Channel Modelling ETIM10. Channel models

Transcription

1 Channel Modelling ETIM10 Lecture no: 6 Channel models Fredrik Tufvesson Department of Electrical and Information Technology Lund University, Sweden Fredrik.Tufvesson@eit.lth.se Fredrik Tufvesson - ETIN10 1

2 Content Modelling methods Okumura-Hata path loss model COST 231 model Indoor models Wideband models COST 207 (GSM model) ITU-R model for 3G Directional channel models Multiantenna (MIMO) models Ray tracing & Ray launching Fredrik Tufvesson - ETIN10 2

3 Channel measures Fredrik Tufvesson - ETIN10 3

4 Modeling methods Stored channel impulse responses realistic reproducible hard to cover all scenarios Deterministic channel models based on Maxwell s equations site specific computationally demanding Stochastic channel models describes the distribution of the field strength etc mainly used for design and system comparisons Fredrik Tufvesson - ETIN10 4

5 Narrowband models Review of properties Narrowband models contain only one attenuation, which is modeled as a propagation loss, plus large- and small-scale fading. Path loss: Often proportional to 1/d n, where n is the propagation exponent (n may be different at different distances). Large-scale fading: Log-normal distribution (normal distr. in db scale) Small-scale fading: Rayleigh, Rice, Nakagami distributions... (of amplitudes and not in db-scale) Fredrik Tufvesson - ETIN10 5

6 Okumura s measurements Extensive measurement campaign in Japan in the 1960 s. Parameters varied during measurements: Frequency Distance Mobile station height Base station height Environment MHz km 1 10 m m medium-size city, large city, etc. Propagation loss is given as median values (50% of the time and 50% of the area). Results from these measurements are displayed in figures in the appendix Fredrik Tufvesson - ETIN10 6

7 Okumura s measurements excess loss FIGURE 7.12 in appendix Excess loss [db] Distance [km] These curves are only for h b =200 m and h m =3 m Frequency [MHz] 900 MHz and 30 km distance Fredrik Tufvesson - ETIN10 7

8 The Okumura-Hata model Background In 1980 Hata published a parameterized model, based on Okumura s measurements. The parameterized model has a smaller range of validity than the measurements by Okumura: Frequency Distance Mobile station height Base station height MHz 1 20 km 1 10 m m Fredrik Tufvesson - ETIN10 8

9 The Okumura-Hata model How to calculate prop. loss Metropolitan areas Small/mediumsize cities Suburban environments Rural areas ( ) km LO H= A+ Blog d + C ( 0 MHz ) ( b ) ( m ) A= log f log h a h B= log ( ( hm )) ( ( hm )) ( h ) ah ( m ) = ( 1.1log( f0 MHz ) 0.7) ( f0 MHz ) h ( 1.56log 0.8) log for f 200 MHz 2 b 3.2 log for f 400 MHz m Fredrik Tufvesson - ETIN C = 0 0 ( f ) 2 0 MHz 2 log / ( f ) 0 MHz ( f0 MHz ) h b and h m in meter 4.78 log log 40.94

10 The COST 231-Walfish-Ikegami model The Okumura-Hata model is not suitable for micro cells or small macro cells, due to its restrictions on distance (d > 1 km). The COST 231-Walfish-Ikegami model covers much smaller distances, is better suited for calculations on small cells and covers the 1800 MHz band as well. Frequency Distance Mobile station height Base station height MHz km 1 3 m 4 50 m Fredrik Tufvesson - ETIN10 10

11 The COST 231-Walfish-Ikegami model How to calculate prop. loss L= L0 + Lmsd + Lrts Free space Building multiscreen Roof-top to street BS MS Details about calculations can be found in the appendix. d Fredrik Tufvesson - ETIN10 11

12 Motley-Keenan indoor model For indoor environments, the attenuation is heavily affected by the building structure, walls and floors play an important rule PL PL 0 10nlog d/d 0 F wall F floor distance dependent path loss sum of attenuations from walls, 1-20 db/wall sum of attenuation from the floors (often larger than wall attenuation) site specific, since it is valid for a particular case Fredrik Tufvesson - ETIN10 12

13 Wideband models Tapped delay line model often used N ht, τ = α texp jθ t δ τ τ ( ) ( ) ( ) i= 1 ( ) ( ) i i i Often Rayleigh-distributed taps, but might include LOS and different distributions of the tap values Mean tap power determined by the power delay profile Fredrik Tufvesson - ETIN10 13

14 Power delay profile Often described by a single exponential decay P sc () τ exp( τ / Sτ ) τ 0 = 0 otherwise log( Psc( τ )) τ delay spread though often there is more than one cluster P() τ = k P S c k c τ, k P sc 0 ( τ τ ) τ 0 c 0, k otherwise log( Psc( τ )) τ Fredrik Tufvesson - ETIN10 14

15 arrival time If the bandwidth is high, the time resolution is large so we might resolve the different multipath components Need to model arrival time The Saleh-Valenzuela model: h Double-exponential L l 0 K k 0 shape for the ray power: k,l T l k,l cluster arrival time (Poisson) ray arrival time (Poisson) Fredrik Tufvesson - ETIN10 15

16 Wideband models COST 207 model for GSM The COST 207 model specifies: FOUR power-delay profiles for different environments. FOUR Doppler spectra used for different delays. IT DOES NOT SPECIFY PROAGATION LOSSES FOR THE DIFFERENT ENVIRONMENTS! Fredrik Tufvesson - ETIN10 16

