ArticlePDF Available

Characterization of Spatial Consistency of Cluster Channels in Urban Environments at 24 and 60 GHz

Authors:

Abstract and Figures

Currently, millimeter-wave (mm-wave) based wireless communications in the 30 to 300 GHz frequency band are attracting new attention because the existing frequency band below 6 GHz is very congested. In order to design and assess mm-wave systems, it is essential to characterize its propagation channel considering the spatial consistency (SC) for non-stationary user equipment. The 3GPP has proposed an SC procedure that provides spatial correlation to clusters for beam-tracking evaluation in dynamic scenarios. To characterize SC, the 3GPP follows only the LSP-dependent correlation distance, whereas this work proposed a parametric method to obtain a more reliable and accurate SC using cluster tracking, which provides a cluster visible region. This letter reports on the evaluation of SC of clusters in terms of cluster visible region based on the obtained measured data in two different mm-wave frequencies, 24 GHz and 60 GHz, in urban environments. The visible region is estimated by tracking the available clusters using the four-dimensional spatiotemporal multipath parameters: the departure/arrival angles (azimuth), delay time, and cluster power. The obtained visible regions for two different frequencies and scenarios are compared and confirmed the dependencies of SC of the clusters on frequency and environment.
Content may be subject to copyright.
IEEE ANTENNAS AND WIRELESS PROPAGATION LETTERS, VOL.XX, NO.XX, SEPTEMBER 2023 1
Characterization of Spatial Consistency of Cluster
Channels in Urban Environments at 24 and 60 GHz
Naoya Suzuki, Hibiki Tsukada, Riku Takahashi, Banibrata Bag Member, IEEE, Minseok Kim Senior Member,
IEEE
Abstract—Currently, millimeter-wave (mm-wave) based wire-
less communications in the 30 to 300 GHz frequency band
are attracting new attention because the existing frequency
band below 6 GHz is very congested. In order to design
and assess mm-wave systems, it is essential to characterize its
propagation channel considering the spatial consistency (SC) for
non-stationary user equipment. The 3GPP has proposed an SC
procedure that provides spatial correlation to clusters for beam-
tracking evaluation in dynamic scenarios. To characterize SC,
the 3GPP follows only the LSP-dependent correlation distance,
whereas this work proposed a parametric method to obtain a
more reliable and accurate SC using cluster tracking, which
provides a cluster visible region.This letter reports on the
evaluation of SC of clusters in terms of cluster visible region
based on the obtained measured data in two different mm-wave
frequencies, 24 GHz and 60 GHz, in urban environments. The
visible region is estimated by tracking the available clusters using
the four-dimensional spatiotemporal multipath parameters: the
departure/arrival angles (azimuth), delay time, and cluster power.
The obtained visible regions for two different frequencies and
scenarios are compared and confirmed the dependencies of SC
of the clusters on frequency and environment.
Index Terms—Channel sounding, millimeter-wave, channel
model, clusters, spatial consistency.
I. INTRODUCTION
THE 5G network based on below 6 GHz band is
now highly congested due to the ever-increasing de-
mand for mobile traffic. Therefore, full-scale deployment of
an ultrahigh-speed 5G network is expected to utilize the
28 GHz or higher millimeter-wave (mm-wave) bands for
next-generation mobile communication systems. The channel
propagation characteristics in mm-wave bands significantly
differ from previously used sub-6 GHz channel characteristics.
Therefore, for the design and development of wireless systems
using mm-wave bands, it is essential to have a channel model
that reasonably reflects the propagation characteristics of the
mm-wave band. Adaptive antenna algorithms and channel
state feedback to properly simulate channel impairments and
variations require spatial consistency (SC) to ensure correlated
channel response over a short time and distance. Spatial con-
sistency is also essential to train and deploy massive multiple-
input multiple-output (MIMO) and multi-user beamforming
[1]. Several organizations and research communities have
This research has been conducted under the contract ”R&D for the
realization of high-precision radio wave emulator in cyberspace” (JPJ000254)
made with the Ministry of Internal Affairs and Communications of Japan.
(Corresponding author: Minseok Kim)
The authors are with Graduate School of Science and Technology, Niigata
University, Niigata, Japan
Hibiki Tsukada is currently with NTT Access Network Service Systems
Laboratories, Kanagawa, Japan.
recently been involved in channel measurements and modeling
using beyond 6 GHz bands to build next-generation (Beyond
5G/6G) mobile communications systems infrastructure. These
channel measurement and analysis of the mm-wave channel
characteristics reveals that out of many new additional channel
modeling components, SC is a mandatory one that must be
taken into consideration during mm-wave channel modeling
[2]–[8].
