CSE 562: Mobile Systems & Applications Quals Course – Systems Area Shyam Gollakota
First Mobile Phone 1973 2
Goal of this course • Have an understanding of state of the art mobile systems research • Explore applications that are capable with mobile devices 5
Course material 1. Signal processing fundamentals 2. Acoustic device and device-free tracking 3. Physiological sensing using phones and speakers 4. IMW tracking and GPS localization 5. Wi-Fi localization and sensing 6. Designing and building IoT device hardware 6
Course material 7. Backscatter systems 8. Mobile privacy and security 9. Robotics mobile systems 7
Grading 3 hands-on assignments (20+20+20% in all) • One every two weeks • Requires programming phones, microcontroller, etc. Class presentation of one paper (10%) Final research project (30%) • Proposal due on May 1 • 2-3 person project 8
Signal processing basics (Slides by Nirupam Roy)
Model for a signal (frequency, amplitude, and phase)
Model for a signal (frequency, amplitude, and phase) A . sin(𝜄) A . 𝑑𝑝𝑡(𝜄)
Model for a signal (frequency, amplitude, and phase) sin(𝜄) 1 � cos(𝜄) 𝑔 𝑑𝑧𝑑𝑚𝑓𝑡 𝑞𝑓𝑠 𝑡𝑓𝑑𝑝𝑜𝑒 𝜄 = 2𝜌𝑔𝑢 2𝜌 𝑏𝑜𝑚𝑓𝑡 𝑞𝑓𝑠 𝑑𝑧𝑑𝑚𝑓
Frequency, Amplitude, and Phase 2𝜌 𝑏𝑜𝑚𝑓𝑡 𝑞𝑓𝑠 𝑑𝑧𝑑𝑚𝑓 𝑔 𝑑𝑧𝑑𝑚𝑓𝑡 𝑞𝑓𝑠 𝑡𝑓𝑑𝑝𝑜𝑒 A � Phas time = t second e A . sin(2𝜌𝑔𝑢) A . sin(𝜄) = = A . sin(2𝜌𝑔𝑢 + 𝜚) -- with initial/additional phase ϕ
Frequency, Amplitude, and Phase Frequency: 2Hz Amplitud e Time (sec) 1.0 1.2 0.0 0.2 0.5 0.7 0 5 0 0 5 5 Amplitud Frequency: 4Hz e Time (sec) 1.0 1.2 0.0 0.2 0.5 0.7 0 5 0 5 0 5
Frequencies of an arbitrary signal Frequency: 2Hz Amplitud e Time (sec) 1.0 1.2 0.0 0.2 0.5 0.7 0 5 0 0 5 5 Amplitud e Time (sec) 1.0 1.2 0.0 0.2 0.5 0.7 0 5 0 5 0 5
The concept of the Fourier series Amplitud e Time (sec) 1.2 0.7 1.0 0.0 0.2 0.5 5 5 0 0 5 0 Amplitud e Time (sec) 1.2 0.7 1.0 0.0 0.2 0.5 5 5 0 0 5 0
Time Domain and Frequency Domain Approx. square wave 1 . sin 2𝜌𝑔𝑢 1 3 . sin 2𝜌. 3𝑔𝑢 1 5 . sin 2𝜌. 5𝑔𝑢 1 7 . sin 2𝜌. 7𝑔𝑢
Time Domain and Frequency Domain w T e i i v m n e i a d m o o m d a y i n c n v e i u e q w e r F
Analogy: Food coloring chart Basis for food colors
Analogy: Food coloring chart Basis for food colors
Analogy: Food coloring chart Basis for food colors
Analogy: Food coloring chart Basis for food colors
Time Domain and Frequency Domain Amplitud e Time (sec) 1.2 0.7 1.0 0.0 0.2 0.5 5 5 0 0 5 0 = A 1 . sin 2𝜌𝑔 1 𝑢 + B 1 . 𝑑𝑝𝑡(2𝜌𝑔 1 𝑢) + A 2 . sin 2𝜌𝑔 2 𝑢 + B 2 . 𝑑𝑝𝑡(2𝜌𝑔 2 𝑢) + A 3 . sin 2𝜌𝑔 3 𝑢 + B 3 . 𝑑𝑝𝑡(2𝜌𝑔 3 𝑢) + …
Time Domain and Frequency Domain Fourier Transform Amplitude Amplitude FFT IFFT 0 2 4 6 8 0.0 0.5 0.2 Frequency (Hz) 0 0 5 Time Time domain Frequency domain FFT = Fast Fourier Transform IFFT = Inverse Fast Fourier Transform
Frequency band Amplitude 0 2 6 4 8 Frequency (kHz) A 4 kHz frequency band starting at 2 kHz What is bandwidth? What is center frequency?
Spectrogram
Spectrogram [ [ FFT of overlapping windows [ [ of samples FFT (Spectrogram) FFT FFT FFT
Spectrogram
Analog Digital Spectrogram plot Physical signal on computer (voice) FFT Time varying voltage signal A collection of numbers
Analog vs Digital World Temperature Voltage Thermo-couple Time Analog Digital
Analog Digital Spectrogram plot Physical signal on computer (voice) FFT Time varying voltage signal A collection of numbers ?
