EE 250 | Layer 1 - PHY Layer | Unit 5 |

EE 250 Unit 5
Physical (PHY) Layer

© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 2

IoT Architecture

https://www.quicsolv.com/internet-of-things/how-iot-works/

© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 3

Basic Embedded Device Architecture

SENSORS

LOW POWER WIRELESS

ENERGY SOURCE
MICROPROCESSOR RADIO TRANSCEIVER

ACTUATORS

This Photo by Unknown Author is

licensed under CC BY-SA
© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 4

LAYER 1 OVERVIEW

© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 5

Physical Layer Overview

• From bits to voltages and radio waves
▪ Interacting with the physical world to communicate 1s and 0s
• Much more to it…

110100010

© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 6

PHY Concerns and Choices

• PHY Concerns / Goals:
▪ Distance vs. power
o Signals attenuation and EMI (Electro-magnetic interference)
▪ Bandwidth and spectrum sharing
• The PHY layer can choose/control:
▪ Signaling/Modulation
o How we transmit 1s and 0s, timing/clocking, etc.
▪ Coding
o Data bits vs. transmitted bits (what might be gained if we convert data bits to some
alternate bits before modulation)

We will focus on wireless PHY aspects.

© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 7

Time vs. Frequency Domain

SIGNALS PRIMER AND

IMPORTANT MATHEMATICAL
FOUNDATIONS
© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 8

Wireless Signals
• Radio Frequency (RF) signaling (EM
waves) using antenna and modulators
▪ These waves are governed by Maxwell's
equations
o Maxwell's equations essentially unified, related,
and augmented other known laws of physics like
Gauss' law, Faraday's law, etc.
o http://www.maxwells-equations.com/

• We now deal with sinusoids and NOT

digital square waves
• Use of specific frequency bands
▪ WiFi (802.11), Zigbee (802.15.4) use
unlicensed frequency bands in the ISM
(Industrial, Scientific, and Medical) bands
such as 915 MHz, 2.4GHz, 5GHz etc.
https://sites.google.com/site/specquest/part-1-electromagnetic-spectrum
https://commons.wikimedia.org/wiki/File:Electromagneticwave3D.gif
© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed. https://commons.wikimedia.org/wiki/File:Datos_digitales_-_Se%C3%B1ales_analogicas.png
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 9

Sinusoidal Signals
• A sinusoid is described by:
▪ Frequency (or inverse of period, i.e., 1/T,)
denoted by f, is the signal's rate of change
and is measured in Hertz or cycles per
second
▪ Peak amplitude, A, is the maximum strength
of a signal and is measured in volts or amps
▪ Phase, 𝜙, describes the position of the
waveform relative to time 0
• Example: Sine wave representation:
A sin(2𝜋𝑓𝑡 + 𝜙)

https://www.nist.gov/sites/default/files/styles/960_x_960_limit/public/images/2016/09/15/period_1.gif?itok=iLypB1r5
© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 10

Composite Signals (1)

• Suppose we added several sinusoids together at each point in time to get a
composite sinusoid.
• What would the resulting signal look like? Random garbage?

Individual sine waves of

1*f, 3*f, 5*f, 7*f
1

0.8

0.6

0.4

0.2

-0.2

-0.4

-0.6

-0.8

-1
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.1

© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 11

Composite Signals
• A composite signal is the sum of many sinusoids with various
amplitude, frequencies, and phases
▪ Useful signals are generally composite signals

Individual sine waves of A composite signal:

1*f, 3*f, 5*f, 7*f Sum of those sine waves
1 1

0.8 0.8

0.6 0.6

0.4 0.4

0.2 0.2

0 0

-0.2 -0.2

-0.4 -0.4

-0.6 -0.6

-0.8 -0.8

-1 -1
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.1 0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.1

© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 12

Time vs. Frequency Domain

• Applying Fourier analysis we can break a
composite signal into its component
sinusoids and plot their amplitude in the
frequency domain

Component Sine Waves - Freq.

1
Composite signal – Time Domain FourierDomain
Series of Composite Signal
1
0.8
0.9
f = 60Hz
0.6
0.8
1*f Component
0.4
0.7 3*f Component

Fourier Coefficient
0.2
0.6
0 5*f Component
0.5
-0.2
0.4 7*f Component
-0.4
0.3
-0.6 9*f Component
0.2
-0.8
0.1
-1
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.1 0
0 500 1000 1500 2000 2500 3000 3500 4000
Hz
© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
https://i1.wp.com/play.fallows.ca/wp/wp-content/uploads/sites/4/2017/01/Signal-Analysis.jpg
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 13

Fourier Analysis and Transform

• By applying variations of the Fourier transform or analysis we can:
▪ Decompose any time domain signal to its component frequencies
▪ Construct any time domain signal using some set of sinusoids at various frequencies

Plot of fight song Fourier Series of Sound

1 0.035

0.8
0.03
0.6

0.4 0.025

Fourier Coefficient
0.2
Amplitude

0.02
0 Note: A sinusoid in one domain is
0.015 an impulse in the other.
-0.2

-0.4
0.01
-0.6
0.005
-0.8

-1 0
0 0.5 1 1.5 2 2.5 3 3.5 4 0 500 1000 1500 2000 2500 3000 3500 4000
Time Domain seconds or distributed.
© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, Frequency Domain Hz
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 14

