Page 343 - Integrated Wireless Propagation Models
P. 343

I n - B u i l d i n g   ( P i c o c e l l )   P r e d i c t i o n   M  o d e l s    321

               The wall attenuation factor P ·  W(K) is shown in the formula, where K is the number of
                        P
               floors and  i s the number of walls. The formula can be enhanced to count and sum up
               the loss of each wall and floor if the material is different; thus, the penetration loss is
               different.
               5.4.1 . 1    Comparison of the  Keenan-Motley Model  and  the  Enhanced Lee Model  In this sec­
               tion, the Lee and Keenan-Motley models are made comparison from their path-loss
               predictions in two situations: LOS and NLOS. Many different values are applied to the
               Keenan-Motley Model's parameters. In the LOS situation, three different attenuation
               factors-0.2 dB/m,  0.15 dB/m,  and 0.1 dB/m-are  applied  to the Keenan-Motley
               model for getting the path losses that are used to compare with Lee model. In the NLOS
               situation, the wall loss of 10 dB is applied to the attenuation factors of 0.2-dB/m and
               0.15-dB/m cases. The path losses from Keenan-Motley model for these two cases are
               compared with that from the Lee prediction model in two scenarios: the scenario of
               multiple rooms in the middle of the floor and the scenario of one room at the end of the
               floor. The general formula of Lee in-building path-loss formula is shown in Eq. (5.3.5.1).
               The equation of the Keenan-Motley model is expressed in Eq. (5.2.8.1.1) as

                      L (dB) = 32.5 + 20 log(j) + 20 log(d) + K ·  F(K) + P ·  W(K) + D(d-Db)   (5.4.1.3)
                  Each symbol of Eq. (5.4.1.3) is described in Eq. (5.2.8.1.1). The first three terms are
               for LOS only. The entire equation is for NLOS. The results are described below.

               5.4. 1.2   LOS  Situation  Here we show the performances of both models in the LOS
               situation. The path loss after the breakpoint, using the attenuation factors of 0.2 dB/m
               and 0.1 dB/m, are compared with the enhanced Lee model with the slope based on the
               LOS loss formula shown in Fig. 5.4. . 3. Besides, a medium difference is used to compare
                                            1
               the predictions between the medium values from the Keenan-Motley model and the
               values from the Lee model.
                  As shown in Fig. 5.4.1.1, the medium differences among all the curves are within
               8 dB at any spot over 200  m.  This means that the differences in predicted values
               between the two models are small. Therefore, the user can choose to use either one of
               these. The Lee model appears to be more optimistic than the Keenan-Motley model in
               the LOS situation. It pretty much follows the 0.1-dB/m attenuation curve. Usually, in
               the LOS situation, the RF link should suffer less attenuation over a long distance. On
               some occasions, perhaps the 0.2-dB/m loss may have to calculate the path loss in the
               scenarios of behind and around building for the Keenan-Motley model, which is not
               in a LOS situation. The Lee model has predicted these scenarios (around the building),
               stated in Sec. 5.3.5.3.
                  The enhanced Lee model shows an 8-dB stronger signal strength at the end of the
               coverage distance as compared with the result obtained from Keenan-Motley model
               with a scenario of using a 0.2-dB/m attenuation factor. The Keenan-Motley model
               sometimes predicts the results conservatively.

               5.4. 1.3   NLOS Situation  Here we show the performances of both models in the NLOS
               situation. There are two major scenarios shown in the NLOS situation. One scenario is
               that to many rooms being located in a straight line along the radial path. The other
               scenario is that of one room being located at the end of the building.
   338   339   340   341   342   343   344   345   346   347   348