1 CS 6910 – Pervasive Computing Spring 2007 Section 3 (Ch.3): Mobile Radio Propagation Prof. Leszek Lilien Department of Computer Science Western Michigan University Slides based on publisher’s slides for 1 st and 2 nd edition of: Introduction to Wireless and Mobile Systems by Agrawal & Zeng © 2003, 2006, Dharma P. Agrawal and Qing-An Zeng. All rights reserved. Some original slides were modified by L. Lilien, who strived to make such modifications clearly visible. Some slides were added by L. Lilien, and are © by Leszek T. Lilien. Requests to use L. Lilien’s slides for non-profit purposes will be gladly granted upon a written request.

Copyright © 2003, Dharma P. Agrawal and Qing-An Zeng. All rights reserved 2 Chapter 3 Mobile Radio Propagation

Copyright © 2003, Dharma P. Agrawal and Qing-An Zeng. All rights reserved 3 Outline of Chapter 3 We skip a lot of detailed information from this Chapter Introduction Types of Waves Speed, Wavelength, Frequency Radio Frequency Bands Propagation Mechanisms Radio Propagation Effects Free-Space Propagation Land Propagation Path Loss Fading: Slow Fading / Fast Fading Delay Spread Doppler Shift Co-Channel Interference The Near-Far Problem Digital Wireless Communication System Analog and Digital Signals Modulation Techniques

Copyright © 2003, Dharma P. Agrawal and Qing-An Zeng. All rights reserved Introduction This section covers the distinguishing features of mobile radio propagation Model of wireless mobile channel Time-varying communication path between 2 terminals Fixed BS and mobile MS Mobility introduces new challenges into radio propagation © 2007 by Leszek T. Lilien

Copyright © 2003, Dharma P. Agrawal and Qing-An Zeng. All rights reserved 5 Speed, Wavelength, Frequency SystemFrequencyWavelength AC current60 Hz5,000 km FM radio100 MHz (88-99, ) 3 m Cellular (original) 800 MHz37.5 cm Ka band satellite20 GHz15 mm Ultraviolet light10 15 Hz10 -7 m Light speed = Wavelength x Frequency = 3 x 10 8 m/s = 300,000 km/s

Copyright © 2003, Dharma P. Agrawal and Qing-An Zeng. All rights reserved 6 Speed, Wavelength, Frequency – cont. © 2007 by Leszek T. Lilien ** OPTIONAL ** Broadcast Frequencies (cf. : AM radio bands: long waves (now 153–279 kHz, historically up to 413 kHz) medium waves (520–1,710 kHz in the Americas) short waves (2,300–26,100 kHz) TV Band I (Channels 2 - 6) = 54MHz - 88MHz (VHF) FM Radio Band II = 88MHz - 108MHz (VHF) TV Band III (Channels ) = 174MHz - 216MHz (VHF) TV Bands IV & V (Channels ) = 470MHz - 806MHz (UHF) Most common cellular bands now in uses worldwide: 850/900/1800/1900 MHz E.g., China uses yet another

Copyright © 2003, Dharma P. Agrawal and Qing-An Zeng. All rights reserved Types of Radio Waves Transmitter Receiver Earth Sky wave Space wave Ground wave Troposphere ( km) Stratosphere ( km) Mesosphere ( km) Ionosphere ( km) Ground, space, and sky waves Cellular systems use ground & space waves

Copyright © 2003, Dharma P. Agrawal and Qing-An Zeng. All rights reserved 8 Radio Frequency Bands Classification Band InitialsFrequency RangeCharacteristics Extremely lowELF< 300 Hz Ground wave Infra lowILF300 Hz - 3 kHz Very lowVLF3 kHz - 30 kHz LowLF30 kHz kHz MediumMF300 kHz - 3 MHzGround/Sky wave HighHF3 MHz - 30 MHzSky wave Very highVHF30 MHz MHz Space wave Ultra highUHF300 MHz - 3 GHz Super highSHF3 GHz - 30 GHz Extremely highEHF30 GHz GHz Tremendously high THF300 GHz GHz AM long wave AM med. wave AM short wave FM, TV TV, cellphones WLAN