The Third Generation Partnership Project (3GPP) channel
model is one of the most popular for modeling the next-
generation wireless communications out of several proposed
mm-wave channel modeling approaches [4]. The 3GPP chan-
nel model has incorporated the features to meet the additional
requirement for modeling 5G systems by extending the pre-
vious drop-based model. However, in practical propagation
scenarios, the arriving angle of the narrow beam to a non-
stationary user equipment (UE) changes continuously with
its movement, which may incur a loss of up to 30 dB for
an offset of just a few degrees between the beam scanning
angle and the angle of the incoming ray. Therefore, to ensure
SC of clusters, the delay and angular characteristics (angles
of arrival and departure; AOAs and AODs) of the multipath
components (MPCs) in each cluster must vary consequently
and realistically as a function of UE position. However, in
the previous 3GPP spatial channel model (SCM), temporal
simulation for the beamforming is not taken into consideration,
and in the case of mobile user simulation, cluster AOAs, and
delays are generated independently without consideration of
the previous cluster parameters, i.e., this model does not follow
the spatiotemporal correlation. To encounter these issues,
Wang et al. [9] proposed a time-variant model where the
MPCs angular characteristics change with the cluster death
and birth process synchronized with spatial shift of UE.
In the current version of the 3GPP channel model [4], a
simple and analytically tractable method has been included to
maintain the SC of clusters, where the changes of UE position
lead to changes in clusters’ delay and angular characteristics
simultaneously.
Most previous studies regarding SC have either relied on
simulation-based evaluations or on the measurements con-
ducted in indoor environments [10]–[13]. However, investi-
gating the SC in outdoor environments as well as indoor
environments is also essential.Therefore, this paper evaluated
SC utilizing the measured data in an outdoor environment. In
this study, only the clusters generated by small objects (random
clusters) are investigated for quasi-deterministic (QD) channel
models [14] because the SC of the deterministic clusters
generated by large objects can be explicitly represented by ray-
IEEE ANTENNAS AND WIRELESS PROPAGATION LETTERS, VOL.XX, NO.XX, SEPTEMBER 2023 2
(a) Urban Micro (UMi). (b) Urban Macro (UMa).
Fig. 1: Measurement environments.
tracing (RT). In the current 3GPP channel model, correlation
distances of large-scale parameters (LSPs) are provided by
some categorized scenarios as an SC parameter. However,
following the approach in [13], [15], this paper characterizes
the cluster visible regions (survival distance), which impose
cluster-level SC, by tracking the clusters measured in two
outdoor environments (UMi and UMa) under two different
frequencies (24 GHz and 60 GHz). Therefore, this paper un-
ravels the spatiotemporal nature of mm-wave channels. Indeed,
it provides insightful results for more accurate system design
and evaluation under diverse circumstances and scenarios
that would direct the development of next-generation mobile
communication with enhanced accuracy and acceptability.
II. MEASUREMENT CAMPAIGN
A. Hardware Specification of the Channel Sounder
The channel sounder used in the measurement consists of
a baseband circuit and an RF circuit [16]. The local oscillator
in the transmitter (Tx)/receiver (Rx) circuit generates carrier
signals at 24.15 GHz and 58.32 GHz. The RF circuit in the
60 GHz band uses a 16-element linear array with a built-in RF
module, and the RF circuit in the 24 GHz band uses a 2 ×8-
element planar array. The equivalent isotropic radiated power
(EIRP), the sum of the output power, beamforming gain, and
antenna element gain, is about 32 dBm in the 24 GHz band
and about 41 dBm in the 60 GHz band. The signal bandwidth
is 200 MHz in the 24 GHz band and 400 MHz in the 60 GHz
band. Therefore, the delay resolution is 5.0ns and 2.5ns,
respectively. The array antenna can scan an angular range from
45to +45by switching beams at high speed. Combining
these four antennas on both the Tx and Rx sides, the AOD
and AOA can be scanned in the 360range, which enables
a simultaneous scan measurement in two frequency bands in
about 5 minutes. The beam’s half-power beamwidth (HPBW)
is about 15in the horizontal plane and about 45in the
vertical plane for both the Tx and Rx sides in the 24 GHz
band. In the case of the 60 GHz band, the horizontal plane is
about 6, the vertical plane is about 45on the Tx side, and
about 18on the Rx side.
B. Measurement Environments
Two different types of environments, namely Urban Micro
(UMi) and Urban Macro (UMa), as shown in Fig. 1, were se-
lected for the channel measurement campaigns and described
below.