Bandwidth
• Bandwidth (BW) is the range of A

frequencies present in a composite

BW = 928-902
signal = 26 MHz
f
▪ BW = freqHi - freqLo 902
MHz
928
MHz

▪ Width of the used spectrum Fourier Series of Sound

0.035

• We will see that the greater the 0.03

bandwidth, the higher the bit rate 0.025

Fourier Coefficient
0.02

we can support. 0.015

0.01

0.005

0
0 500 1000 1500 2000 2500 3000 3500 4000
Hz

BW of the signal above is

about 3190 Hz =
3200 highest - 10 lowest freq.
© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 15

Product of Sinusoids
• Recall, the various trigonometric product-to-sum and sum-to-
product identities
1
▪ cos 𝛼 ∙ cos 𝛽 = ∙ cos 𝛼 − 𝛽 + cos 𝛼 + 𝛽
2
1
▪ sin 𝛼 ∙ sin 𝛽 = ∙ cos 𝛼 − 𝛽 − cos 𝛼 + 𝛽
2

Key: The product of two sinusoids at specific frequencies yields

components at the sum and difference of the input frequencies.

𝑓ℎ𝑖 𝑓ℎ𝑖 0.035

Fourier Series of Sound

1 0.03

0.025

cos 𝛼 ∙ ෍ cos 𝛽𝑖 = ෍ ∙ cos 𝛼 − 𝛽𝑖 + cos 𝛼 + 𝛽𝑖

Fourier Coefficient
0.02

2
0.015

0.01

𝑖=𝑓𝑙𝑜 𝑖=𝑓𝑙𝑜
0.005

0
0 500 1000 1500 2000 2500 3000 3500 4000
Hz

© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 16

Baseband and Carrier Signals A

0.035

0.03

0.025
Fourier Series of Sound

Fourier Coefficient
0.02

•
0.015

0.01

Baseband signals 0.005

0
0 500 1000 1500 2000
Hz
2500 3000 3500 4000

▪ Run from 0 to maximum freq. component 3 kHz

f

o Ex: voice is usually 300 to 3 kHz A

o Thus BW = 2700 Hz, but Baseband signal is 3KHz
▪ Often use this term to refer to the original signals we
f
want to communicate 3 kHz

▪ Are often easily distorted

A Carrier Freq. Carrier Freq.
▪ Multiple signals in the same frequency range can
corrupt each other
• We often modulate the baseband signal on to 915
kHz
921
kHz
f

different carrier signals to separate them and make

them amenable to transmission A

• A broadband transmission enables two or more 912-918 =

6 KHz = BW
918-924 =
6 KHz = BW
communication channels to share the bandwidth of f
a communications channel 912
KHz
918
KHz

© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed. Broadband Transmission
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 17

Filtering
• Filtering is the process of removing undesired
frequency components from a composite signal
• Low-pass filter (LPF)
▪ Allows low frequencies to pass but removes high
frequencies
Ideal Low-Pass Filter (LPF)
▪ Ideal LPF transfer function for some cutoff Response
frequency, fc
1, 𝑓 ≤ 𝑓𝑐
o 𝐻 𝑓 =
𝑆𝑜𝑢𝑡
𝑆𝑖𝑛
=ቊ
0, 𝑓 > 𝑓𝑐
▪ Hard to achieve ideal
• Band-pass filter
▪ Allows frequencies in some range to pass but
attenuates others
• More on this in a future unit…
Real LPF Response

https://i.stack.imgur.com/nLwKi.png
https://www.mathworks.com/help/signal/examples/practical-introduction-to-digital-filter-design.html
© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 18

dB, dBm, and SNR

SIGNAL STRENGTH, SNR, AND DECIBELS

© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 19

Can You Hear Me Now?

• Successful transmission requires the receiver be able to "hear" the
transmitter's signals
• Two common barriers to this are:
▪ Attenuation and fading N

▪ Sources of Noise
S
• Often need to compute the
ratio of:
Pt Pr
▪ Receive Power (Pr) to
Transmit Power (Pt)
▪ Signal Power (S) to
Noise Power (N)

© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 20

Electromagnetic Wave Propagation:

Ideal Inverse-Squared Law
• As an ideal spherical EM wave radiates outwards in vacuum, the received
power decreases inversely with the square of distance: 𝑃𝑟 ∝ 𝑃𝑡Τ𝑑 2
▪ See http://hyperphysics.phy-astr.gsu.edu/hbase/Forces/isq.html

From: https://i.stack.imgur.com/Yec8x.png
© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 21

Signal to Noise Ratio

• When we send signals, noise can occur
▪ EMI (Electro-magnetic interference)
▪ Multipath fading
o Reflections of a signal can interact with the
original signal in an additive or subtractive way
• The 2.4GHz band where 802.11bgn and
802.15.4 operate is fairly occupied with many
sources of noise
o Microwaves, cordless phones, car alarms, baby
monitors, etc.
o Other sources: Internal circuitry itself, thermal
noise, neighboring frequencies

• SNR = Signal to Noise Ratio =

𝑺
𝑵
▪ Ratio of signal power to noise power

© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 22

Interference From Other Networks

From Fuhr, Peter & Manges, Wayne & Schweitzer, Patrick & Kagan, Hesh. (2010). Wireless Technologies
Implications for Power Systems.
https://www.researchgate.net/publication/254994580_Wireless_Technologies_Implications_for_Power_Systems
© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 23