Analog Digital Spectrogram plot Physical signal on computer (voice) FFT Time varying voltage signal A collection of numbers ADC Analog-to-Digital Converter
Sampling theorem Amplitud e Time (sec) 1.2 0.7 1.0 0.0 0.2 0.5 5 5 0 0 5 0
Sampling theorem Amplitud e Time (sec) 1.2 0.7 1.0 0.0 0.2 0.5 5 5 0 0 5 0 Amplitud e Time (sec) 1.2 0.0 0.5 0.7 1.0 0.2 5 0 0 5 0 5 [ 0.34 0.22 0.09 0.21 0.30 0.08 0.09 ]
Sampling theorem Amplitud e Time (sec) 1.2 0.7 1.0 0.0 0.2 0.5 5 5 0 0 5 0 f s = 1/T = Sample rate (or sampling T = Sampling interval frequency) Clock
1-dimensional sampling
1-dimensional sampling 2-dimensional sampling
1-dimensional sampling 2-dimensional sampling 3-dimensional sampling
Sampling theorem Amplitud e Time (sec) 1.2 0.7 1.0 0.0 0.2 0.5 5 5 0 0 5 0 Amplitud e Time (sec) 0.7 1.0 1.2 0.0 0.2 0.5 5 0 5 0 5 0
Sampling theorem Amplitud e Time (sec) 1.2 0.7 1.0 0.0 Aliasing: Two signals become indistinguishable 0.2 0.5 5 5 0 0 5 0 after sampling Amplitud e Time (sec) 0.7 1.0 1.2 0.0 0.2 0.5 5 0 5 0 5 0
Aliasing Amplitud e Time (sec) 1.2 0.7 1.0 0.0 0.2 0.5 5 5 0 0 5 0 Amplitud e Time (sec) 0.7 1.0 1.2 0.0 0.2 0.5 5 0 5 0 5 0
Aliasing
Aliasing in real life https://www.youtube.com/watch?v=QOwzkND_ooU
How to find a good sample rate?
How to find a good sample rate? Nyquist sampling theorem: In order to uniquely represent a signal F(t) by a set of samples, the sampling rate must be more than twice the highest frequency component present in F(t). If sample rate is f s and maximum frequency we want record is f max , then f s > 2f max
Amplitud e Time (sec) 1.0 1.2 0.0 0.2 0.5 0.7 0 5 0 0 5 5 Nyquist frequency = Maximum alias-free frequency for a given sample rate. Nyquist rate = Lower bound of sample rate for a signal
Nyquist f max = 12 Hz f max = 6000 Amplitude Amplitude Hz 0 4 12 8 16 0 2 6 4 8 Frequency (Hz) Frequency (kHz) A 4 kHz frequency band starting at 2 kHz
Commonly, the maximum frequency in human voice is 4 kHz, what sample rate will you use in your audio recorder?
Aliasing: A real life scenario Nyquist frequency Amplitud e 10k 20k 30k 40k 50k 60k 70k 80k 90k 100k Frequency spectrum
Aliasing: A real life scenario Nyquist frequency Amplitud e 10k 20k 30k 40k 50k 60k 70k 80k 90k 100k Frequency spectrum
Aliasing: A real life scenario Nyquist frequency Amplitud e 10k 20k 30k 40k 50k 60k 70k 80k 90k 100k Frequency spectrum We need a “Low-pass filter” to remove unwanted high frequency signals
Anti-aliasing filter Nyquist frequency Amplitud e 10k 20k 30k 40k 50k 60k 70k 80k 90k 100k Frequency spectrum Sensor ADC Anti-aliasing Filter
Anti-aliasing filter Nyquist frequency Amplitud e 10k 20k 30k 40k 50k 60k 70k 80k 90k 100k Frequency spectrum Sensor ADC Anti-aliasing Filter
Anti-aliasing filter Anti-aliasing filter Amplitud e 10k 20k 30k 40k 50k 60k 70k 80k 90k 100k Frequency spectrum Sensor ADC Anti-aliasing Filter
Model for a signal (frequency, amplitude, and phase) A . sin(𝜄) A . 𝑑𝑝𝑡(𝜄)
Model for a signal (frequency, amplitude, and phase) sin(𝜄) 1 � cos(𝜄) How can we incorporate both Sine and Cosine in the equation?
Model for a signal (frequency, amplitude, and phase) sin(𝜄) 1 � cos(𝜄) 𝑑𝑝𝑡 𝜄 + sin(𝜄) 1.
Model for a signal (frequency, amplitude, and phase) sin(𝜄) 1 � cos(𝜄) < 𝑑𝑝𝑡 𝜄 , sin 𝜄 > 𝑑𝑝𝑡 𝜄 + sin(𝜄) 1. 2.
Model for a signal (frequency, amplitude, and phase) sin(𝜄) 1 � cos(𝜄) < 𝑑𝑝𝑡 𝜄 , sin 𝜄 > 𝑑𝑝𝑡 𝜄 + sin(𝜄) 1. 2. 𝑑𝑝𝑡 𝜄 + 𝑘 sin 𝜄 3.
Complex numbers Imaginary 𝑘 = −1
Complex numbers
Complex numbers
Complex numbers Imaginary axis j8 = multiply by "j" 0 Real – 8 8 axis -j8 � � � � � � � — � �
Complex numbers and Natural exponential
� � � � � � � � � � � � � � Complex numbers and Natural exponential � e j � = 1 + j � + (j � ) 2 + (j � ) 3 + (j � ) 4 + (j � ) 5 � + 2! 3! 4! 5! � � � � � � � � � � � �
� � � � � � � � � � � � � � � � � � � � � � � � � � � � Complex numbers and Natural exponential � � e j � = 1 + j � + (j � ) 2 + (j � ) 3 + (j � ) 4 + (j � ) 5 � � � � � � � + � 2! 3! 4! 5! � � � � � = 1 + j � - � 2 2! - j � 3 3! + � 4 4! + j � 5 5! - � 6 � 6! � � � � � � � � � � � �
Recommend
More recommend