Fig. 2: Tx and Rx positions at measurement site.
1) UMi: For propagation channel measurements under the
UMi environment, a street around the shopping complex in
Yokohama City, one of the metro cities in Japan, was selected,
as shown in Fig. 1(a). In this figure, the Tx, indicated by
the yellow marker, is the base station (BS), located near the
intersection at a height of 3.0m, and Rx is the UE, with
an antenna height of 1.5m, marked by the blue line. Large
commercial facilities, trees, traffic signals, signboards, and
scatterers such as automobiles and pedestrians surrounded the
receiving points. A cart was used to carry the receiver point
along the blue line to measure the double-directional channel
impulse responses under non-stationary UE. The measure-
ments were performed for 15 points at 1.0m intervals along
the blue line, keeping this Tx and Rx setup.
2) UMa: Similarly, for measuring the propagation channel
under the UMa environment, the area around Chinatown in
Yokohama City was selected, as shown in Fig. 1(b). Here, to
increase the height of the Tx antenna from the ground, the Tx
is deployed on the roof of a building. Therefore, the height of
the Tx from the ground became 34.0m, whereas the height
of the Rx remained the same at 1.5m, as shown in Fig. 2.
Following the same Rx setup, the receiver terminal was carried
along two different routes, namely, the E-Route and the S-
Route, as shown in Fig. 1(b), and channel measurements were
performed for 15 points at an interval of 1.0m on each path.
III. EXPERIMENTAL PRO CE SS
A. MPC Extraction and Clustering
The data acquired in this measurement is a three-
dimensional impulse response array of the AOA, AOD, and
delay time. The double-directional azimuth delay power spec-
trum (DDADPS) [17] can be obtained by squaring the absolute
value of each element. The Sub-Grid CLEAN algorithm [18],
a high-resolution channel parameter estimation algorithm, is
used to extract the multipath components (MPCs) from the
DDADPS. The K-powerMeans (KPM) clustering algorithm
[19] is used to classify the extracted multipath components as
clusters of similar angle and delay parameters. The appropriate
number of clusters Kis determined by visually inspecting the
KPM results on a panoramic photograph of the measurement
environment and conformed with the physical reflective ob-
jects.
In this study, extracted MPC datasets consisting of the data
acquired in the 24 GHz and 60 GHz bands measurements are
combined and clustered at two frequencies simultaneously. By
IEEE ANTENNAS AND WIRELESS PROPAGATION LETTERS, VOL.XX, NO.XX, SEPTEMBER 2023 3
(a) 24 GHz. (b) 60 GHz. (c) Propagation paths of D clusters.
Fig. 3: Measured PDPs and cluster tracking results in UMi.D-clusters are identified by comparing with RT.
(a) 24 GHz. (b) 60 GHz. (c) Propagation paths of D clusters.
Fig. 4: Measured PDPs and cluster tracking results in UMa E-Route. D-clusters are identified by comparing with RT.
(a) 24 GHz. (b) 60 GHz. (c) Propagation paths of D clusters.
Fig. 5: Measured PDPs and cluster tracking results in UMa S-Route. D-clusters are identified by comparing with RT.
doing this, the clusters obtained at two different frequencies
can be associated with each other. Here, the path losses caused
by the different frequencies are compensated [20]. In this
case, based on the free-space path loss formula, 7.66 [dB]
(difference from 24 GHz) is added to/subtracted from the path
gain of each MPC at 60 GHz before/after the clustering.
B. Visible Region
The 3GPP SCM is widely acknowledged as a standard 5G
channel model. According to the authors in [9], the MPCs
should be changed with the UE position and considered as
a time-varying function of UE’s position. In connection with
this, authors in [9] have proposed a way to achieve realistic SC
property, where the SC is imposed by the correlation distance
of LSPs. However, since the SC may differ from the used LSP,
a cluster visible region, which is obtained by tracking clusters
from birth to extinction, obviously offers a more practical and
accurate SC [13], [15].
The visible region is calculated as described as follows.
The clusters at a receiver point are connected to the cluster
with the smallest distance between their centroids among the
clusters at the next Rx point. The cluster centroid at a Rx point
kis expressed as xjk={ϕT,jk, ϕR,jk, τjk,PGjk}, where
jk= 1, . . . , Mk,Mkbearing the total number of clusters and
ϕT,jk,ϕR,jk,τjkand PGjkdenote the radiation angle/arrival
angles (azimuth), delay time, and cluster power, respectively.