17 Wideband models COST 207 model for GSM Four specified power-delay profiles P [ db] RURAL AREA P [ db] 30 τ [ µs ] τ [ µs] TYPICAL URBAN P [ db] BAD URBAN P [ db] 30 τ [ µs ] τ [ µs] Fredrik Tufvesson - ETIN HILLY TERRAIN

18 Wideband models COST 207 model for GSM Four specified Doppler spectra P ντ, s ( ) i (, ) P ντ s i CLASS GAUS1 τ 0.5 µ s 0.5 µ s < τ 2 µ s i i ν max 0 (, ) P ντ s +ν max i ν max 0 (, ) P ντ s +ν max i GAUS2 τ 2 µ s i RICE > Shortest path in rural areas ν max 0 +ν max ν max 0 +ν max Fredrik Tufvesson - ETIN10 18

19 Wideband models COST 207 model for GSM Doppler spectra: CLASS GAUS1 GAUS2 P [ db] RURAL AREA P [ db] TYPICAL URBAN First tap 10 RICE 20 here τ [ µs 30 ] τ [ µs] P [ db] BAD URBAN P [ db] 30 τ [ µs ] τ [ µs] Fredrik Tufvesson - ETIN HILLY TERRAIN

20 Transfer function, Typical urban Fredrik Tufvesson - ETIN10 20

21 Wideband models ITU-R model for 3G The ITU-R model specifies: SIX different tapped delay-line channels for three different scenarios (indoor, pedestrian, vehicular). TWO channels per scenario (one short and one long delay spread). TWO different Doppler spectra (uniform & classical), depending on scenario. THREE different models for propagation loss (one for each scenario). The standard deviation of the log-normal shadow fading is specified for each scenario. The autocorrelation of the log-normal shadow fading is specified for the vehicular scenario Fredrik Tufvesson - ETIN10 21

22 Wideband models ITU-R model for 3G ns Fredrik Tufvesson - ETIN10 22

23 Directional channel models The spatial domain can be used to increase the spectral efficiency of the system Smart antennas MIMO systems Need to know directional properties How many significant reflection points? Which directions? Model incoming angle (direction of arrival) and outgoing angle (direction of departure) to scatterers Model independent of specific antenna pattern Fredrik Tufvesson - ETIN10 23

24 Double directional impulse response TX position RX position number of multipath components for these positions h t, r TX, r RX,,, N r 1 h t, r TX, r RX,,, delay direction-of-departure direction-of-arrival h t, r TX, r RX,,, a e j Fredrik Tufvesson - ETIN10 24

25 Double directional impulse response with slightly different notation: Time and location is omitted here! Fredrik Tufvesson - ETIN10 25

26 Physical interpretation Ω τ l Ψ Fredrik Tufvesson - ETIN10 26

27 Angular spread E s,,, s,,, P s,,, double directional delay power spectrum DDDPS,, P s,,, d angular delay power spectrum ADPS, DDDPS,, G MS d τ l angular power spectrum APS APDS, d power P APS d Fredrik Tufvesson - ETIN10 27

28 Geometry-based stochastic channel models Assign positions for scatterers according to given distributions Derive impulse response given the scatterers and distributions for the signal properties. Used in the COST 259 model, COST 273, COST 2100, WINNER 3GPP/3GPP Fredrik Tufvesson - ETIN10 28

29 Geometry-Based Stochastic Channel Model (GSCM) Create an imaginary map for radio wave scatterers (clusters) Local cluster MS 2 Cluster MS 1 Local cluster BS Cluster Fredrik Tufvesson - ETIN10 Courtesey: K. Haneda, Aalto Uni. 29

30 Fredrik Tufvesson - ETIN10 30 MIMO channel channel matrix = ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( Tx Rx Rx Rx Tx Tx τ τ τ τ τ τ τ τ τ τ M M M M M M h h h h h h h h h! " # " "!! H signal model ( ) ( ) ( ) = = 1 0 D t t τ τ H τ x y

31 Deterministic modeling methods Solve Maxwell s equations with boundary conditions Problems: Data base for environment Computation time Exact solutions Method of moments Finite element method Finite-difference time domain (FDTD) High frequency approximation All waves modeled as rays that behave as in geometrical optics Refinements include approximation to diffraction, diffuse scattering, etc Fredrik Tufvesson - ETIN10 31

32 Ray launching TX antenna sends out rays in different directions We follow each ray as it propagates, until it either Reaches the receiver, or Becomes too weak to be relevant Propagation processes Free-space attenuation Reflection Diffraction and diffuse scattering: each interacting object is source of multiple new rays Predicts channel in a whole area (for one TX location) Fredrik Tufvesson - ETIN10 32

33 Ray tracing Determines rays that can go from one TX position to one RX position Uses imagining principle Similar to techniques known from computer science Then determine attenuation of all those possible paths Fredrik Tufvesson - ETIN10 33

34 Example: Ray tracing Required base station power to connect to a WCDMA cell phone. Example from Stuttgart. Courtesey: Awe-communications Fredrik Tufvesson - ETIN10 34

35 Example: Ray tracing Coverage for a WCDMA cell phone. Example from Stuttgart. Courtesey: Awecommunications Propagation Models Fredrik Tufvesson - ETIN10 35