processing. (1) The antenna captures a wide signal. In the lab, we use the KMA 50500 antenna and the VHF/UHF Log Periodic Dipole Antenna, or LPDA. (2) The radio receiver transfers a small band of this signal to the (3) sound card, which, in turn, digitalizes the signal. (4) This digital signal is synchronized by the GPS and (5) then transferred in to the computer.
Day Two
Today was a continuation of the radio system’s setup. We also learned how to set up a DAQ. (1) Connect Antenna with the receiver. (2) Connect Audio Output with the sound card. (3) Open the Control Panel.
Day Three
We had a chance to see MARIACHI’S radio system in action. Using a program called, WinRadio G313, we recorded our own radio frequencies. The system included many broadcast such as, FM, AM, Aeronautical, TV, etc.; however, we only recorded frequencies in the AM broadcast because it was the easiest for our antenna to pick up.
Day Four
Today, we utilized a program, MatLab, to generate signals and listen to sounds.
Generate Signal with MatLab
(1) Using the array pattern "array=start:step:stop", we generated an array, t, which atart at 0 and stop at 1 with step 0.01. Which looked like this:
>> t=0:0.01:1
(2) We generated a sinusoid signal y(t) with t that was generated in the last step:
>> y(t)=sin (2*pi*ft) f=2
(3) We plotted y(t) with the plot function in MatLab.
We played around with the frequencies and with the "step" in order to observe that the higher the frequency, the less periods the graph will have.
Listening to the sound
(1) Using the syntax, "sound (sampledSignal,frequency)", we sent the previously generated one-dimensional array to the audio card in our PC. Using the values in "sampledSignal" real numbers in the range of [-1.0,1.0], we listened to various intervals,(by changing the "stop" previously mentioned) of the frequency 8000 Hz. The signal used was y(t)=sin (2*pi*8000t).
At .01 sec (or t=0:0.1:0.01), the tone was very short and pitched. At 2 seconds (or t=0:0.1:2), the tone was much longer but had the same sound as the previous tone.
(2) We listened to the sound by changing amplitudes. For example, we typed in "sound(0.5*sampledSignal,frequency)" to get the amplitude of signal to 0.5. Then, we changed the amplitude of signal to 2, resulting in "sound(2*sampledSignal,frequency). The difference between these two signals is that the latter was much louder, and higher.
(3) We listened to the sound by changing frequency. Using the same formula of "sound(sampledSignal,frequency)", we divided the frequency by two and then multiplied it by two, resulting in two different sounds. The first sound of the signal that was divided by 2 had bass and was very low, but loud. The secound signal had a much higer and louder sound.
Day Five
Today, we added noise to the sound we had produced in Day Four, saved and loaded .WAV file through MatLab, and worked on a small project.
Noise additive to Sound
(1) To add Gaussian Noise to a sound, we had to use the syntax "randn(N,M)". We generated a sound signal "y(t)=sin(2*pi*830*t)", with the sampling frequency of 8000Hz and a 2 second interval.
(2) Using this function, we generated a noise signal having the same length as y(t). “randn(1,length(y))*0.1”
(3) Then, we listened to the noise by inserting “sound(randn(1,length(y))*0.1, frequency)” into MatLab. This produce a sound similar to static heard on the radio.
(4) We listened to the sound of our signal that we generated on Day Four using the syntax, “sound(sampledSignal,frequency)”, which resulted in a long, high pitched sound. After this, we added both sounds together using the syntax “sound(sampledSignal+randn(1,length(y))*0.1,frequency).” This produced both the long, high pitch and the static heard in the previous step.
(5) We also listened to the static and pitch at different frequencies. At a frequency of 0.5, “sound(sampledSignal+randn(1,length(y))*0.5,frequency)”, we heard a louder and clearer noise than previously. At a frequency of 0.01, “sound(sampledSignal+randn(1,length(y))*0.01,frequency)”, we heard a sound with no static because the Gaussian Noise did not register at this frequency.
WAVE File Saving and Loading
During this part of the assignment, we inserted sounds from other files outside of MatLab, such audio samples from the PC, into the system in order to be heard and eventually plotted.
(1) Using the syntax, “[readSignal, frequency] = wavread(wavFileName)”, we listened to various files under various frequencies.
(2) We listened to ‘1.wav’ by inserting:
“wavwrite (y(t),8000,’1.wav’)”
t:0:1/8000:2
“sound(wavread(‘1.wav’))”
“y= wavread (‘1.wav’))"
"[y,c]= wavread ('1.wav')"
MatLab yields c=8000 as the frequency, then produces a short beep.