The distance between clusters in the four-dimensional param-
eter space is calculated as
cjk,jk+1 =
4
X
l=1
xl
jk,jk+1 minjk,jk+1
xl
jk,jk+1
maxjk,jk+1
xl
jk,jk+1 minjk,jk+1
xl
jk,jk+1
(1)
where xl
jk,jk+1 =|xl
jkxl
jk+1 |denotes the distance between
parameters of dimension l. At the same time, if the distance
obtained between clusters is more than 15 ns in delay time
and the departure and arrival angles are more than 20apart,
then the cluster is excluded as an outlier. Moreover, in such
scenarios where a cluster disappears temporarily, and there
is no cluster present between two adjacent UE positions that
satisfy the above-mentioned conditions, then to calculate the
IEEE ANTENNAS AND WIRELESS PROPAGATION LETTERS, VOL.XX, NO.XX, SEPTEMBER 2023 4
0 5 10
Visible Region [m]
0
0.2
0.4
0.6
0.8
1
CDF
24GHz
60GHz
(a) Route followed in UMi.
0 5 10
Visible Region [m]
0
0.2
0.4
0.6
0.8
1
CDF
24GHz
60GHz
(b) Route followed in UMa (E-Route).
0 5 10
Visible Region [m]
0
0.2
0.4
0.6
0.8
1
CDF
24GHz
60GHz
(c) Route followed in UMa (S-Route).
Fig. 6: Visible region for random clusters in each scenario.
TABLE I: Visible region of random clusters.
Urban Micro (UMi) E-Route (UMa) S-Route (UMa)
24 GHz 60 GHz 24 GHz 60 GHz 24 GHz 60 GHz
Number of random clusters 29 25 10 8 12 10
Visible region (mean) [m] 5.41 4.16 4.23 3.23 5.58 4.58
visible region, the next adjacent UE position is taken into
consideration for the same and follows the same procedure.
IV. RES ULT S AN D ANALYSIS
Figs. 3–5 (a) and (b) exhibit the location characteristics of
the power delay profile (PDP) for each receiving point for each
scenario. These figures exhibit the resultant clusters and the
associated cluster centroids plotted with the same color and
marker. Only clusters with a cluster visible region of 3m or
greater are plotted here for each frequency band.
In each figure, the propagation paths are clearly identifiable.
The clusters are assumed to be deterministic if these can be
represented appropriately by RT; otherwise, they are treated
as random clusters. Reflecting objects were identified using
the clustered MPCs and panoramic photographs taken at
each measurement point during channel measurement, and the
propagation path was determined by comparing them with the
RT results. White lines surround the deterministic clusters (D-
clusters), direct waves, or wall reflections in each figure of
Figs. 3–5. The tiny red circles around 200,500, and 400 ns
in Figs. 3–5 indicate direct waves, which are observed at
all receiver points. In all cases, the delay time increases
monotonically indicating the Rx moves away from Tx. The
single wall reflection in Fig. 5 and the double wall reflection
in Fig. 4 show that the D-clusters exist at both frequencies.
In addition, Figs. 3–5 (c) shows the propagation path of
the D-cluster identified by comparing RT and measurement
results. On the contrary, in Figs. 3–5, all the clusters except
those surrounded in white are random clusters that cannot be
represented by the RT.
The detailed observation revealed that the dominant scat-
terers are interactions with small objects such as signboards,
billboards, and things that cannot be represented in the 3D
model. A large path loss is observed when the Fresnel zone
of a propagation path is partially blocked by small objects such
as tree leaves and thin poles on the street [20]. This, thereby,
reduces the visible regions and the number of associated
clusters. The 60 GHz band is prone to significant propagation
loss due to the smaller Fresnel zone radius.
Fig. 6 shows the CDF representations of the visible regions
of random clusters where the blue and red lines in the figures
represent the 24 GHz and 60 GHz band, respectively. The
comparison of the CDF of the two different frequency bands
shows that the visible region in the 24 GHz band is more
significant than in the 60 GHz band for all scenarios. It can
also be observed that some clusters are only present in the case
of the 24 GHz band, which indicates the frequency dependence
of visible clusters. The mean values of the visible regions and
the number of clusters for each scenario are tabulated in Table
I. Comparing the results for the two different frequency bands
shows that the mean values of the visible region differ by
more than 1m. As shown in Table I, the number of clusters
in the 24 GHz band is a little more than in the 60 GHz band,
which is prominently due to the fact that the 60 GHz band
suffers more for its shorter wavelength and is affected more
by the blockage and other attenuation factors present in the
environment. Moreover, the presence of small objects, such as
signs and billboards, that are relatively flat and easily reflective
in the vicinity of the receiver following the S-Route results in
fewer clusters, most of which are of extended visible regions.
