Page 46 - ITU Journal: Volume 2, No. 1 - Special issue - Propagation modelling for advanced future radio systems - Challenges for a congested radio spectrum

ITU Journal: ICT Discoveries, Vol. 2(1), December 2019

(OLOS) to NLOS when the Rx turns around the 4.2 Urban intersections
corner, the Tx was fixed, while the Rx was moved to
different locations as in Fig. 4. The antennas were To develop next generation wireless systems for
placed at a typical height of a vehicle at 1.55 m and vehicles, detailed propagation models are needed.
both the Tx and Rx antennas with 30° HPBW were Particularly, there is a need for a spatially consistent
rotated with 30° to scan the complete azimuth model that can reflect the non-stationarities of the
domain at 0° elevation to synthesize the vehicular channel and that can support multi-
omnidirectional power delay profile (PDP), as in antenna configurations at both ends, either for
Fig. 5. Five Rx positions were measured to show the beamforming purposes, or for spatial multiplexing,
transition from LOS to being obstructed by the or just for diversity. Previously, member
vegetation in the corner and then being blocked by institutions of the COST 2100 and IC1004 actions
Building B. Position 1 was measured twice: Position have derived a GSCM for V2V communication in
1a without parked vehicles in the scenario (blue highway scenarios and a non-stationary model for
bullets), and Position 1b with parked vehicles in the vehicular communication, but a detailed
surrounding area. The resulting total link budget is measurement-based propagation model supporting
shown in Fig. 6, where there is a total maximum loss multiple antenna configurations with realistic
(path loss plus obstruction loss) of approximately spatio-temporal characteristics in urban
17 dB at 60 GHz. intersections has been lacking until now.
The COST IRACON channel model for urban
intersections [11] is a GSCM based on a street
geometry defined by a map. It can thus represent
typical intersection scenarios or specific
intersections. Specular scatterers are randomly
dropped, with a given density (number of specular
scatterers per unit area), in the simulation area in
bands along the walls according to the geometry of
Fig. 4 – Measurement set-up in the corner scenario the intersection. These scatterers are then labeled
as first, second and third order reflection points.
Similarly, diffuse scatterers are also randomly
placed along with the walls, but in wider bands. For
simulations of specific intersections, especially for
simulations of wider or more open intersections,
scatterers can also be dropped in areas that are not
Fig. 5 – Synthetic omnidirectional PDP for position 1b aligned with the walls. These scatterers typically
represent contributions from lamp posts or larger
street signs, which typically can be observed in
measurements. Table 3 summarizes the intensity, χ,
of different types of scatterers as well as the width,
W, of the bands of scatterers along the walls.
Table 3 – Intensity and width of the bands for random
scatterer drop for the different types of MPCs
Type Order χ (m ) W (m)
−2
Wall 1 st 0.044 3
Wall 2nd 0.044 3
Fig. 6 – Received power vs Tx-RX distance Wall 3 rd 0.044 3
Non-wall 1 st 0.034 User defined
Diffuse, wall 1 st 0.61 12
Diffuse, non-wall 1 st 0.61 User defined

30 © International Telecommunication Union, 2019