(3) We repeated these steps with the file ‘2.wav’ and MatLab calculated a frequency of 4000, and with the file ‘3.wav’ (this produced a louder beep) and it calculated a frequency of 16000 (this produced a higher but shorter beep).
Small Project
Our small project was to record a 10 second radio signal, using WinRadio G313i, from an AM station.
Day Six
Today, we used the Graphical User Interface (GUI) to study the time and frequency representaions of sound signals. GUI is a user interface built with graphical objects, such as buttons, text fields, sliders, and menus. The goal of the GUI program is to verify signals generated in previous classes and learn frequency characteristic of signals.
(1) We loaded a file at a specified start and stop time into the GUI program and invoked the file in order to play it.
(2) We then generated and plotted 3 types of signals, which were pure sinusoids, pure noises and mixture of sinusoids and noises.
The image above is sinusoid GUI 5, as we called it in the lab. We generated it by changing the GUI interface to the following settings: "SamplingFreq=200, Start= 0, Stop= 1, Comp= 1, and F1= 20."
The image above is sinusoid GUI 6, which has a noise signal. We generated it by changing the GUI interace to the following settings: "SamplingFreq= 200, Start= 0, Stop= 1, Comp= 1, F1= 20, and NoiseLevel= 0.5."
(3) We also plotted two Windows .WAV file for fun.
The above graph is of BLIP.wav.
The above graph is of Chimes.wav.
WISE PROJECT (Days Seven and Eight)
Day Seven
(Task 1)
We used what we have learned in the MARIACHI Lab in our ending project. First, we made time plots of the data of the signals that the NSL Lab recorded Dec. 15, 2007. Data was recorded every 0.1 second up to 5 minutes. We organized this data in Microsft Excel and then plotted the signal power (in decibel) vs. time, the signal power (not in decibel, but in watts), the estimated noise power (in decibel) vs. time, and the estimated noise power (not in decibel, in watts) vs. time.
a) The above image is the graph of Signal Power (in dB) vs. Time (in seconds). Excel terms "Plot C and B".
b) The above image is the graph of Signal Power (not in dB) vs. Time (in seconds). Excel terms "Plot E and B".
c) The above image is the graph of Estimated Noise Power (in dB) vs. Time (in seconds). Excel terms "Plot D and B".
d) The above image is the graph of Estimated Noise Power (not in dB) vs. Time (in seconds). Excel terms "Plot F and B".
Day Eight
(Task 2)
Aforementioned, we organized the data given by the signal. The Excel sheet resembled this format:
where column A is the Date; column B is the Time (in seconds); column C is the Signal Power (in dB); column D is the Estimated Power (in dB); column E is the Signal Power (not dB); column F is the Estimated Noise Power (not dB).
In Task 2, we had to compare the Signal Power (not dB) with column G, which was the Estimated Noise Power (not dB) multiplied by 10^(6). Basically G= F X (10^6). We performed this 'extensive' mathematical function to determine the signal with a threshold (which is 10^6 times bigger than the estimated noise at the same instant). If the signal power is greater than the threshold at that instant, then the mark detection indicator would be 1, other wise it is 0. To represent this logic, we used Column H,the detection marker and in Excel, we utilized the logic "IF(E#>G#,1,0)". # = the number corresponding to that cell. Then, we plotted the detection indicators with time. Excel terms "Plot H and B."
(Task 3)
We also had to count the number of events in this 5 minutes. The continuous 1 detection indicators is counted as one event. For example,
01101001
should be counted as three events.
011 01 001
We counted these events by using this logic in Excel as column I = "IF(AND(H# =0,H# =1),1,0". The count was 125. Then, we had to modify our detection indicators with a 101-111 rule, meaning that if there is 101 in the detection indicator then we had to change it into 111. So the same example will look like...
01111001
We had to plot this modified detection indicator with time; the number of these particular events were 111. Through Excel, we calculated this by making column J = "IF(AND(H# =1,H# =0,H# =1),1,0)". This means the program will find the 101 sequence in column H. Then, we made a column K = SUM(H#,J#). So the 101 will add to the 1 of each event, to give 111. Finally the logic of "IF(AND(K# =0,K# =1),1,0)"= L's column, give us the number 111 events that occur in this 5 minutes. This count is 111. Finally, we plotted the modified detection indicators with time. Excel terms "Plot K and B."