V. CONCLUSION
This letter provides a detailed insight into the SC of clusters
for non-stationary UE under mm-wave propagation channels
in an urban environment. To characterize the SC, the cluster
visible region was obtained by tracking the clusters obtained
from the measurement data. The investigation results show
that the number of random clusters with a visible region of
3m or more is greater in UMi than in UMa, establishing
the environmental dependencies of SC. Another significant
observation is that irrespective of the environment, the visible
region is more (1m) at 24 GHz than at 60 GHz, whereas
the number of clusters is more at 24 GHz, which implies the
frequency dependencies of SC.
IEEE ANTENNAS AND WIRELESS PROPAGATION LETTERS, VOL.XX, NO.XX, SEPTEMBER 2023 5
REFERENCES
[1] S. Ju and T. S. Rappaport, “Simulating Motion - Incorporating Spatial
Consistency into NYUSIM Channel Model,” in Proc. 2018 IEEE 88th
Vehicular Technology Conference (VTC-Fall), 2018, pp. 1–6.
[2] International Telecommunication Union, Technical, “ITU-R M.2412-0.
Guidelines for evaluation of radio interface technologies for IMT-2020,”
released: Oct. 2017.
[3] S. Rangan, T. S. Rappaport, and E. Erkip, “Millimeter-Wave Cellular
Wireless Networks: Potentials and Challenges, Proc. IEEE, vol. 102,
no. 3, pp. 366–385, 2014.
[4] “Study on channel model for frequencies from 0.5 to 100 GHz.
(3GPP TR 38.901 version 16.1.0 Release 16),” (Released: Nov. 2020).
[Online]. Available: https://www.etsi.org/deliver/etsi tr/138900 138999/
138901/16.01.00 60
[5] S. Dai, N. Abdellatif, M. Kurras, S. Jaeckel, and L. Thiele, “Spatial
Consistency Validation on Massive SIMO Covariance Matrices in the
Geometry-Based Stochastic Channel Model QuaDRiGa,” in Proc. WSA
2020; 24th International ITG Workshop on Smart Antennas, 2020, pp.
1–6.
[6] T. S. Rappaport, Y. Xing, G. R. MacCartney, A. F. Molisch, E. Mellios,
and J. Zhang, “Overview of Millimeter Wave Communications for Fifth-
Generation (5G) Wireless Networks—With a Focus on Propagation
Models,” IEEE Trans. Antennas Propag., vol. 65, no. 12, pp. 6213–
6230, 2017.
[7] S. Ju and T. S. Rappaport, “Simulating Motion - Incorporating Spatial
Consistency into NYUSIM Channel Model,” in Proc. 2018 IEEE 88th
Vehicular Technology Conference (VTC-Fall), 2018, pp. 1–6.
[8] L. Bai, Z. Huang, and X. Cheng, “A Non-Stationary Model With Time-
Space Consistency for 6G Massive MIMO mmWave UAV Channels,
IEEE Trans. Wirel. Commun., vol. 22, no. 3, pp. 2048–2064, 2023.
[9] Y. Wang, Z. Shi, L. Huang, Z. Yu and C. Cao, “An extension of
spatial channel model with spatial consistency, in Proc. 2016 IEEE 84th
Vehicular Technology Conference (VTC-Fall), Montreal, QC, Canada,
Mar. 2016.
[10] G. Yu, L. Tian, J. Zhang, and C. Wang, “The method to implement
5g channel model with spatial consistency, in Proc. 2018 IEEE/CIC
International Conference on Communications in China (ICCC), 2018,
pp. 736–740.
[11] M. Shafi, J. Zhang, H. Tataria, A. F. Molisch, S. Sun, T. S. Rappaport,
F. Tufvesson, S. Wu, and K. Kitao, “Microwave vs. Millimeter-Wave
Propagation Channels: Key Differences and Impact on 5G Cellular
Systems,” IEEE Commun. Mag., vol. 56, no. 12, pp. 14–20, 2018.
[12] K. Zaman and M. M. Mowla, “A Millimeter Wave Channel Modeling
with Spatial Consistency in 5G Systems,” in Proc. 2020 IEEE Region
10 Symposium (TENSYMP), 2020, pp. 1584–1587.
[13] W. Sloane, M. Shafi, C. Gentile, G. Woodward, P. Martin, P. Tang,
J. Zhang, and C. Lai, “Analysing the 3GPP Spatial Consistency Pro-
cedure Through Channel Measurements,” in Proc. 2021 IEEE 32nd
Annual International Symposium on Personal, Indoor and Mobile Radio
Communications (PIMRC), 2021, pp. 860–865.