Measuring Ratios: Bels and Decibels

• Usually want to measure the ratio of two signals
▪ Transmitted vs. received signal, Signal to noise, amplification of one range of
frequency vs. another in a composite signal
• A common method of measuring ratios is Bels
𝑃
▪ 1 𝐵𝑒𝑙 = 𝑙𝑜𝑔10 ( )
𝑃0
o If P = 10P0, P has a gain of 1 Bel
o If P = P0, P has a gain of 0 Bels
• To make the granularity finer, we often use decibels (dB):
𝑃
▪ 1 𝑑𝐵 = 10 ∗ 𝑙𝑜𝑔10 ( )
𝑃0
o If P = 100P0, P has a gain of 20dB

© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 24

Measuring Power: dBm

• Radio power is typically measured in a dBm Power Ratio

logarithmic scale, using dBm as units which 10 dBm 10 mW

normalizes the ratio to 1 mW 20 dBm 100 mW
• From mW to dBm: 30 dBm 1W
𝑷𝒐𝒘𝒆𝒓 𝒊𝒏 𝒎𝑾
▪ 𝒅𝑩𝒎 = 𝟏𝟎 ∗ 𝒍𝒐𝒈𝟏𝟎 -3 dBm 0.5 mW
𝟏𝒎𝑾
▪ Thus 1 mW corresponds to 0 dBm -50 dBm 0.01 μW
• From dBm to mW: 3 dBm 2 mW
𝒅𝑩𝒎ൗ
▪ 𝒎𝑾 = 𝟏𝟎 𝟏𝟎
15 dBm 32 mW
• What do the following correspond to? 50 dBm 100 W
▪ 10 dBm, 20 dBm, 30 dBm, -3dBm, -50dBm
-30 dBm 1 μW
▪ 2mW, 32mW, 100 W, 1 micro-Watt

Note: 3dB = 2x Power and

© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
-3dB = 0.5x Power
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 25

mW to dBm Conversion Trick

• Rely on the conversions:
▪ Factor of 10±1 in power is equivalent to addition of ±10 in dB
▪ Factor of 2±1 in power is equivalent to addition of ±3 in dB
• Examples:
▪ 40x power gain: 40=10∙2∙2 corresponds to 10+3+3 = +16 dB gain
▪ -15 dB gain: -3 + -3 + -3 + -3 + -3 corresponds to 2-5=1/32 power gain
▪ 58 dB = 7*10 - 4*3 corresponds to 107∙2-4 = 10E6 / 16 ≈ 625,000x power gain

© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 26

WiFi and Cell Receive Signal Strength

https://eyesaas.com/wi-fi-signal-strength/
© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed. https://www.rvwithtito.com/articles/stay-connected-4g-signal-booster
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 27

Modulation and Coding

PHY TASKS

© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 28

Overview of PHY Tasks

• Two of the main PHY tasks in wireless that we will focus on are
coding and modulation
• Coding refers to altering the data bits to a different set of bits often
called "chips" for the purpose of mitigating fading (attenuation,
interference, multipath), synchronization, error correction,
spectrum spreading, and DC balancing
▪ In 802.15.4 every 4 data bits are converted to 32 chips
• Modulation refers to using one signal to alter another, usually for
the purpose of moving signals of one frequency to another

https://www.springer.com/cda/content/document/cda_downloaddocument/9781461413769-c2.pdf?SGWID=0-0-45-1335410-p174190785
© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 29

Functional Block Diagram of Wireless PHY Layer

• Data bits are converted to multiple transmitted bits (aka chips)

• Chips are then modulated and transmitted

1101000101001 Channel
Modulator Tx Front End
Encoder
Data Bits

Bit to Chips to Amplifier,

Chips Voltages Filtering, etc.

Channel
1101000101001 Demodulator Rx Front End
Decoder
Data Bits
Chips to Voltages to Amplifier,
Bits Chips Filtering, etc.
© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 30

MODULATION

© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 31

Modulation Overview
• Modulation is the process of
altering the characteristics of one
signal based on another
• Usually we take the original
message (or baseband) signal and
use it to alter a carrier
signals/frequency
• Here we show an analog signal
modulating a higher frequency
carrier

https://upload.wikimedia.org/wikipedia/commons/a/a4/Amfm3-en-de.gif
© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed. https://qph.fs.quoracdn.net/main-qimg-a187fa629ad758ceb796511a43c89680
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 32

Amplitude Modulation/Demodulation
• From Waltenegus, Fundamentals of Wireless Sensor Networks

© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 33

Amplitude Modulation
𝛽 𝛼
• Message = 𝑆𝑀 cos(2𝜋𝑓𝑚 𝑡 + 𝜙𝑚 ) and 𝐶𝑎𝑟𝑟𝑖𝑒𝑟 = 𝑆𝑐 cos(2𝜋𝑓𝑐 𝑡 + 𝜙𝑐 )
• To modulate the signal we just multiply the carrier signal by the message signal (for ease of
analysis, assume the phase is 0)
▪ 𝑠𝑚𝑜𝑑 𝑡 = 𝑆𝑀 cos 2𝜋𝑓𝑚 𝑡 𝑆𝑐 cos 2𝜋𝑓𝑐 𝑡 (use product of cos identity)
𝑆𝑀 𝑆𝑐
𝑠𝑚𝑜𝑑 𝑡 = ∙ cos 2𝜋 𝑓𝑐 + 𝑓𝑀 𝑡 + cos 2𝜋 𝑓𝑐 − 𝑓𝑀 𝑡
2
𝛼 𝛽 𝛼 𝛽

