Content uploaded by Othman Zahid

Author content

All content in this area was uploaded by Othman Zahid on Jan 18, 2022

Content may be subject to copyright.

URSI GASS 2020, Rome, Italy, 29 August - 5 September 2020

Clutter Loss Measurements and Modeling at 26 GHz Band

J. Huang, O. Zahid, and S. Salous*

Department of Engineering, Durham University, Durham, DH1 3LE U.K, *sana.salous@durham.ac.uk

Abstract

Clutter loss refers to the power loss caused by clutter

objects such as buildings and vegetation. The

determination of clutter loss is crucial for the deployment

of wireless communication networks. In this paper, we

present results of clutter loss measurements at 26 GHz in

a campus environment. The ITU-R P.2108-0 statistical

clutter loss model is applied to model the clutter loss and

the cumulative distribution functions (CDFs) of the

measured and modeled clutter losses are compared, which

show good agreement.

1 Introduction

As defined in ITU-R P.2108-0 [1], clutter refers to objects

such as buildings and vegetation, which are on the surface

of the earth but not terrain. Clutter loss is defined as the

difference in the transmission loss or basic transmission

loss with and without the presence of terminal clutter at

either end of the path with all other path details being the

same.

The determination of clutter loss is crucial for the

deployment of wireless communication networks, which

needs to be obtained from real clutter loss measurements.

Clutter loss measurements require one end of the link to

be above clutter level while the other terminal is in the

clutter. An illustration of clutter loss measurement setup is

shown in Fig. 1. At the transmitter (Tx) side, a horn

antenna is usually placed above the clutter for example on

the rooftop of a building and down tilted, while the

receiver (Rx) is behind clutter and equipped with an

omni-directional antenna.

Figure 1. An illustration of clutter loss measurements.

Only a few clutter loss measurements have been reported

in the literature below sub-6 GHz band, and in the

millimeter wave (mmWave) band. In [2], clutter loss

measurements were conducted in three cities in the US in

the 1755-1780 MHz band. The median clutter loss was

found to be 20-27 dB. In [3], the independent and joint

clutter loss and building entry loss were measured at 3.5

GHz. The results showed that the clutter loss and building

entry loss cannot be treated as multiplicative, and the

combined loss was related to the details of building

geometry and surrounding environment. In [4], clutter

loss measurements were conducted at two bands around

28 GHz and 38 GHz in an urban low-rise environment.

The clutter loss was found to depend on the placements of

the Tx and Rx antennas among buildings. In [5], clutter

loss measurements were conducted at 27 GHz in urban

campus environment. The maximum clutter loss was

about 35 dB. In [6, 7], clutter loss measurements were

conducted at 26 GHz and 40 GHz in suburban

environments in the range of 120-350 m. The clutter loss

with respect to the elevation angle range of 0°-5° was

analyzed. In addition, ray tracing simulation results were

compared with the measurement results. In [8], the clutter

loss and building entry loss were measured at 26 GHz.

Building entry loss was higher than clutter loss above the

median value due to measurements taken on the higher

floors. In [9], above rooftop channel measurements were

conducted in a suburban environment at 32.4 GHz. The

paths along the road, between houses, and on the roof

were measured. A site-specific path loss and clutter loss

model was proposed. In [10], an overview of the new

developed ITU-R P.2108-0 model was given. Extensive

measurements in Aalborg, Gothenburg, and Tokyo were

presented. The clutter loss in Aalborg was measured in a

residential area at 18 GHz band with distances up to 1.4

km. A clear increasing trend of clutter loss was found up

to 0.8 km, while it became constant for larger distances.

The clutter loss in Tokyo was measured at 2.2, 4.7, 26.4,

and 66.5 GHz in the range of 0.26-1.2 km.

In this paper, we present clutter loss measurements at 26

GHz, as it is one of the “pioneer bands” in Europe and

identified by ITU WRC-19 recently for fifth generation

(5G) high data rate transmission. The results are

compared with the ITU-R P.2108-0 statistical clutter loss

model.

The remainder of the paper is organized as follows.

Section 2 describes the 26 GHz clutter loss measurements.