[14] A. Maltsev, A. Pudeyev, I. Karls, I. Bolotin, G. Morozov, R. Weiler,
M. Peter, and W. Keusgen, “Quasi-deterministic approach to mmWave
channel modeling in a non-stationary environment, in Proc. 2014 IEEE
Globecom Workshops (GC Wkshps), 2014, pp. 966–971.
[15] C. Lai, R. Sun, C. Gentile, P. B. Papazian, J. Wang, and J. Senic,
“Methodology for multipath-component tracking in millimeter-wave
channel modeling,” IEEE Trans. Antennas Propag., vol. 67, no. 3, p.
1826–1836, Mar. 2019.
[16] M. Kim, H. Tsukada, K. Kumakura, R. Takahashi, N. Suzuki,
H. Sawada, and T. Matsumura, “A 24/60-GHz Dual-Band Double-
Directional Channel Sounder Using COTS Phased Arrays, in Proc.
2022 IEEE International Conference on Communications (ICC), Seoul,
Korea, May 2022.
[17] M. Kim, S. Tang, and K. Kumakura, “Fast Double-Directional Full
Azimuth Sweep Channel Sounder Using Low-Cost COTS Beamforming
RF Transceivers, IEEE Access, vol. 9, pp. 80 288–80 299, 2021.
[18] M. Kim, T. Iwata, S. Sasaki, and J. ichi Takada, “Millimeter-Wave Radio
Channel Characterization Using Multi-Dimensional Sub-Grid CLEAN
Algorithm,” IEICE Trans. Commun., vol. E103.B, no. 7, pp. 767–779,
2020.
[19] N. Czink, P. Cera, J. Salo, E. Bonek, J.-p. Nuutinen, and J. Ylitalo,
“A Framework for Automatic Clustering of Parametric MIMO Channel
Data Including Path Powers, in Proc. 2006 IEEE Vehicular Technology
Conference, 2006, pp. 1–5.
[20] M. Kim, N. Suzuki, H. Tsukada, and R. Takahashi, “Frequency De-
pendence of Millimeter-Wave Urban Macrocell Multipath Cluster Chan-
nels,” in Proc. 2023 European Conference on Antennas and Propagation
(EuCAP), Florence, Italy, Mar. 2023.
... To date, numerous research groups and organizations worldwide have undertaken campaigns to measure radio characteristics across a wide range of mm-wave bands and in different scenarios. For example, extensive propagation measurements in both outdoor and indoor urban environments at various frequencies, including 28 GHz, 38 GHz, 60 GHz, and 73 GHz, were conducted [1]- [3], and mm-wave propagation channel characteristics in urban scenarios at 28 GHz were investigated [4], [5]. Many other groups [6]- [8] are similarly involved in measurement campaigns investigating mm-wave propagation channel characteristics in various scenarios and channel parameters for new wireless communication technologies such as beamforming, massive MIMO, etc.; these efforts contribute to developing a suitable channel model for 5G or beyond wireless communication systems [9]- [13]. ...
... In this framework, deterministic paths generated by ray tracing results calibrated by using the above-mentioned proposed method and random paths generated by measured site-specific statistical parameters of LSPs and SSPs are combined to produce accurate channel responses. The spatial consistency procedure in the 3GPP model is also applied using the measured sitespecific correlation distances [38]. Furthermore, an inhouse channel model simulator, called CPSQDSIM, was developed for grid-wise channel data (so-called Path-GridData) generation. ...
... Advantageously, the deterministic clusters obtained by ray tracing are inherently spatially consistent because the evolution of angles and delays are calculated based on the geometry of the environment. Further, the channel model also generates spatially consistent random clusters following the 3GPP spatial consistency procedure with measured site-specific correlation distance [38]. ...
Article
Full-text available
To design a reliable communication system leveraging millimeter-wave (mm-wave) technology-gaining popularity for its capacity to deliver multi-gigabit-per-second data rates-it is essential to consider the site-specific nature of mm-wave propagation. Conventional site-general stochastic channel models often fail to reproduce channel responses in specific usage scenarios or environments accurately. To enable high-precision channel simulation that captures site-specific characteristics, this paper proposes a channel modeling framework leveraging the widely accepted 3GPP map-based hybrid channel modeling approach and provides a detailed methodology for applying it to real-world scenarios with practical examples. First, an extensive measurement campaign was conducted in typical urban macro and micro cellular environments using an in-house dual-band (24/60 GHz) double-directional channel sounder. The mm-wave channel behavior was then characterized, with a focus on differences between the two frequencies. Following this, site-specific large-scale and small-scale channel properties were parameterized. As a critical component for enhancing prediction accuracy, this paper introduces an exponential decay model for the power-delay characteristics of non-line-of-sight clusters, which are often significantly overestimated by deterministic prediction tools. Finally, using the in-house channel model simulator (CPSQDSIM) developed for grid-wise channel data (PathGridData) generation, a marked improvement in prediction accuracy over the existing 3GPP map-based channel model was demonstrated.
