4.radio Propagation.pdf

Radio Propagation Theory Electric wave propagation in mobile environment

Radio Wave propagation models 

Radio wave propagation models are used in the network planning process to predict the average received signal strength at a given distance from the transmitter for large scale (>1km) and for small-scale or fading propagation models.



For such fading propagation models the variability of the signal strength at a close spatial proximity to a particular location is of major interest.

Radio wave propagation Mechanism Under ideal conditions radio waves propagate in free space from the transmitter to the receiver straight in all directions without being disturbed.

In real environments the received signal power is influenced by the following effects: • reflection • scattering • diffraction at edges • shadowing behind buildings or walls • fading • refraction.

Radio wave propagation mechanism –contd..

Signal Propagation Ranges

Multipath propagation

Fig. Multi-path propagation

Fig: Inter-symbol interference

General Classification

General classification of propagation models

Free Space and Log Distance Model

Fig: Free Space or Log Distance Model

Definition of Path Loss and Path Loss Exponent  The Path Loss is the difference (in dB) between the transmitted power (PT) and the received power (PR).  It represents signal level attenuation caused by free space propagation, reflection, diffraction and scattering.  The path loss is a fundamental unit for radio network planners, since it affects the minimum number of basestations (dimensioning) and the necessary output power (link budget). ( Here n is called pathloss exponent2)  The higher the frequency the higher the path loss!

Fig: Definition of path loss exponent

Okumura-Hata Model The advantage of the Okumura-Hata model is that the equations require only the following four parameters and needs therefore a very short computation time: • frequency (f = 150…1500 MHz) • distance between transmitter and receiver (r = 1 …20 km) • antenna height of transmitter (hBS = 30…200 m) • antenna height of transmitter (hMS = 1…10 m).

The isotropic Okumura-Hata model presumes a quasi flat surface without e.g. buildings, trees, walls, etc. It is not applicable when the BS is below the rooftops. Nevertheless it is the most applied model in network planning.

To improve the accuracy of this model an effective antenna height has been introduced. Be aware that the terrain profile between BS and MS is not taken into account.

Fig: Okumura-Hata model - Effective antenna height

The path loss is calculated according the following formulas:

Neither hills between BS and MS nor local effects around the receiver (reflections, shadowing, etc.) are taken into account.

Okumara-Hata Model Example

Okumara-Hata Model Example Contd..

COST-Hata Model The new model is known under the name: COST-Hata model and is defined for the following parameters: • frequency (f = 1500…2000 MHz) • distance between transmitter and receiver (r = 1 …20 km) • antenna height of transmitter (hBS = 30…200 m) • antenna height of transmitter (hMS = 1…10 m).

Figure: COST-Hata model example

Knife-Edge Model

The knife-edge effect is explained by Huygens' Principle, named for the Dutch physicist Christiaan Huygens, which states that a welldefined obstruction to an electromagnetic wave acts as a secondary source, and creates a new wave front. This new wave front propagates into the geometric shadow area of the obstacle.

Fig: Diffraction loss as function of v

Ray-Tracing Method Deterministic models, like Ray-Tracing, are most accurate. They are site-specific and require enormous number of geometric site information and are therefore only feasible for very limited areas like indoor pico cells.

COST-Walfish-Ikegami Model The most important semi-empirical model that is developed especially for urban areas with BS antennas placed below the roof tops is called COST-Walfish-Ikegami. It is defined for the following parameters: • frequency (f = 800…2000 MHz) • distance between transmitter and receiver (r = 20 …5000 m) • antenna height of transmitter (hBS = 4…50 m) • antenna height of transmitter (hMS = 1…3 m).

Fig: COST-Walfish-Ikegami model

Further parameters that are shown in the figure are: h: mean value of building heights (m) b: mean value of building separation (m) w: mean value of street widths (m) φ : mean angle between direct radio path and street (°).

COST-Walfish-Ikegami model

Even though the COST-Walfish-Ikegami model is suitable only for chessboard-like inner cities, it is commonly used for all kinds of city centres. Be aware that this model shows often significant inaccuracies, when using it for irregular building pattern like historically grown towns. The main advantage of this semi-empirical model is its short computation time.