Spectrum of baseband signal

with bandwidth B.
Recall: cos 𝛼 ∙ cos 𝛽 = Fundamentals of WSNs, Fig
1 5.12
∙ cos 𝛼 − 𝛽 + cos 𝛼 + 𝛽
2

Spectrum of
modulated
signal.
Fundamentals of
WSNs, Fig 5.14

© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 34

Amplitude DeModulation
• Message = 𝑆𝑀 cos(2𝜋𝑓𝑚 𝑡 + 𝜙𝑚 ) and 𝐶𝑎𝑟𝑟𝑖𝑒𝑟 = 𝑆𝑐 cos(2𝜋𝑓𝑐 𝑡 + 𝜙𝑐 )
• Modulated signal:
𝑆𝑀 𝑆𝑐
𝑠𝑚𝑜𝑑 𝑡 = ∙ cos 2𝜋 𝑓𝑐 + 𝑓𝑀 𝑡 + cos 2𝜋 𝑓𝑐 − 𝑓𝑀 𝑡
2
• To demodulate an AM signal we just multiply the received signal by the carrier frequency again
𝐾𝑆𝑀 𝑆𝑐
▪ 𝑠𝑑𝑒𝑚𝑜𝑑 𝑡 = 𝑆𝑐 cos 2𝜋𝑓𝑐 𝑡 ∙ ∙ cos 2𝜋 𝑓𝑐 + 𝑓𝑀 𝑡 + cos 2𝜋 𝑓𝑐 − 𝑓𝑀 𝑡
2
𝛼 𝛽1 𝛽2
𝛼 + 𝛽1 𝛼 + 𝛽2
𝐾𝑆𝑀 𝑆𝑐 2 𝛼 − 𝛽1
▪ 𝑠𝑑𝑒𝑚𝑜𝑑 𝑡 = ∙ cos 2𝜋 2𝑓𝑐 + 𝑓𝑀 𝑡 + cos 2𝜋 2𝑓𝑐 − 𝑓𝑀 𝑡 + 𝟐 ∙ cos(2𝜋𝑓𝑚 𝑡)
2 𝛼 − 𝛽2
▪ Notice that the demodulated signal as two components at 2𝑓𝑐 and one at 𝑓𝑀
▪ We can use a Low-Pass Filter to remove the high frequency carrier wave and recover and amplify the
message signal

Low Pass
Filter

© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 35

Digital Modulation
• Modulation is the process of varying one or more of the key
parameters (amplitude, frequency, phase) of a carrier signal in
relation to the information signal

© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed. 35
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 36

Digital Modulation
• Modulating digital signals
means we only need to
signal 0s and 1s
▪ Often called keying rather
than modulation
• Can use amplitude (ASK),
frequency (FSK), or phase
(PSK) to modulate our bits

https://commons.wikimedia.org/wiki/File:Datos_digitales_-_Se%C3%B1ales_analogicas.png
© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 37

D/A – Modulation Techniques

© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed. 37
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 38

Rectangular vs. Polar Form

• Recall (learn?) from trigonometry
▪ 𝐴 ∙ cos 𝑥 + 𝐵 ∙ sin 𝑥 = 𝑀 ∙ cos 𝑥 + 𝜃 where
𝐵
𝑀 = 𝐴2 + 𝐵2 and 𝜃 = tan−1
𝐴
• If we add a cosine and sine wave of the
SAME frequency, we get a sinusoid with
altered amplitude and phase BUT THE SAME
frequency
• Or a sinusoid with some arbitrary magnitude In these diagrams, we
are showing Amplitude

and phase can be broken into the sum of a and Phase.

We assume the
cosine and sine waves of that same frequency is a known
constant.
frequency with amplitudes A and B
© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 39

Phase and Amplitude

Im
(sin) (1.73,1)
• Fact: a cos and sin wave at the same 1
frequency are "orthogonal" (i.e. independent
where a change in one does not affect our
ability to recover the other) -1 1 Re
(cos)
▪ Recall 𝑒 𝑖𝑡 = cos 𝑡 + 𝑖 ∙ sin(𝑡) and the complex -1
plane
▪ The dot product of the two is 0 Reminder: we assume the
frequency is a known constant.

• The point is that we can modulate the x (real / 1.73 ∙ cos 2𝜋𝑓𝑡 + 1 ∙ sin 2𝜋𝑓𝑡

cosine) component separately from the y can be combined to

(imaginary / sine) component, mix them to a 𝜋
= 2 ∙ cos 2𝜋𝑓𝑡 +
single sinusoid with a certain magnitude and 6
phase, and then separate them back can be separated back to

= 1.73 ∙ cos 2𝜋𝑓𝑡 + 1 ∙ sin 2𝜋𝑓𝑡

© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 40

BPSK, QPSK, & Constellation Diagrams

-1 +1
• The PSK plot shown on the previous slide is
Re(cos)
more aptly called Binary PSK (BPSK)
• The receiver need only be capable of
distinguishing two phases BPSK

• Quadrature Phase Shift Keying modulates a Data: 0 0 1 1 0 1

Phase:-Π -Π Π Π -Π Π
cosine and sine wave separately to create a Amp: -1 -1 +1 +1 -1 +1

single sinusoid with a given amplitude

[magnitude] and phase

Phase to bit mappings – aka "Constellation" diagrams

© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 41

QPSK Modulation
• The bits from the incoming bitstream at rate, R, are alternately distributed to
the I and Q stream
• The I (in-phase) stream modulates the cos wave using BPSK and the Q
(quadrature-phase) stream does the same with the sin wave
• The two signals are then added and transmitted => QPSK