The ITU-R P.2108-0 statistical clutter loss model is

presented in Section 3. The measurement and modeling

results and analysis are given in Section 4. Finally,

conclusions are drawn in Section 5.

2 26 GHz Clutter Loss Measurements

The custom-designed channel sounder [11] is used to

conduct 26 GHz clutter loss measurements along three

routes at Durham University, behind buildings indicated

as (R1), (R2), and (R3), shown in Fig. 2(a)-(c). The Tx is

located at a height of about 18.2 m with a down tilt angle

of 12°, as shown in Fig. 2(d). The measured frequency

band is 24.68-27.68 GHz at a sweep repetition frequency

of 1.22 kHz. A horn antenna with 20 dBi gain and 18°

beamwidth is used at the Tx side, while an omni-

directional antenna is used at the Rx side with a height of

1.6 m. The main beam of the Tx antenna is adapted to

cover the three measurement routes, as shown in Fig. 2(e).

A total of 102, 150, and 47 positions were measured in R1,

R2, and R3, respectively. The corresponding Tx-Rx

distances were 50-80 m, 60-100 m, and 100-105 m for the

three routes, respectively.

(a) (R1)

(b) (R2)

(c) (R3)

(d) (Tx location)

(e) Measurement setup

Figure 2. Clutter loss measurement environments.

3 Statistical Clutter Loss Model

In ITU-R P.2108-0 recommendation, there are three

clutter loss models, for different terminal environments

and frequency ranges. The statistical clutter loss model for

the frequency range of 2-67 GHz is applied to model the

26 GHz clutter loss.

In the statistical clutter loss model, the input parameters

include frequency f (GHz), Tx-Rx distance d (km), and

percentage of locations p (%). The clutter loss not

exceeded for p% of locations for the terrestrial to

terrestrial path is given as

log

dB

(1)

(2)

(3)

where Q–1(p/100) is the inverse complementary normal

distribution function, and are the long-range and

short-range clutter loss, respectively [10]. An illustration

of the median clutter loss calculated from the model is

shown in Fig. 3. As the frequency increases, the median

clutter loss becomes larger. Below the distance of about 1

km, the median clutter loss shows a linear relationship

with the distance in log scale. When the distance becomes

larger, the median loss does not change, i.e., goes to the

saturation range.

Figure 3. An illustration of the median clutter loss

calculated from the ITU-R P.2108-0 clutter loss model.

4 Results and Analysis

The measured power delay profile (PDP) is expressed as

PDP

(4)

where L is the number of multipath components (MPCs)

with powers above the threshold, and are the power

and delay of the lth MPC, respectively.

The clutter loss can be calculated as

CL

(5)

(6)

FSPL

(7)

where P and FSPL are the total path loss and free space

path loss, respectively.

The measured PDPs and path loss variations of R1, R2

and R3 are shown in Figs. 4-6, respectively. As can be

seen, R1 shows a low probability of clutter loss, which is

possible for a short link. R2 has the largest clutter loss as

the buildings are high and dense, which causes higher

losses. R3 has the smallest clutter loss variation, as the

distance variations are over a small range. The

measurement results show high correlation with the

building geometry and surrounding environments.

(a) Measured PDPs of R1.

(b) Path loss variations of R1.

Figure 4. Measured PDPs and path loss variations of R1.

(a) Measured PDPs of R2.

(b) Path loss variations of R2.

Figure 5. Measured PDPs and path loss variations of R2.

(a) Measured PDPs of R3.

(b) Path loss variations of R3.

Figure 6. Measured PDPs and path loss variations of R3.

The measured cumulative distribution functions (CDFs)

for R1, R2, and R3 are shown in Fig. 7. R1 has the

smallest clutter loss, while R2 has the largest clutter loss.

A comparison of the measured and modeled clutter losses

is shown in Fig. 8. For the statistical clutter loss model, as

the distance increases from 70 m to 90 m, the clutter loss

tends to be larger. For the measurement results, the

distance is in the range of 50-105 m. To avoid statistical

bias, the combined clutter loss values are re-sampled at

some distances. The measured combined CDF is closest

to the statistical clutter model when d = 80 m. The

measured median clutter loss is 11 dB, which is similar to

the statistical clutter model when d = 80 m. The

comparison shows a good match between measured and

modeled clutter loss.