... In this framework, deterministic paths obtained through ray tracing with calibration by the developed exponential decay model and measured site-specific statistical parameters of LSPs and SSPs are combined to generate accurate channel responses. The spatial consistency procedure in the 3GPP model is also applied using the measured site-specific correlation distances [44]. Furthermore, an in-house channel model simulator, called CPSQDSIM, was developed for grid-wise channel data (so-called PathGridData) generation. ...
... Physical space Cyber space advantageous that the deterministic clusters obtained by ray tracing are inherently spatially consistent because the evolution of angles and delays are calculated based on the geometry of the environment. Further, the channel model also generates spatially consistent random clusters following the 3GPP spatial consistency procedure with measured site-specific correlation distance [44]. ...
... The BS antenna was mounted at 3.0 meters on the sidewalk, and the MS at each marked point had an antenna height of 1.5 meters. In Fig. 3(b), the measurement points from WP9 to WP23 are closely located for spatial consistency measurement [44]. The distances between the BS and MS antennas ranged from 27 to 105 m. ...
Preprint
Full-text available
To design a reliable communication system utilizing millimeter-wave (mm-wave) technology, which is gaining popularity due to its ability to deliver multi-gigabit-per-second data rate, it's essential to consider the site-specific nature of the mm-wave propagation. Conventional site-general stochastic channel models are often unsatisfactory for accurately reproducing the channel responses under specific usage scenarios or environments. For high-precision channel simulation that reflects site-specific characteristics, this paper proposes a channel model framework leveraging a widely accepted 3GPP map-based hybrid channel modeling approach, and it provides a detailed recipe to apply it to an actual scenario using some examples. First, an extensive measurement campaign was conducted in typical urban macro and micro cellular environments using an in-house dual-band (24/60 GHz) double-directional channel sounder. Subsequently, the mm-wave channel behavior was characterized, focusing on the difference between the two frequencies. Then, the site-specific large-scale and small-scale channel properties were parameterized. As an essential component for improving prediction accuracy, this paper proposes an exponential decay model for power delay characteristics of non-line-of-sight clusters, of which powers are significantly overestimated by deterministic prediction tools. Finally, using the in-house channel model simulator (CPSQDSIM) developed for grid-wise channel data (PathGridData) generation, a significant improvement in prediction accuracy compared with the existing 3GPP map-based channel model was demonstrated.
Conference Paper
Full-text available
The use of millimeter-wave (mm-wave) bands has been expanding in new wireless communication systems such as fifth-generation (5G) mobile systems and WiGig WLANs. Since mm-waves have quasi-optical propagation characteristics, it is necessary to develop a channel model that can cover site-specific characteristics of various environments. For this reason, a large amount of channel measurement data measured in a wide range of frequency bands covering mm-wave in various environments should still be required. In particular, it is necessary to obtain the angular channel characteristics on both sides of the transmitter and receiver, but this requires considerable time and cost. The authors' previous study developed a 60-GHz double-directional (D-D) channel sounder using commercial-off-the-shelf (COTS) phased-array beamforming RFs, and it has currently been extended to realize 24/60-GHz dual-band simultaneous measurement. In this paper, we detailed the design of the dual-band D-D channel sounder. As a demonstration of the operation, some measurement results taken in typical urban macro-cell (UMa)/micro-cell (UMi) scenarios were quickly reported.
Article
Full-text available
This paper presents a multipath component (MPC) parameter estimator that uses a double-directional (D-D) channel sounding system in WiGig millimeter-wave bands. It employs low-cost commercial-off-the-shelf (COTS) RF transceivers capable of beam steering within an azimuth angle range of ±45°; the transmit power is approximately 31 dBm, equivalent isotropic radiated power (EIRP), including a total antenna gain of approximately 19 dBi. The half-power beam widths in the azimuth and elevation planes are 6° and 45°, respectively. A fast D-D channel acquisition in 360° full azimuth range can be achieved by developing a 4×4 multiple-input-multiple-output (MIMO) time division-multiplexing (TDM) scheme that enables simultaneous use of four antenna arrays at both sides of the transmitter and receiver. The beam switching for all combinations (48×48=2304) with 12 beams at each array requires less than 200 ms, excluding optional procedure related to data storage. A high-resolution MPC extraction method was developed from the D-D angular delay power spectra obtained using the sub-grid CLEAN algorithm, which is applicable when phase coherence among multiple snapshots obtained via the beam-switching measurement cannot be guaranteed. Furthermore, the operation of the developed system was validated via a measurement at 58.32 GHz in an office floor of a university laboratory. From the data analysis, the cluster properties and the corresponding scattering processes were identified.