Copyright © 2003, Dharma P. Agrawal and Qing-An Zeng. All rights reserved Propagation Mechanisms Ideal propagation in free space (no obstacles) Radio signals can penetrate simple walls to some extent Large structure, a hill – difficult to pass through Propagation effects due to obstacles Reflection Propagation wave impinges on an object which is large as compared to wavelength E.g., the surface of the Earth, buildings, walls, etc. Diffraction Radio path between transmitter and receiver obstructed by surface with sharp irregular edges Waves bend around the obstacle, even when LOS (line of sight) does not exist Scattering Objects smaller than the wavelength of the propagation wave E.g. foliage, street signs, lamp posts

Copyright © 2003, Dharma P. Agrawal and Qing-An Zeng. All rights reserved 10 Radio Propagation Effects h b – heights of BS antenna, h m - heights of MS antenna

Copyright © 2003, Dharma P. Agrawal and Qing-An Zeng. All rights reserved 11 Quality of signal reaching receiver via different types of radio waves Direct propagation is the best Reflection is 2nd best Diffraction is 3rd best Scattering is 4th best If no direct-path waves (LOS waves) can reach receiver mainly by reflection or diffraction

Copyright © 2003, Dharma P. Agrawal and Qing-An Zeng. All rights reserved Free-space Propagation The received signal power P r at distance d: A e is effective area covered by the transmitter (on the receiver’s side - e.g., receiver’s antenna) G t is the transmitting antenna gain (a better antenna has a higher gain) P t is transmitting power (Assuming that the radiated power is uniformly distributed over the surface of the sphere.) Transmitter Distance d Receiver hbhb hmhm

Copyright © 2003, Dharma P. Agrawal and Qing-An Zeng. All rights reserved 13 ** SKIP ** Antenna Gain For a circular reflector antenna Gain G =  (  D / ) 2  = net efficiency (depends on the electric field distribution over the antenna aperture, losses, ohmic heating, typically 0.55) D = diameter thus, G =  (  D f /c ) 2, c = f (c is speed of light) Example: Antenna with diameter = 2 m, frequency = 6 GHz, wavelength = 0.05 m G = 39.4 dB Frequency = 14 GHz, same diameter, wavelength = m G = 46.9 dB * The higher the frequency, the higher the gain for the same-size antenna

Copyright © 2003, Dharma P. Agrawal and Qing-An Zeng. All rights reserved 14 Free-space Path Loss Definition of path loss L P : Path loss in free space (no obstacles) : where f c is the carrier frequency The greater the f c, the higher is the loss (see next slide) P t is transmitted signal power P r is received signal power

Copyright © 2003, Dharma P. Agrawal and Qing-An Zeng. All rights reserved 15 Example of Path Loss (Free-space)

Copyright © 2003, Dharma P. Agrawal and Qing-An Zeng. All rights reserved Land Propagation The received signal power: where: G t is the transmitter antenna gain, G r is the receiver antenna gain, L is the propagation loss in the channel, i.e., L = L P L S L F Fast fading (long-term f.) Slow fading (short-term f.) Path loss

Copyright © 2003, Dharma P. Agrawal and Qing-An Zeng. All rights reserved 17 Fast and Slow Fading and Path Loss Fast fading (short-term fading) – Microscopic aspect of a channel for mobile comm. For a moving MS, it represents fading over its every step Due to scattering of signal by objects near transmitter => Changing wave diffractions Slow fading (long-term fading) – Variation of propagation loss in a local area (k * 10m) Due to obstacles (e.g., buildings) For an MS, it represents overall average fading over short distance (e.g., a couple of blocks) traveled by MS Path loss Propagation loss over long distances (more below) © 2007 by Leszek T. Lilien