Figure 7. CDFs of the measured clutter loss.

Figure 8. Comparison of the measured and modeled

clutter losses.

5 Conclusions

In this paper, results of measurements conducted at 26

GHz using the customized channel sounder at Durham

University were presented. The measurement data were

processed to extract clutter loss. The ITU-R P.2108-0

statistical clutter loss model was utilized to model the

clutter loss at 26 GHz band. The clutter loss is found to be

related to the building geometry and surrounding

environment and the measured and modeled clutter loss

have shown good agreement.

6 Acknowledgements

The authors would like to acknowledge the support of

WaveComBE project, under Horizon 2020 research and

innovation program, grant agreement No. 766231, the

European Joint Research Centre, JRC, Ispra, Italy, EPSRC

under grant EP/I01049/1, PATRICIAN, and Intel, USA. The

authors would also like to acknowledge Amar Al-Jzari

and Mohamed Abdulali at Durham University for their

help in conducting the channel measurements.

7 References

1. ITU-R P.2108-0, Prediction of clutter loss, June 2017.

2. C. Hammerschmidt, R. Johnk, “Extracting clutter

metrics from mobile propagation measurements in the

1755–1780 MHz band,” in Proc. IEEE

MILCOM’16,

Baltimore, MD, USA, Nov. 2016, pp. 1-6.

3. R. Rudd, D. Wu, V. Ocheri, M. Nekovee,

“Independent and joint statistics of clutter loss and

building entry loss—Initial measurements,” in Proc.

IEEE PIMRC’18, Bologna, Italy, Apr. 2018, pp. 1-2.

4. Y. Yoon, J. Kim, K. Kim, M. Kim, Y. Chong, “Clutter

loss characteristic in mm-Wave bands for small urban

environment,” in Proc. ICTC’17, Jeju, South Korea,

Oct. 2017, pp. 1-3.

5. R. F. Rudd, M. Nekovee, “Millimetre-wave

propagation in urban clutter for 5G systems,” in Proc.

LAPC’17, Loughborough, UK, Nov. 2017, pp. 1-2.

6. B. Montenegro-Villacieros, J.-M. Chareau, J. Bishop,

P. Viaud, T. Pinato, M. Basso, “Clutter loss

measurements in suburban environment at 26 GHz and

40 GHz,” in Proc. EuCAP’18, London, UK, Apr. 2018,

pp. 1-4.

7. B. Montenegro-Villacieros, J. Bishop, J.-M. Chareau,

“Clutter loss measurements and simulations at 26 GHz

and 40 GHz,” in Proc. EuCAP’19, Krakow, Poland,

Mar.-Apr. 2019, pp. 1-5.

8. S. Salous, F. Tufvesson, K. Turbic, L. M. Correia, T.

Kürner, D. Dupleich, C. Schneider, D. Czaniera, B. M.

Villacieros, “IRACON propagation measurements and

channel models for 5G and beyond,” ITU Journal: ICT

Discoveries, 2, 1, Nov. 2019, pp. 1-9.

9. Y. K. Yoon, K. W. Kim, Y. J. Chong, “Site prediction

model for the over rooftop path in a suburban

environment at millimeter wave,” International Journal

of Antennas and Propagation, 2019, Article ID

1371498, Apr. 2019, pp. 1-12,

doi:10.1155/2019/1371498.

10. J. Medbo, C. Larsson, B.-E. Olsson, F. S. Chaves, H.

C. Nguyen, I. Rodriguez, T. B. Sorensen, I. Z. Kovács,

P. Mogensen, K.-T. Lee, J. Woo, M. Sasaki, W.

Yamada, “The development of the ITU-R terrestrial

clutter loss model,” in Proc. EuCAP’18, London, UK,

Apr. 2018, pp. 1-6.

11. S. Salous, S. M. Feeney, X. Raimundo, and A. A.

Cheema, “Wideband MIMO channel sounder for radio

measurements in the 60 GHz band,” IEEE Trans.

Wireless Commun., 15, 4, pp. 2825-2832, Apr. 2016.