-1 +1 +1 +1

1 0 0 1 0 1 1 1

+1 -1 -1 +1

http://www.rfwireless-
world.com/Terminology/QPSK-vs-
© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed. QPSK Modulation OQPSK-vs-pi-4QPSK.html
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 42

QPSK Demodulation
• The incoming signal is multiplied by a cos and sin wave
▪ Recall our discussion of AM and how this yields a component at twice the carrier frequency and a
component at the message frequency
• The low pass filter removes the carrier frequency
• A 1 or 0 will have a constant (aka DC) positive or negative offset and can be detected with a
threshold detection

(Integrator)
+1 +1 +1 -1

QPSK Demodulation
1 1 0 1 0 1 1 0
+1 -1 -1 +1

(Integrator)

https://www.researchgate.net/profile/Rj_Green/publication/224611110/figure/fig3/AS:302813562327043@1449207
787770/Schematic-diagram-of-a-quadrature-phase-shift-keying-QPSK-demodulator.png
© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 43

QPSK Matlab Demo

© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 44

QPSK QPSK Constellation

10 11
45⁰
• Allows…
135⁰

▪ Twice as many bits per second using the same bandwidth

as BPSK ..OR..
00 01
▪ Half the bandwidth required by BPSK
-135⁰ -45⁰
• Essentially we are trying to distinguish 4 phases
• By modulating both amplitude and phase of the I & Q
signals (Quadrature Amplitude Modulation) we can
transfer more bits per symbol (but requires better SNR
to avoid errors)
• Used by IEEE 802.15.4
• Good Explanation:
▪ https://www.labvolt.com/downloads/39865_F0.pdf

https://www.labvolt.com/downloads/39865_F0.pdf
https://www.electronicdesign.com/sites/electronicdesign.com/files/uploads/2013/10/1107_
EE_evm_F4.gif
© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 45

2k-PSK and QAM

• Note: 2k-PSK implies k bits per signal
• To the right is 8-PSK (k=3) 2k-PSK
• As k increases in 2k-PSK, phases get closer to
each other and is harder to differentiate during
demodulation, making the receiver more likely
to make mistakes (and thus bit errors) due to
noise
• By modulating the amplitude of both the I & Q
8-QAM

signals (Quadrature Amplitude Modulation =

QAM) we can produce many combinations of
Magnitude/Amplitude AND Phase (yielding
more bits per signal)

© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed. 45
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 46

QAM
• Using Quadrature Amplitude Modulation (QAM) we can transfer
more bits per symbol
▪ N-QAM can send log2(N) bits per sinusoid using different magnitude/phase
combinations
▪ Requires better SNR to avoid errors
▪ Later 802.11 WiFi standards can switch from BPSK to 1024 QAM based on
signal quality

16 QAM 64 QAM
Constellation Constellation

© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed. https://www.headendinfo.com/32qam-64qam-128qam-256qam/
 EE 250 | Layer 1 - PHY Layer | Unit 5 |47

Helpful Resources
• https://www.allaboutcircuits.com/textbook/radio-frequency-analysis-design/

© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 48

Using 802.15.4 Zigbee as a case study

CODING (AND 802.15.4)

© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 49

Bit vs. Symbol vs. Chip Rate and Baud

Chips Symbols

1101000100101 Tx Front
Coder Modulator
End
Data Bits

Bit to Chips to Amplifier,

Chips Voltages Filtering, etc.

• For more robust communications, data bits are coded to longer sequences of
bits (aka chips).
▪ Say, every 4 data bits are translated to 8 chips.
• Now consider a modulation scheme that specified 16 distinct amplitude/phases
(i.e. 16 QAM), or symbols
▪ Thus, 16 unique symbols correspond to 4-bits of information
• Thus if we have a data bit stream at 10 kbps…
▪ It is converted to 20 kChips/s which is then transmitted at 5 kSym/s
• The rate of transmitted symbols is known as the baud rate
© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 50

802.15.4 2.4GHz PHY Coding

Why convert 4-bits to 32-bits?

This is an example of what is
called (Direct-Sequence)
Spread Spectrum approaches.
More bits may make symbols 0000
more discernable and allow a 0001
few bits to be in error & still
recoverable
Faster transmission may spread
the power from a narrow
Data bits

frequency band to a wider band

(has some desirable features)

https://cdn.rohde-
schwarz.com/pws/dl_downloads/dl_appli
cation/application_notes/1gp105/1GP10
5_1E_Generation_of_IEEE_802154_Sig
nals.pdf
1111
© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 51

Orthogonal Vectors (Cross Correlation)

• Orthogonal codes: Correlation between two vectors is 0
• Cross-correlation (between 2 symbols):
𝑪 = ෍ 𝒔𝟏 𝒊 ∙ 𝒔𝟐 𝒊
𝒊
• 802.15.4 codes are quasi-orthogonal

Cross-Correlation: DS0: 11 01 10 01 11 00 00 11 01 01 00 10 00 10 11 10
C(DS0,DS1)=0 DS1: 11 10 11 01 10 01 11 00 00 11 01 01 00 10 00 10

Recall:
• digital 1 = +1 and
• digital 0 = -1

© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 52

Auto-Correlation
• Auto-correlation (shifted version of self):
𝐶 = ෍𝑠 𝑖 ∙ 𝑠 𝑖 − 𝜏 , 1≤𝜏<𝐿
𝑖