Copyright © 2003, Dharma P. Agrawal and Qing-An Zeng. All rights reserved 18 Schematic Diagram of Propagation Loss (and Fading) Signal Strength (dB) Distance Path Loss Slow Fading (Long-term fading) Fast Fading (Short-term fading)

Copyright © 2003, Dharma P. Agrawal and Qing-An Zeng. All rights reserved 19 © 2007 by Leszek T. Lilien

Copyright © 2003, Dharma P. Agrawal and Qing-An Zeng. All rights reserved Path Loss (for Land Propagation) Path loss as a fcn of distance L p = A d α where: A and α: propagation constants d : distance between transmitter and receiver Value of  : normally - below 2 in some waveguides - 2 in free space, in typical urban areas - 4 is for relatively lossy environments in some environments, such as buildings, stadiums & other indoor environments

Copyright © 2003, Dharma P. Agrawal and Qing-An Zeng. All rights reserved 21 Path Loss Path loss in decreasing order: Urban area (large city) Urban area (medium and small city) Suburban area Open area Path loss

Copyright © 2003, Dharma P. Agrawal and Qing-An Zeng. All rights reserved 22 Path Loss (Urban, Suburban and Open areas) Urban area: where Suburban area: Open area: Notice how L PU is subtracted from in L PS and L PO - h b, h m – see slide 20 (Units: MHz, m, etc., in parentheses)

Copyright © 2003, Dharma P. Agrawal and Qing-An Zeng. All rights reserved 23 Example of Path Loss ( Urban Area: Large City ) Should be on top of this f c list

Copyright © 2003, Dharma P. Agrawal and Qing-An Zeng. All rights reserved 24 Example of Path Loss ( Urban Area: Medium and Small Cities )

Copyright © 2003, Dharma P. Agrawal and Qing-An Zeng. All rights reserved 25 Example of Path Loss (Suburban Area) Compare loss for, e.g., 30 km & 1500 MHz with Large/Small Urban

Copyright © 2003, Dharma P. Agrawal and Qing-An Zeng. All rights reserved 26 Example of Path Loss (Open Area) [repeated for completeness] Compare loss for, e.g., 30 km & 1500 MHz with Large/Small Urban & Suburb.

Copyright © 2003, Dharma P. Agrawal and Qing-An Zeng. All rights reserved Slow Fading Slow fading (shadowing or log-normal fading) of a signal = long- term spatial and temporal variations in signal strength over “large enough “ distances Distance is “large enough “ if it produces gross variation in signal strength for the overall path between the transmitter and receiver Slow fading obeys log-normal distribution (hence one of its names) - The pdf of the received signal level is given in decibels (dB) by where M is the true received signal level m in dB (i.e., 10log 10 m) M is the area average signal level (i.e., the mean of M)  is the standard deviation in dB

Copyright © 2003, Dharma P. Agrawal and Qing-An Zeng. All rights reserved 28 ** SKIP ** Log-normal Distribution M M 22 p(M) The pdf of the received signal level (log-normal distribution)

Copyright © 2003, Dharma P. Agrawal and Qing-An Zeng. All rights reserved Fast Fading Fast fading (short-term fading) of a signal = rapid fluctuations in the spatial and temporal signal characteristics caused by local mulipath Multipath – since scattered, diffracted, and reflected signals gets to receiver’s antenna over many paths Fast fading occurs over distances of about half a wavelength Example: VHF and UHF (30 MHz MHz & 300 MHz - 3 GHz) Vehicle traveling at 30 mph It passes through several fades in a second Mobile radio signal consists of: Fast fading signal Superimposed on a local mean value Local mean value: constant over a small area / varies slowly as the receiver moves © 2007 by Leszek T. Lilien

Copyright © 2003, Dharma P. Agrawal and Qing-An Zeng. All rights reserved 30 Modeling fast fading: When receiver is far from transmitter No direct radio waves between transmitter & receiver Reflected waves stronger Modeled by the Rayleigh distribution (below) Useful for modeling when no LOS When receiver is close to transmitter Direct radio wave between transmitter & receiver is stronger than other waves Stronger than reflected, diffracted, or scattered Modeled by the Rician distribution (below) © 2007 by Leszek T. Lilien Fast Fading - cont. 1