Article
Full-text available
In radio channel measurements and modeling, directional scanning via highly directive antennas is the most popular method to obtain angular channel characteristics to develop and evaluate advanced wireless systems for high frequency band use. However, it is often insufficient for ray-/cluster-level characterizations because the angular resolution of the measured data is limited by the angular sampling interval over a given scanning angle range and antenna half power beamwidth. This study proposes the sub-grid CLEAN algorithm, a novel technique for high-resolution multipath component (MPC) extraction from the multi-dimensional power image, so called double-directional angular delay power spectrum. This technique can successfully extract the MPCs by using the multi-dimensional power image. Simulation and measurements showed that the proposed technique could extract MPCs for ray-/cluster-level characterizations and channel modeling. Further, applying the proposed method to the data captured at 58.5 GHz in an atrium entrance hall environment which is an indoor hotspot access scenario in the fifth generation mobile system, the multipath clusters and corresponding scattering processes were identified.
Article
In this paper, a novel unmanned aerial vehicle (UAV) space-time-frequency (S-T-F) non-stationary channel model with time-space consistency for sixth generation (6G) massive multiple-input multiple-output (MIMO) millimeter wave (mmWave) wireless communication systems is proposed. In the proposed model, the line-of-sight (LoS) transmission and non-LoS (NLoS) transmission through ground reflection, single-clusters, and twin-clusters are modeled. Meanwhile, the three-dimensional (3D) continuously arbitrary trajectory and the self-rotation of UAV are imitated. To capture the time-space consistency and S-T-F non-stationarity simultaneously, a new UAV-related non-stationary modeling algorithm that integrates the visibility region (VR), frequency-dependent path gain, and survival probability is developed for the first time. In this algorithm, the UAV-related parameters are considered, including UAV’s height, 3D moving velocity, and self-rotation angles. In the proposed model, the calculation of channel impulse response (CIR) is developed, which considers the cluster density index influenced by communication scenarios, frequency, UAV’s height, and the distance between transceivers. Some important channel statistical properties, such as S-T-F correlation function (STF-CF), Doppler power spectral density (DPSD), and stationary interval, are derived. Finally, simulation results match well with ray-tracing-based results, which verifies the utility of the proposed model.
Conference Paper
The enormous demand of wireless traffic is leading researchers to utilize the huge bandwidth of millimeter wave (mmWave) bands for higher data transmission and larger capacity channel modeling for new wireless standard 5G. However, mmWave channel modeling is still a challenge as it is highly weather sensitive and oppressive for short range line-of-sight (LOS) requirement. In this paper, the characteristics of spatial channel modeling is simulated for 73 GHz mmWave bands using NYUSIM. Spatial consistency channel models for moving users and channel models for static users without spatial consistency consideration have been compared in terms of different channel parameters for both LOS and non-LOS (NLOS) environment. Simulation results indicates that the received power is higher for LOS and NLOS considering spatial consistency channel model compared to the channel models without spatial consistency in urban micro cell scenario. In addition, several power delay profiles and dynamic channel characteristics are presented considering both spatial consistency and without spatial consistency.
Article
We describe an extensive channel-measurement campaign, including 325 unique transmitter-receiver configurations, conducted in a lecture room with our 3D double-directional 60-GHz channel sounder. The receiver was mounted on a mobile robot, with 40-cm spacing between channel acquisitions, enabling the tracking of clustered multipath components in the multi-dimensional delay-angle space. To mitigate against angle-estimation error and multipath blockage, we introduce a robust tracking algorithm based on the Assignment Problem. For the purpose of validation, the clusters were transformed from the delay-angle space onto a 2D map of the environment and compared against the locations of cluster-generating reflectors, such as walls and tables. The location errors were typically within 30-50 cm. The clusters identified were then reduced to a stochastic map-based channel model, including reflection loss and dispersion characteristics such as the Ricean K-factor and angular spread. Given the 0.5-ns delay resolution of the channel sounder and angle-estimation error around 2°, the parameters were reported with high fidelity.