Combinations of Models, Applicability The mentioned empirical models (Okumura-Hata and COST-Hata) are only applicable for a quasi flat surface. In combination with the deterministic model Knife-Edge they can be extended to hilly terrain or mountainous areas: • Okumura-Hata and Knife-Edge • COST-Hata and Knife-Edge. Data for empirical models can be collected from propagation measurements. Empirical (e.g. Okumura-Hata) or semi-empirical models (e.g. COSTWalfish- Ikegami) includes some free parameters which can be tuned in order to fit the model with measured samples. This makes the model applicable in certain environment.

Fig: Applicabilityof propagation models

Remark: Ray-tracing can be applied to urban or rural environments, but due to the very long computation time it is technically not feasible.

Fading Effects

Fading effects

Fig: Fading effects

Slow Fading

Fig: Slow fading - Q function

Fast Fading

Fig: Fast fading - Doppler shift

• Thank You

Radio Environment Features  MS antenna is not very high above the ground surface, so the received signal mainly consist of multiple reflection waves generated by the dispersion objects near the MS.  MS is often in the constant moving state, so the received signal amplitude and phase will vary with time and place.  Wave-guide effect in city streets.  Man-made noise:like industrial noise

 Sever interference: co-frequency, neighbor frequency and mutual modulation interference.

Propagation Mode • Electric wave propagation mode The line-of-sight direct wave and reflection wave, most of them are multiple reflection waves generated by the dispersion objects near the MS. Multiple path reflection include two type of signals • Reflection from far place generated by dispersion • Reflection within 50-400 wave length generated by the reflection and dispersing objects near the near MS.

Field intensity The real received signal level is formed by overlaying the fast fading upon the slow fading signals. r (t )  m(t )  r0 (t )

r (d )  m(d )  r0 (d )

Where d is the distance, t is time. m(x)--local mean value,which is synthesized by long-term fading and space loss,slow fading,long-term fading. r0(x)-short-term fading,fast fading,Rayleigh fading local mean value change much with distance than time.

Field intensity • Definition of local mean value 1 m( x )  2L

x L

 r ( y )dy

x_l

The average value of receiving signal. 2L: 40 , 36-50 sampling points.

• The geographic conditions and objects in the service area are basically not changeable in a certain period of time, so the local mean value m(x) is a fixed value for the fixed BTS in a certain place.

Field intensity test  Suppose the car speed is v, the test equipment (e.g. high speed sweep generator) samples n receiving powers of the same point in a second, the wavelength is λ（1/3m for 900MHz, and 1/6m for 1800MHz）, then:  40λ/v*n<=50 and 40λ/v*n>=36  For instance, at 900MHz, the test equipment samples 25 data of the same frequency point, then the reasonable car speed is 24~33km/h.

Propagation Loss • Definition:The power attenuation between output signal of transmitting antenna and that of the receiving antenna. • The relationship between the propagation loss and distance is proportional.around 40db/dec.

Propagation Model

In the free space, the electric wave transmitted will not generate the reflection, refraction, dispersion, diffraction and absorption, instead there will be only attenuation caused by diffusion, so its basic transmission loss Lbf can be calculated with the formula below: Lbf  32.5  20 log f MHz  20 log d km

Propagation Model

 Okumura-Hata 1、Application conditions: 150~1500MHz（GSM900）； Effective BTS antenna height :30~200m； MS antenna height: 1~10m； Communication distance: 1~20km； Quasi-smooth topography、urban downtown、suburb、open air、hills、mountain、waters；

Propagation Model 2、Propagation loss formula(Urban downtown): Lb城  69.55  26.16 lg f 13.82 lg hb  a(hm )  (44.9  6.55 lg hb )(lg d )

3、Correction factor： City type：Suburb、Open air、open air、Rural area； Topography：Hill、Slope、Isolate mountain、Sea(lake) mixed； City dedicated correction factor：Street 、Building density；

4、Total path loss： L  Lb  K street

 0 K s R K  u  h  S (a)    K sp   K mr K  im  Q0 0   Qr

Propagation Model For 1800Mhz use the COST-231Model

Application condition: GSM900/1800； Effective BTS antenna height :30~200m； MS antenna height: 1~10m； Communication distance: 1~35km； Quasi-smooth topography、urban downtown、suburb、open air、hills、mountain、waters；

Lb城  46.3  33.9 lg f  13.82 lg hb  a(hm )  (44.9  6.55 lg hb )(lg d )

Propagation Model

General Model 1、Application condition: GSM900/1800； Effective BTS antenna height :30~200m； MS antenna height: 1~10m； Communication distance: 1~35km(may be farther)； Quasi-smooth topography、urban downtown、suburb、open air、hills、mountain、waters；

2、Basic idea： Combine Okumura-Hata with Cost-231 formula； Consider clutter correction factor； All correction factor K can be modified by test data.