Copyright © 2003, Dharma P. Agrawal and Qing-An Zeng. All rights reserved 31 Rayleigh Distribution The pdf of the Rayleigh Distribution r – envelope of the fading signal;  - standard deviation r P(r)  =1  =2  =3 r m =  r m is middle value of envelope signal (i.e. such value that: P(r ≤ r m ) = 0.5)

Copyright © 2003, Dharma P. Agrawal and Qing-An Zeng. All rights reserved 32 Rician Distribution r Pdf p(r)  = 2  = 1  = 0  = 1  = 3 The pdf of the Rician Distribution r – envelope of the fading signal  - amplitude of the direct signal (  = 0 means no direct signal)  = 0 means no direct signal => Rayleigh distr. - see the same shape as for Rayleigh distr. with  = 1 For large  (= very strong signal), Rician distr. can be approximated by Gaussian distribution

Copyright © 2003, Dharma P. Agrawal and Qing-An Zeng. All rights reserved 33 Modeling fast fading – cont.: When receiver is far from transmitter Modeled by the Rayleigh distribution When receiver is close to transmitter Modeled by the Rician distribution Generalized model for any distance – Nakagami distrib. Rayleigh distribution is its special case Rician distribution can be closely approximated by it “Can often be better” than Rician © 2007 by Leszek T. Lilien Fast Fading - cont. 2

Copyright © 2003, Dharma P. Agrawal and Qing-An Zeng. All rights reserved 34 Characteristics of Instantaneous Amplitude Instantaneous amplitude => potentially fast changing signal Characteristics of instantaneous amplitude Level crossing rate at a specified threshold (signal level) Average number of times per second that the signal envelope crosses the threshold in positive-going direction Fading rate Number of times that the signal envelope crosses the middle value r m in positive-going direction per unit time Depth of fading Ratio of mean square value of fading signal and minimum value of fading signal Average depth of fading often used instead Fading duration Duration for which signal is below a given threshold

Copyright © 2003, Dharma P. Agrawal and Qing-An Zeng. All rights reserved Doppler Effect (Doppler Shift) Doppler effect - occurs when a wave source and a receiver are moving towards or away from each other When they are moving toward each other, the frequency of the received signal is higher than the source frequency When they are moving away from each other, the frequency is lower than the source frequency Received signal frequency is where: f C - source frequency, f D - the Doppler shift in frequency Doppler shift is where: v - the moving speed - the wavelength of the carrier of the source signal MS Signal from source Moving direction of the receiver (MS) 

Copyright © 2003, Dharma P. Agrawal and Qing-An Zeng. All rights reserved 36 *** SKIP *** Moving Speed Effect [LTL:] This textbook figure is not consistent with text referring to it. It is confusing in the context of the Doppler effect (which has to do with signal frequency not signal strength, as the figure suggests).

Copyright © 2003, Dharma P. Agrawal and Qing-An Zeng. All rights reserved Delay Spread When a signal propagates from a transmitter to a receiver, it can suffer one or more reflections [ Agrawal & Zeng] Now, instead of single-path direct signal we have a multipath reflected signal reaching the receiver More descriptively: Each reflection of the original signal produces a new (fainter) “child” signal that follows its own path The path followed by the “child” signal is determined by reflection parameters When all “children” signals are put together, we have multiple reflected signals (all originating from the original signal) following a multipath towards the receiver Some of the “children” signals may be too faint to reach the receiver Some of the “children” signals may be reflected in wrong direction(s) and never reach the receiver © 2007 by Leszek T. Lilien

Copyright © 2003, Dharma P. Agrawal and Qing-An Zeng. All rights reserved 38 Delay Spread – cont. Different paths in the multipath have different lengths, so the time of arrival for each of many reflected “child” signals is different This spreads out the received signal => is called “delay spread” If we had direct signal only, there would be one signal path, no children signals, and no delay spread See Figure (next)

Copyright © 2003, Dharma P. Agrawal and Qing-An Zeng. All rights reserved 39 Delay Spread Delay Signal Strength The signals reflected from nearby reflectors The signals reflected from intermediate reflectors The signals reflected from distant reflectors Q1: A signal reflected by a nearby reflector arrives sooner than a signal reflected by a more distant reflector. Why?