110110011100001101010010001011100
011011001110000110101001000101110

Auto-correlation: C(DS0,DS0+1) = -3

1101100111000011010100100010111000
0011011001110000110101001000101110

Auto-correlation: C(DS0,DS0+2) = -6

© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
https://ntnuopen.ntnu.no/ntnu-xmlui/bitstream/handle/11250/2369531/349061_FULLTEXT01.pdf?sequence=1
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 53

Receiving the Spread Spectrum Signal

• At the receiver can correlate (take the inner-product of) the input signal with
each of the 16 possible PN sequences and choose the symbol (i.e. convert back
to the 4-data bits) whose sequence had the highest correlation
▪ A few chips may be in error but the correlation will still be high

DS0: 11 01 10 01 11 00 00 11 01 01 00 10 00 10 11 10

Correlation
x Σ to 0000 seq.
…
PN sequence for 0000
Incoming Max
signal
... 01 00 10 00 10 11 10 Correlation
x Σ to 1111 seq.

PN sequence for 1111

© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 54

Direct Sequence Spread Spectrum

• What would happen to the bandwidth if rather than sending a bit every 4
microseconds, we send a bit every 0.5 microseconds?
• 802.15.4 converts every 4 data bits to 32 chip PN (PseudoNoise) sequence
▪ This increases the bandwidth needed from 250Kbps to 2Mbps
• Why convert 4-bits to 32-bits?
• Spread spectrum signaling
▪ Move from a single, narrow band to spread the signal over greater
bandwidth
o Side note: Can make the signal harder to detect and harder to jam
o The ratio of the higher chip rate to the data rate is know as the spreading factor (i.e. 32
chips / 4 data bits = 8x spreading factor)
▪ Known as Direct Sequence Spread Spectrum (DSSS)

© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 55

Spread Spectrum Communications

Baseband signal

Spreading Spreading

Data bits Chip seq.

(lower BW) x (higher BW)

Spreading PN sequence

Chip seq. Data bits

(higher BW) x Σ (lower BW)

Despreading PN sequence
Despreading

© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
https://www.allaboutcircuits.com/textbook/radio-frequency-analysis-design/selected-topics/understanding-spread-spectrum-rf-communication/
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 56

Benefits of DSSS
• Tolerant of noise and interference in portions of the
used spectrum
• Resilient to multipath effects

© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 57

802.11 Coding & Modulation

• Rather than spread spectrum techniques used by 802.15.4, 802.11ac uses OFDM while 802.11ax
uses a variant known as OFDMA
▪ Breaks one channel to multiple (say, K) narrow-band subcarriers
▪ With K sub-carriers, a data stream of bits at rate R is demultiplexed (alternated) to each of the K
subcarriers so each carrier sends data at a rate of R/K
o Subcarrier 0 transmits bits 0, K, 2K, 3K, … while subcarrier 1 transmits bits 1, K+1, 2K+1, 3K+1, etc.
▪ If certain frequencies experience fading and cause errors, FEC and the coding scheme can help recover
from or correct the errors since only 1 out of K bits are missing/wrong

Bit Stream: 0111… 0111

One OFDM channel Subcarriers

© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed. http://www.ni.com/white-paper/3740/en/
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 58

802.11 Modulation
• Channels can choose an appropriate modulation scheme
based on link conditions and other environmental
variables
▪ Subcarriers can adaptively choose
BPSK, QPSK, 16-, 64-, 256-, or 1024-QAM

© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 59

MIMO
• Multiple-In, Multiple-Out
▪ Several antenna all transmitting at the same time
(even potentially over the same channel)
• Can trade data rate for reliability
• Advantages
▪ Spatial multiplexing: Multiple independent data
streams over the same channel
o Can use multipath to our advantage so that the different
streams are received at different times (different paths)
o Use of orthogonal codes for each stream
▪ Reliability / Beamforming: Use of multiple
transmitters to strengthen one stream at the receiver

© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed. This Photo by Unknown Author is licensed under CC BY-SA
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 60

LORA PHY

© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 61

LoRa PHY
• Operating at 13 frequency channels around 915 MHz in USA, 868
MHz in Europe
• Low data rate but very long range (few miles outdoors)
• PHY uses Chirp-spread spectrum (CSS) signaling
▪ Can tune Tx power, bandwidth, and spreading factor to achieve data-rate vs.
power requirements
o Higher Tx Power (Better Range, Less battery life)
o Higher Spreading Factor (Better Noise Immunity/Range,
Lower data rate)

© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 62

Chirp Spread Spectrum

• LORA uses chirp spread spectrum techniques rather than QPSK with
DSSS
• A chirp is a modulation technique where the signal transitions from
flow to fhigh (or fhigh to flow ) linearly over time

S(f)

f
B

Spectrum of the chirp pulse with

bandwidth B and a roll-off factor of 0.25 Up-Chirp in the time domain Frequency vs. time

http://www.ieee802.org/802_tutorials/03-November/15-03-0460-00-0040-IEEE-802-CSS-Tutorial-part1.ppt
© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 63

LORA Modulation
• Jumps in frequency correspond to data values being modulated
• Data is decoded by looking at the frequency transitions and use
the constant slope to infer the initial frequency, f0, of each chirp.