Propagation Model 3、Propagation loss formula: 1、One-stage: Lb城  K1  K 2 lg d  K 3 lg hb  K 4 lg d lg hb  K 5  K clutter  K d Ld  a(hm )  K street

2、Two-stage: K1  K 21 lg d  K 3 lg hb  K 4 lg d lg hb  K 5  K clutter  K d Ld  a(hm )  K street d  d0  Lb城   K1  K 22 lg d  K 3 lg hb  K 4 lg d lg hb  K 5  K clutter  K d Ld  a(hm )  K street  ( K 21  K 22 ) lg d 0 d  d 0

Propagation Model COST-231-Walfish-Ikegami 1、Application condition: GSM900/1800：

MS antenna height 1~10米； Communication distance:20m~5km；

2、Basic idea： The macro cellular model basis is: the propagation loss between the BS and the MS is dependent on the surrounding environment of the MS; but within 1km, the direction of the buildings around the BS and the streets seriously affect the propagation loss between the BS and the MS, so the macro cellular model mentioned earlier is not suitable for the forecast within 1km.

Propagation Model 3、Propagation loss formula： Visual pass:

Lb  42.6  26 lg d( km)  20 lg f ( MHz )

Non-visual pass:

Lb  L0  Lrts  Lmsd

L0—— transmission loss in free space Lrts——diffraction and dispersion loss from roof to street Lmsd——multi-shield diffraction loss

Propagation Model Micro Cell: Cost-231-Walfish-Ikegami Model 900M in Urban Area: Okumura-Hata Model 1800M in Urban Area: COST-231 Model

Correction of propagation model The necessity of propagation model correction As the propagation environment of the empirical propagation model is different from the actual propagation environment, it is necessary to test the electric wave propagation in a typical environment in the area where the GSM network is to be constructed, and use the test data to correct the propagation model and enhance the correctness of the propagation forecast.

Field intensity test method Select the test station To select the typical propagation environment in the service area, such as densely-populated urban area, general urban area and suburban area. To select a station that can cover as more clutter types as possible (based on the electric map); The antenna of the test station shall be higher by 5m than the obstacles within the range of 150~200m; for every artificial environment, it would be better to select three or more test stations to eliminate the location factor as much as possible.

Field intensity test method Determine the parameters related to the test station Use the omni-directional antenna Ensure the cleanness of the test frequency point Record the longitude and latitude of the test station, its antenna height, antenna type, feeder cable loss, transmitting power of the transmitter, gain of the receiver

Field intensity test method Determine the test route 1. Get the test data of different distances in different directions; 2. There should be at least 4~5 test data in a certain distance to eliminate the location effect; 3. Try to go through various clutters; 4. Try to avoid selecting the expressways or wider highways. The optimal selection is the road that is not more than 3m in width.

 way test path

Spiral test path

Field intensity test method Requirement for the electronic map

Rural areas: 50m and 100m Cities and suburbs:20m Micro cells: 5m

Correction factor Lb =K1+K2lgd+K3lghb+K4lgdlghb+K5+K6+K7 • K1 (146.83 900MHz)(156.65 1800MHz)

K2  K3  K4  K7 

44.9 -13.82 -6.55 0.5

Modify K5、K6 K5 Artificial environment factor K6 clutter factor

Correction factors

Correction factor in common urban area Factor K1 K2 K3 K4 K5 City 156.65 44.9 -13.8 -6.55 -9.99 Clutter factor K6: Water: -4.52 Open air in suburb area: -2.24 Open air in city: -8.49 Grass: -4.21 Middle density build below 20m: 3.11

K7 0.5