Copyright © 2003, Dharma P. Agrawal and Qing-An Zeng. All rights reserved 40 Delay Spread – cont. 1 © 2007 by Leszek T. Lilien Q1: A signal reflected by a nearby reflector arrives sooner than a signal reflected by a more distant reflector. Why? A: First note that the direct path is the shortest path from the xmitter to the receiver. A signal reflected by a nearby reflector already traveled most of the distance d as a direct signal. In other words, such signal already traveled most of the distance d along the shortest path to the receiver. A signal reflected further away from the receiver was reflected earlier, so it traveled more of the distance d as a multipath reflected signal, i.e., its components (children signals) traveled more of the distance d along a longer (not the shortest) path to the receiver.

Copyright © 2003, Dharma P. Agrawal and Qing-An Zeng. All rights reserved 41 Delay Spread (again) Delay Signal Strength The signals reflected from nearby reflectors The signals reflected from intermediate reflectors The signals reflected from distant reflectors Q2: A signal reflected by a nearby reflector is stronger than a signal reflected by a more distant reflector. Why?

Copyright © 2003, Dharma P. Agrawal and Qing-An Zeng. All rights reserved 42 Delay Spread – cont. 2 © 2007 by Leszek T. Lilien Q2: A signal reflected by a nearby reflector is stronger than a signal reflected by a more distant reflector. Why? A: A signal reflected by a nearby reflector already traveled most of the distance d as the strongest (single-path & direct) signal. A signal reflected further away from the receiver was reflected earlier, and traveled more of the distance d as a weaker (multipath & reflected) signal. To be precise, it traveled as a multipath collection of reflected children signals, collectively weaker than the original single-path & direct signal would be.

Copyright © 2003, Dharma P. Agrawal and Qing-An Zeng. All rights reserved 43 Delay Spread – cont. 3 © 2007 by Leszek T. Lilien Typical delay spread: 3 microsec. in a city area 10 microsec. in a hilly terrain

Copyright © 2003, Dharma P. Agrawal and Qing-An Zeng. All rights reserved 44 © 2007 by Leszek T. Lilien

Copyright © 2003, Dharma P. Agrawal and Qing-An Zeng. All rights reserved Intersymbol Interference (ISI) Intersymbol Interference (ISI) - caused by delay spread Example of ISI: 101… digital signal sent from xmitter to rcver 4 multipaths due to reflections Single-path direct signal  4 multipath reflected children signals Each child generated by another reflection of the original signal See figures in the next 2 slides Observation – case of ISI: In Slide +2 multipath components for symbol “1” arrive at the time when only multipath components for symbol “0” should be arriving => symbol “1” interferes with symbol “0” © 2007 by Leszek T. Lilien [Agrawal&Zeng]

Copyright © 2003, Dharma P. Agrawal and Qing-An Zeng. All rights reserved 46 Intersymbol Interference (ISI) Example Modified by LTL Figure: Xmitter and receiver are synchronized Red broken-line arrows show synchronization “Short delay”: all 4 delayed multipath signals showing NULL  1 arrive within duration of their “time slot “ Time Transmission signal Received signal (Short delay) Propagation time Delayed multipath signals NULL

Copyright © 2003, Dharma P. Agrawal and Qing-An Zeng. All rights reserved 47 Intersymbol Interference (ISI) Example Modified by LTL Figure Xmitter and receiver are synchronized Red broken-line arrows show synchronization “Long delay”: only 1 delayed multipath signal showing NULL  1 arrives within duration of its “time slot” 3 delayed multipath signals arrive too late (within the next “time slot”) Time Transmission signal Received signal (Long delay) Delayed multipath signals Propagation time NULL