Time

https://www.mdpi.com/1424-8220/16/9/1466/htm
© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 64

Chirps and Spreading Factor

• Can make each symbol time longer or shorter (i.e. Spreading Factor) to trade
reliability and range for data rate
▪ Higher SF (i.e. longer duration per symbol) => Lower data rates but better reliability and
range

© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
https://www.sghoslya.com/p/lora-is-chirp-spread-spectrum.html
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 65

How distance affects the signal strength

PROPAGATION AND PATH LOSS

© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 66

RF wave propagation
• As an ideal spherical EM wave radiates outwards
in vacuum, the received power decreases with
distance per an inverse-squared law, i.e.
inversely with the square of distance: 𝑃𝑟 ∝
1Τ𝑑 2
▪ See http://hyperphysics.phy-
astr.gsu.edu/hbase/Forces/isq.html
• In practice, due to reflection, scattering and
absorption effects the propagation is not so
simple. A more general model is used: 𝑃𝑟 ∝
1Τ𝑑 𝜂 , From: https://i.stack.imgur.com/Yec8x.png

▪ η is called the path loss exponent and is

determined empirically.
▪ Typically, η is between 1 to 4.

© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 67

Introduction to Path Loss Model

• Receivers experience signal attenuation due to distance and the medium being
traversed

https://www.datarespons.com/drones-wireless-video/
http://www.tranzeo.com/allowed/Tranzeo_Link_Budget_Whitepaper.pdf
© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 68

Introduction to the Path Loss Model

• Signal strength is also lost due to multi-path effects (e.g. reflections,
diffraction, scattering)
▪ Can be additive or subtractive
▪ If phase shift Φ = π?
▪ If phase shift Φ = 2π?
Original received signal

Received reflection
Φ
Delay to multipath

http://www.wirelesscommunication.nl/reference/chaptr03/pathloss.htm
© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 69

Auto-Correlation
• 802.15.4 codes have beneficial auto-correlation properties
• Auto-correlation (shifted version of self):
𝐶 = ෍𝑠 𝑖 ∙ 𝑠 𝑖 − 𝜏 , 1≤𝜏<𝐿
𝑖

110110011100001101010010001011100
011011001110000110101001000101110

Auto-correlation: C(DS0,DS0+1) = -3

1101100111000011010100100010111011
1011011001110000110101001000101110

Auto-correlation: C(DS0,DS0+2) = -6
© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
https://ntnuopen.ntnu.no/ntnu-xmlui/bitstream/handle/11250/2369531/349061_FULLTEXT01.pdf?sequence=1
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 70

Simplified Path Loss Equation

• To model the signal strength loss, the following "simplified path loss equation"
can be used (either in mW or dBm)

𝑷𝒕𝒎𝒘 𝒅
𝑷𝒓𝒎𝒘 𝒅 = 𝑷𝒓𝒅𝑩𝒎 𝒅 = 𝑷𝒕𝒅𝑩𝒎 − 𝑲𝒅𝑩 − 𝜼 ∙ 𝟏𝟎 ∙ 𝒍𝒐𝒈𝟏𝟎
𝒅 𝜼 𝒅𝟎
𝑲𝒓𝒆𝒇 ∙
𝒅𝟎
In dBm units
In milli-watt units
d0

Parameters
• d is distance
• d0 is a reference distance (say 1m or 100m)
Pt K=(Pt/Pr(d0)) Pr
• Pt is the transmitted power
• Pr is the received power
• K is the unitless path loss at the reference distance d
• η is the path loss exponent

© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 71

Simplified Path Loss Equation

PrdBm
𝑷𝒕𝒎𝒘 𝑷𝒓 𝒅𝟎 = 𝑷𝒕𝒅𝑩𝒎 − 𝑲𝒅𝑩𝒎
𝑷𝒓𝒎𝒘 𝒅 =
𝒅
𝑲𝒓𝒆𝒇 ∙ ( )𝜼
𝒅𝟎

Slope = 𝜼

𝒅
𝑷𝒓𝒅𝑩𝒎 𝒅 = 𝑷𝒕𝒅𝑩𝒎 − 𝑲𝒅𝑩 − 𝜼 ∙ 𝟏𝟎 ∙ 𝒍𝒐𝒈𝟏𝟎
𝒅𝟎
log10(d/d0)

𝜼 = 2 through free space and 2-4 for real environments

(buildings, walls, etc.) and determined empirically
d0
Parameters
• d is distance
• d0 is a reference distance (say 1m or 100m)
• Pt is the transmitted power
• Pr is the received power
• K is the unitless path loss at the reference distance Pt K=(Pt/Pr(d0)) Pr
• η is the path loss exponent
© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed. d
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 72

Measured Path Loss Examples

Location Path Loss Exponent

Free spaces 2
Urban area cellular radio 2.7-3.5
In-building LOS 1.6-1.8
Obstructed in-building 4-6
Obstructed in factories 2-3

https://www.csie.ntu.edu.tw/~hsinmu/courses/_media/wn_17fall/pathloss.pdf
© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 73

Simplified Path Loss Equation

• Random fluctuations due to
shadowing and multi-path PrdBm

reflections - referred to as fading

• Can be modelled as a Gaussian
distribution in dBm scale
▪ Log-normal fading
log10(d/d0)

Original received signal

Received reflection
Φ
Delay to multipath
© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 74

Log-Normal Fading
Can model the vertical
“noisy/random” distribution of
received power values (in dBm)
around the “mean” path loss line
(with respect to log (distance))
as a Gaussian Random
Variable, characterized by a
standard deviation 𝝈. 𝝈2 is
referred to as the fading
variance.

