0% found this document useful (0 votes)
53 views2 pages

Radio Coverage Mapping For Cellular Networks in A High-Rise Urban Area Using Tuned 9999 Model

The document discusses tuning the 9999 radio propagation model using measurements from a cellular network in a high-rise urban area of Manila. The model parameters were tuned using a measurement campaign and both Cellular Expert software and a MATLAB program to minimize the error between predictions and measurements. The tuned model better predicted signal strengths with lower error compared to the default model.
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
53 views2 pages

Radio Coverage Mapping For Cellular Networks in A High-Rise Urban Area Using Tuned 9999 Model

The document discusses tuning the 9999 radio propagation model using measurements from a cellular network in a high-rise urban area of Manila. The model parameters were tuned using a measurement campaign and both Cellular Expert software and a MATLAB program to minimize the error between predictions and measurements. The tuned model better predicted signal strengths with lower error compared to the default model.
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

Radio Coverage Mapping for Cellular Networks in a

High-rise Urban Area using Tuned 9999 Model


Korinne Ella R. Morico*, Julius M. Judan, and Calvin Artemies G. Hilario
Department of Science and Technology - Advanced Science and Technology Institute, Quezon City, Philippines
*email: [email protected]

Abstract – This research aims to provide radio coverage 𝑃𝐿 = 𝑎0 + 𝑎1 log(𝑑) + 𝑎2 log(ℎ𝐵 )