Copyright © 2003, Dharma P. Agrawal and Qing-An Zeng. All rights reserved 48 Burst error– a contiguous sequence of symbols, received over a data transmission channel, such that: (a) the first symbol and the last symbol are in error, and (b) there exists no contiguous subsequence of m correctly received symbols within the error burst The last symbol in a burst and the first symbol in the following burst are separated by m correct bits or more The length of a burst of errors in a frame is defined as the number of bits from the first error to the last, inclusive Error burst – occurrence of burst errors Burst error rate – measures the rate of burst errors ISI can cause burst errors => ISI has impact on the burst error rate of a channel Intersymbol Interference (ISI) – cont. 1 © 2007 by Leszek T. Lilien

Copyright © 2003, Dharma P. Agrawal and Qing-An Zeng. All rights reserved Intersymbol Interference (ISI) – cont. 2 To assure low bit-error rate (BER), digital transmission rate R must be limited by delay spread  d as follows:

Copyright © 2003, Dharma P. Agrawal and Qing-An Zeng. All rights reserved Coherence Bandwidth Gain [cf. = the mean ratio of the signal output of a system to the signal input of the system Gain has wide use to characterize amplifiers Linear phase channel (or filter) [cf. = channel with no distortion due to the frequency-selective delays All frequency components have equal delay times In contrast, in a filter/channel with non-linear phase, delay varies with frequency, resulting in phase distortion Flat channel Passes all frequencies (“spectral components”) with approxi- mately equal gain and linear phase © 2007 by Leszek T. Lilien

Copyright © 2003, Dharma P. Agrawal and Qing-An Zeng. All rights reserved Coherence Bandwidth – cont.1 Coherence bandwidth B c = channel frequency range over which the channel can be considered “flat” In other words: the frequency interval over which two different frequencies f 1 and f 2 of a signal are likely to experience correlated (amplitude) fading [cf. *** SKIP this bullet *** - NOT EXPLAINED WELL – SEEMS INCONSISTENT WITH THE REST B c represents the correlation between 2 fading signal envelopes at frequencies f 1 and f 2 Correlation => a statistical measure B c is a function of delay spread Bigger delay spread => lower correlation Two frequencies that are larger than their coherence bandwidth B c fade independently Concept of coherence bandwidth useful in diversity reception Diversity reception = multiple copies of same message are sent using different frequencies

Copyright © 2003, Dharma P. Agrawal and Qing-An Zeng. All rights reserved Coherence Bandwidth – cont.2 LetS = transmitted signal B = bandwidth of S B c = channels’ coherence bandwidth If B linear transmission of S Only gain and phase of S are changed => S not distorted Intuitively, whole B for S is flat (whole B within B c ) If B > B c => non-linear transmission of S Part of S truncated => S severely distorted Intuitively, only part of B for S is flat (part of B outside of B c ) © 2007 by Leszek T. Lilien

Copyright © 2003, Dharma P. Agrawal and Qing-An Zeng. All rights reserved Cochannel Interference Cells reuse frequencies Same freq assigned to different cells => cells using the same frequency interfere with each other Let:r d - the desired signal r u - the interfering undesired signal  - the protection ratio for which r d   r u (so that interference of r u is limited by  ) (Probability of) cochannel interference between cells reusing a frequency P co = P(r d   r u ) where P(r d   r u ) = probability that r d   r u

Copyright © 2003, Dharma P. Agrawal and Qing-An Zeng. All rights reserved Cochannel Interference Cells having the same frequency interfere with each other r d is the desired signal r u is the interfering undesired signal  is the protection ratio for which r d   r u (so that interference of r u is limited by  ) Cochannel interference between cells using the same frequency P co = P(r d   r u ) where P(r d   r u ) = probability that r d   r u

Copyright © 2003, Dharma P. Agrawal and Qing-An Zeng. All rights reserved 55 The End of Section 3