Modeling is an art - there isn’t one right way to

do it, though some approaches are better (more
realistic or more useful or more elegant) than
others.

© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 75
Real Received Power
(Received Signal Strength - RSS) Measurements

From P. Tarrio, A. M. Bernardos, J. R. Casar, "Weighted Least Squares Techniques for Improved Received Signal
Strength Based Localization," Sensors 11(9), 2011. http://www.mdpi.com/1424-8220/11/9/8569/htm

© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 76

Communication Range
• The minimum received power at which the radio can demodulate a message is
called radio sensitivity threshold affects the communication range
▪ If the threshold is smaller, the range will be larger.
• If we transmit at 0 dBm, KdB = 20dBm, d0 = 1m, and η = 3, and the radio
sensitivity threshold is -80dBm, what is the communication range of the radio?
▪ Does the range get bigger/smaller if the transmit power increases? Bigger
▪ Does the range get bigger/smaller if the path loss exponent η increases? Smaller

𝒅
𝑷𝒓𝒅𝑩𝒎 𝒅 = 𝑷𝒕𝒅𝑩𝒎 − 𝑲𝒅𝑩𝒎 − 𝜼 ∙ 𝟏𝟎 ∙ 𝒍𝒐𝒈𝟏𝟎
𝒅𝟎

−𝟖𝟎 = 𝟎 − 𝟐𝟎 − 𝟑𝟎 ∙ 𝒍𝒐𝒈𝟏𝟎 𝒅
−𝟔𝟎 = −𝟑𝟎 ∙ 𝒍𝒐𝒈𝟏𝟎 𝒅
𝟐 = 𝒍𝒐𝒈𝟏𝟎 𝒅
𝟏𝟎𝟐 = 𝒅 = 𝟏𝟎𝟎𝒎

© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 77

TutorNet Testbed

© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 78

Bluetooth and BLE

• 2.4 GHz band
• 79 channels, 1 MHz each
• 1-3 Mbps, ~1 W power consumption
• Frequency Hopping
• Maximum of 8 devices are allowed to share the bandwidth within the same
network by using non-conflicting sequences for hopping
• Range from 0.5 m to 100m
• A newer version standard is Bluetooth Low Energy (BLE) which trades off data
rate (125 kbps-2 Mbps) for lower power consumption (.01 to 0.5 W)

To learn more, see https://en.wikipedia.org/wiki/Bluetooth

© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 79

Recap / Summary Questions

• Understand the relationship of bandwidth and SNR
• Convert between dB and power ratios
• Understand AM and various digital modulation techniques (ASK,
FSK, PSK, QPSK and n-QAM)
• Given parameters regarding modulation (e.g. QPSK or n-QAM),
coding (chips/data bit), and baud rates, calculate effective data rates
• Utilize the path-loss equation to solve for some unknown value in
the equation using given parameters

© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 80

MORE INFO: 802.15.4

© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 81

IEEE 802.15.4 Low Power Radio Physical Layer

• Lower data rate and power than Wifi

• Two different frequency bands (250kb/s @ 2.4GHz and 20 or 40kbps @ 915 MHz)
• Transmit ~ 0dBm, receive sensitivity: -85dBm, -92dBm
• At 2.4 Ghz
▪ Up to 16 channels at 2.4GHz (PHY layer can help select an appropriate channel)
▪ 250kb/s => 4-bits/symbol (i.e. 16 possible symbols to transmit) => 62.5k Baud
▪ Each of the 16 symbols is a 32-bit sequence that is orthogonal to each other

From https://people.eecs.berkeley.edu/~prabal/teaching/cs294-11-f05/slides/day21.pdf
© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 82

802.15.4 Power and Energy Lifetime

• Following specs were taken from
▪ https://mentor.ieee.org/802.15/dcn/04/15-04-0227-04-004a-ieee-802-15-4-
phy-layer-and-implementation.ppt
• Power budget
▪ Transmit power ~ 50mW
▪ Receive power ~40mW
▪ Idle power ~4 μW
• Take away: If a device checks in at a rate of every few seconds (say
10s) then the energy needed is expected to be less than the lifetime
of AA batteries

© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 83

IEEE 802.15.4 Low Power Radio Physical Layer

• Standard
▪ http://ecee.colorado.edu/~liue/teaching/comm_standards/2015S_zigbee/8
02.15.4-2011.pdf
• Sample parts/datasheets
▪ http://www.ti.com/lit/ds/symlink/cc2538.pdf
▪ https://www.mouser.com/ds/2/268/70005023C-1218178.pdf

© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.
 EE 250 | Layer 1 - PHY Layer | Unit 5 | 84

Frames/Packets/Segments
• Application sends 1 KB = 1024 byte message
▪ TCP adds 24 byte header
▪ IP adds 24 byte header
▪ 802.11 adds 16 byte header
• Suppose the following, how many 802.11 frames are sent?
▪ TCP has max transfer unit of 2048 bytes
▪ IP has max transfer unit of 1536 bytes
▪ 802.11 has max transfer unit of 272 bytes
• Suppose the following, how many 802.11 frames are sent?
▪ TCP has max transfer unit of 576 bytes
▪ IP has max transfer unit of 320 bytes
▪ 802.11 has max transfer unit of 272 bytes

© 2024 by Mark Redekopp. This content is protected and may not be shared, uploaded, or distributed.