Thinking outside the sphere

Weston, Massachusetts Wireless

Drive Test Results,
November 3, 2018

Introduction
The Town of Weston, Massachusetts engaged the services of Isotrope, LLC to gather information
on the availability of personal wireless services, familiarly known as cellular services, throughout
the town. Weston is a predominantly residential community dominated by single family
residential use with some institutional use. The town has limited area dedicated to commercial
and business districts. It is heavily wooded and hilly. It is challenging to provide wireless service
to areas that consist of wooded and hilly terrain with low density of development.

Wireless coverage in Weston is obtained from cell sites approved through zoning permits. The
map below shows the locations of constructed cell sites. Some cell sites host the antennas of more
than one carrier, but several are single-carrier sites.

Figure 1 – Weston Cell Site Map

www.isotrope.im Isotrope, LLC ◦ 503 Main Street ◦ Medfield, MA ◦ 02052 508 359 8833
 Isotrope, LLC
Drive Test Route
To map the existing coverage from each of the wireless carriers, the town and Isotrope developed
a proposed route to cover most of the public ways and accessible private ways, as well as certain
passable fire trails. On October 1st, Isotrope drove the fire trails. On October 25th and 26th, Isotrope
drove the streets. Foliage was present for both surveys. Foliage reduces wireless signal strength,
so it is important to capture the worst case by making surveys with foliage present.

Wireless Environment
The wireless services operate on numerous licensed frequency bands in the radio spectrum, from
700 MHz at the low end to 2300 MHz at the high end. Lower frequencies are less affected by
foliage. Wireless carriers program their networks to manage which of their frequency bands each
user device is connected to, based on the quality of service available at the user’s location.
Typically, the higher frequencies are used closer to the cell sites, and users who move out to the
fringes of a cell site’s coverage may be transferred to a lower frequency band to maintain service.
When better coverage from an adjacent cell site becomes available, the user’s device is “handed
off” to the next cell site.

There is also the possibility of handing a user device over to an older technology. Specifically, the
current technology of choice is fourth generation (“4G”) service provided through the LTE
technology. With a typical turnover rate of about 2 years per user device, most devices in the
market today are equipped to use 4G LTE service. Most cell sites in metropolitan Boston have
been upgraded to provide LTE service. In the event of high volumes of users demanding a
connection, a user might be handed over to 3G service. The coverage areas of 3G and 4G services
are similar because the available frequency bands are the same. For example AT&T uses 700 MHz
and 2100 MHz to provide LTE service and continues to serve 3G users with its 850 MHz band.
Occasionally, the 3G signal available to a user may be better than the 4G signal, and a user device
may be handed over to 3G service, but this is not a general rule.

Test Method
Isotrope’s test rig uses recent models of smart phones subscribed to each carrier. The phones are
mounted inside the windshield to minimize signal attenuation caused by the vehicle. This is a
customary method used by the industry. It bears the caveat that general computer prediction of
coverage assumes the device is fully outdoors, with adjustments to the displayed signal level
thresholds to account for vehicle penetration and building penetration. Hence, most of the time
the windshield-mounted smart phones (known as “UE,” or user equipment, in the trade) are
seeing the equivalent of an outdoor signal level because they are against the windshield (the glass
has negligible impact on signal strength). When the test vehicle is oriented in certain ways
(generally when driving directly away from a cell site) there may be some reduction in signal

2
www.isotrope.im
 Isotrope, LLC
strength measured by the UE, compared to an unobstructed signal outdoors. This does not
materially affect the results because the goal is to obtain a general picture of the signal levels and
gradients from good to poor throughout the town.

The test rig is a ZK Celltest unit that connects to the smartphones and logs the signal
characteristics. The UE is allowed to be handed over to the radio band that the network prefers.
This means that the UE might be on an upper frequency band because it is relatively close to the
cell site and may not be receiving as strong a signal level as it would if it were connecting via a
lower frequency band at the same spot. The network can decide that the signal level of the upper
frequency band is good enough for the purpose, and assign the UE to the upper frequency band
to keep the lower frequency band available to UEs closer to the cell site’s coverage fringe. Because
of this, the signal levels reported are not necessarily the highest levels available, but the levels
that the network determined were the best trade-off between quality and network loading. This
does not materially affect the results for the purposes of this study.

Four UEs were employed, covering the AT&T, Sprint, T-Mobile and Verizon networks. There
are some variations on how the UEs interact with the ZK Celltest logging unit. The Verizon and
Sprint UEs did occasionally switch to 3G service. We noted that the AT&T and T-Mobile UEs did
not show any use of the 3G network. This may have been a result of the way these networks, the
UEs, and the logging unit interact. AT&T and T-Mobile networks evolved from the GSM
technology while the Sprint and Verizon networks evolved from the CDMA technology.
Consequently, we cannot be certain whether the lack of any 3G data on AT&T or T-Mobile
UEs is the result of a network decision not to employ 3G during the survey, the lack of any 3G
service that was better than the 4G service, or the inability of the test system to log 3G data. It is
apparent, based on the data collected on all four services, that this difference between the
AT&T/T-Mobile data sets and the Sprint/Verizon data sets is immaterial.

Results
All four UEs provided reliable data during the road survey. On the fire trails, the data collection
apparatus was mounted on a town vehicle. The Sprint and T-Mobile UEs did not properly connect
to their respective networks. The collected fire-trail data for these two units is not reflective of
their networks’ coverage. We have eliminated these from the dataset.

Data is provided in two formats. Shapefile data is provided for the town GIS professionals to work
with. Google Earth kml/kmz data is provided for quick loading and interactive viewing. This
document contains snapshots of each dataset overlaid on a map of the town.

The attached images show where the 4G LTE signals are considered to be among several
classifications based on received signal strength. Each carrier has slightly different values for its

3
www.isotrope.im
 Isotrope, LLC
classifications, but for consistency, we provide a generalized set of thresholds. The way LTE signal
strength is measured is by obtaining the value of a signal component, not the entire signal. That
component is the “Reference Signal.” The measurement is Reference Signal Received Power
(“RSRP”). Because of this, the signal levels reported for LTE coverage cannot be directly compared
with signal levels for 3G coverage.

The LTE thresholds we chose to depict on the attached maps are:

Color Value Description

Green ≥-85 dBm (LTE RSRP)Excellent service. Also expected to be reliable in larger
non-residential buildings.
Yellow ≥-95 dBm Typical network design target to penetrate residential
buildings
Light blue ≥-115 dBm Typical threshold for outdoor signal coverage
Black <-115 dBm UE could still work at levels as low as -120 dBm or less,
but even outdoors this level is usually unreliable.
Dark blue 3G Indicates UE handed over to 3G service. To avoid
confusion, no 3G signal level is presented on this map.
Orange Road driven but no Indicates UE was not able to lock onto a cell site.
signal reported
Table 1 - Information Shown on Included Maps

4
www.isotrope.im