mapping technique using an accurate propagation model and
+𝑎3 log(ℎ𝐵 ) log(𝑑) − 3.2(log(11.75 ℎ𝑀 ))2 (1)
high-resolution digital terrain map. Radio propagation
simulations were done using the 9999 model which provides +44.49 log(𝑓) − 4.78(log(𝑓))2
steering parameters convenient for calibration. Measurement
campaign was conducted in a high-rise urban area in Metro III. MEASUREMENT CAMPAIGN
Manila. Lee Method was used on the actual measured data to Measurement campaign was conducted in a high-rise urban
smooth out short-term fading, but not eliminate the effects of long-
term fading. Parameter tuning of the 9999 model was performed area in Taguig City, Metro Manila. Received signal power were
such that the simulation fits the received signal power obtained measured using a portable spectrum analyzer, with a multi-
from the measurement campaign. polarized antenna of 3 dBi gain. Measurements were recorded
Keywords — 9999 Ericsson model, radio propagation, on a smartphone, as well as its corresponding GPS coordinates,
measurement campaign as shown in Fig. 1. Receiver bandwidth, frequency interval,
I. INTRODUCTION vehicle speed, and data averaging were based on the ITU
Recommendation SM.1708-1 [3]. Averaging of measured data
With the drastic increase of the number of mobile users was done using Lee Method to smooth out short-term fading,
comes the need to increase the number of cell towers to but not eliminate the effects of terrain variation.
accommodate new users. In the Philippines, there is only 0.14
towers for a thousand users, when the ideal ratio is one tower
for a thousand users [1]. On top of that, cell towers are mostly
crowded in the cities in Metro Manila and Luzon, and a few
places in other parts of the country. This research aims to
provide radio coverage mapping technique using an accurate
propagation model and high-resolution digital terrain map. This
may be used in planning the locations where new cell towers
will be built. Fig. 1. Summarized flowchart of the model tuning process
II. RADIO PROPAGATION SIMULATIONS IV. PARAMETER TUNING OF THE 9999 MODEL
Radio propagation simulations were conducted on Cellular The Cellular Expert has a built-in model tuning function,
Expert using a 5-meter resolution digital terrain elevation which requires the predicted field strength using the 9999
model (DTM). The propagation model used was the 9999 model and the measured data with its coordinates. Tuning of
model, or the Ericsson model. The 9999 model is an extended the 9999 model was done by minimizing the error between the
version of the Hata model which provides steering parameters prediction and the measured data. Only the parameters 𝑎0 and
depending on the environment, making it very convenient for 𝑎1 from Eq. (1) were tuned, since the measurements obtained
calibration [2]. Parameter 𝑎0 is the constant offset which are taken at a roughly constant height relative to the base station
regulates the absolute level of the loss curve. Parameter 𝑎1 is [5]. To verify the results of the model tuning on Cellular Expert,
the distance influence coefficient which regulates the slope of a model tuning program was done on MATLAB. The program
the curve. Parameters 𝑎2 and 𝑎3 are the transmitter height solves for the tuned values for 𝑎0 and 𝑎1 , which minimizes the
influence coefficient and the Okumura-Hata type of mean-squared error between the prediction of the tuned 9999
multiplying factor, respectively. The 9999 model also takes model and the measured data [5]. Field strength maps were
into account the distance 𝑑 of the point with respect to the generated using the tuned 9999 models from both. Results were
transmitter, height of base transmitter ℎ𝐵 , height of mobile compared with the actual measured data using the mean error
antenna ℎ𝑀 , and the frequency 𝑓 . This will be used in the obtained from the error calculation feature on Cellular Expert.
equation for path loss, measured in dB, as in For the rest of this paper, the term tuned model would also mean
tuned 9999 model, or the 9999 model with tuned parameters. Cellular Expert includes the DTM, while the calculation on
The term 9999 model would refer to the 9999 model with the MATLAB does not, hence the discrepancy between the results.
default values for the steering parameters.
V. RESULTS AND ANALYSIS
Field strength map was generated on Cellular Expert using
the 5m resolution DTM, and the 9999 model with the default
values for 𝑎0 and 𝑎1 , shown in Table I. Raw measurements,
from the measurement campaign in the high-rise urban area,
and the averaged data using Lee Method is shown in Fig. 2. The
plot shows that the averaged data has reduced the fluctuations
caused by short-term fading, but still captured the detailed
signal changes due to long-term fading.
Fig. 4. Comparison of
Simulated field strength using the 9999 model, using the
tuned model from Cellular Expert, and using the tuned model
from MATLAB were individually compared with the actual
measured data. The calculated values for the error, expressed
in dB, are presented on Table II.
TABLE II ERROR CALCULATION
Measure 9999 model Cellular Expert MATLAB
Mean Error 29.04 6.05 1.46
Std. Dev. 7.54 5.31 5.61
The mean error using the tuned model from MATLAB is
Fig. 2. Plot of raw measurements and averaged data lesser than the one from Cellular Expert. This coincides with
Using the model tuning function on Cellular Expert and the the plot on Fig. 4, which shows that the plot of the tuned model
program on MATLAB, the new values for 𝑎0 and 𝑎1 are from MATLAB is closer to the plot of the measured data.
presented on Table I. The default values for 𝑎2 and 𝑎3 , which VI. CONCLUSION
are −12.0 and 0.1, respectively, are kept as is.
This paper presents a tuning process for the 9999 model
TABLE I STEERING PARAMETERS
Parameter Default Cellular Expert MATLAB
using simulations and received signal power from actual
𝑎0 36.2 75.88 70.6822 scenario. Results show that the tuned parameters from
𝑎1 30.2 40.05 38.5817 MATLAB and Cellular Expert are roughly around the same
Fig. 3 shows the field strength map of the cell tower using values. Although the tuned model from MATLAB has a lower
the tuned model from Cellular Expert, and the black path mean error, the Cellular Expert’s tuning function may also
represents the coordinates of the measured data from the include additional tuning parameters using the Free Space Loss
measurement campaign. Model for line-of-sight areas, and dominant Knife-Edge
diffraction loss.
REFERENCES
[1] Roxas, P (2019). DICT, telcos identify 1,000 locations for common tower
project — Philippine Inquirer. [online] Inquirer.net. Available at:
https://business.inquirer.net/268291/dict-telcos-identify-1000-locationsfor-
common-tower-project.
[2] A. Saeed, H.U. Rehman, M.H. Masood, “Performance analysis and
comparison of radio frequency propagation models for outdoor environments
in 4G LTE network,” M. S. thesis, Blekinge Institute of Technology, Sweden,
2013.
[3] Cellular Expert Technical Reference, 2015. Available:
https://cellularexpert.freshdesk.com/support/home.
Fig. 3. Field strength map using tuned model [4] International Telecommunication Union, Field-strength measurements
along a route with geographical coordinate registrations, Geneva,
A more visible comparison of the values obtained from the Switzerland, 2011. Available: https://www.itu.int/dms_pubrec/itu-r/rec/sm/R-
field strength map is shown in Fig. 4. Fig. 4 shows the received REC-SM.1708-1-201109-I!!PDF-E.pdf.
power (in dBm), with respect to the distance (in km) from the [5] Benedicic L, Pesko M, Javornik T, Korosec P. A (2014). Metaheuristic
approach for Propagation-Model Tuning in LTE Networks, Informatica, 38,
base transmitter. As mentioned earlier, the simulation on
135–143.

You might also like