Ekspresa Blog Articles

LTE Coverage in Santo Domingo central polygon

Last week we started a cycle of publications of LTE Coverage reports for some of the markets where Ekspresa is present. The first post corresponded to Panama City. This week we will show the results of the measurements in the central polygon of Santo Domingo, comparing the operators Altice, Claro and Viva.

The measurements were collected using the multi-technology solution CellWireless, by Celplan.

We want to highlight that the analysis that we will present in this article is based exclusively on the signal levels (RSRP and CINR) measured on LTE cells. This information, while very valuable, does not provide a complete picture of the quality of the user experience on a mobile network.

To perform a complete benchmarking of the quality of service of a mobile network, additional measurements are required using test terminals (typically smartphones) performing an extensive test protocol for voice and data services. Even though Ekspresa has extensive experience in this type of project, for this series of publications we limited ourselves only to measurements collected with the scanner.

In the next section we present the methodology followed in the study. This includes test conditions, execution dates, and equipment characteristics. Later, we will show an interactive map with the results by operator and channel, as well as a brief analysis of the results.

LTE coverage evaluation methodology

We briefly describe the methodology used for this study.

General considerations

Image of the scanner used to measure LTE Coverage - CellDigitizer
CellDigitizer Scanner

Frequencies

The equipment was configured to collect measurements at the following frequencies:

OperadorBanda# Canal (EARFCN)
Altice1900 MHz700
Altice2100 MHz/1700 MHz (AWS)2100
Altice850 MHz2450
Altice900 MHz 3750
Claro1900 MHz900
Claro2100 MHz/1700 MHz (AWS)2300
Claro2.6 GHz2850
Viva2100 MHz/1700 MHz (AWS)2000
Viva2100 MHz/1700 MHz (AWS)2175
Table 1. Frequencues per operator

We only considered for this study the frequencies used for mobile LTE service. On this occasion, we are not measuring the coverage of residential internet services offered by Claro, Altice and Wind Telecom.

LTE Coverage levels definition

In cellular networks, there are indicators that are used to show the level of signal intensity (received power) and signal quality (signal-to-noise ratio) with which the mobile decodes the cell transmissions.

In LTE, we usually use the RSRP (Reference Signal Received Power) to measure the signal strength, and the RSRQ (Reference Signal Received Quality) and CINR </ strong> (Carrier to Interference and Noise Ratio) to measure the signal-to-noise ratio. The description of the meaning of these metrics is beyond the scope of this article, please review this link if you want to explore these definitions.

In the next analysis we will use the RSRP and the CINR for the evaluation of LTE coverage. In addition to presenting the individual RSRP and CINR values, we define three coverage “levels” or “tiers” that take both metrics into consideration.

Please note that in the field of Radio Frequency analysis the term “coverage” is often used to denote only the level of signal intensity (RSRP), however in this article we will use this word to refer to the combined indicator.

To calculate the coverage level in each bin, the RSRP and CINR samples collected with the scanner are averaged in the hexagonal area (2149m²). After this, the RSRP and CINR of all the frequencies in Table 1 are evaluated in each bin, and the best level of coverage is selected from among all the channels found for each operator. For example, if in a given bin the operator X has -110dBm of RSRP / -10 dB of CINR for EARFCN # 1 (which corresponds to level 1), and has -85dBm RSRP / 25dB CINR (Level 3) for EARFCN # 2, then this bin will be considered level 3.

Results: Who has better LTE Coverage?

In Fig. 1. a map is shown with the comparison of the three operators for the indicator “Coverage Level” described in the methodology. The map is interactive: you can select the metric to evaluate, change the frequency and zoom in or out.

Fig. 1. Coverage levels for the 3 operators. Santo Domingo central polygon

It can be seen that the operator Altice has a larger number of samples with CINR & gt; 10dB. The numerical result can be seen in Fig. 2.

Fig. 2. Comparison of LTE Coverage level for the three operators

Note that Altice has the highest percentage of samples in range 3, followed by Claro and Viva. As will be seen later, Altice has a carrier with CINR levels above the rest, which has an important weight in the indicator.

Let’s examine the RSRP and CINR individually:

RSRP comparison

The following figure shows the RSRP statistics for the three operators, where it can be seen that they all have more than 98% of the samples above -100 dBm.

Fig. 3. RSRP level comparison > -100 dBm

If we raise the threshold to -80dBm, we appreciate that Altice takes the advantage. This makes sense, since Altice has LTE carriers in lower pathloss bands (850MHz and 900MHz).

Fig. 4. RSRP level comparison > -80 dBm

In the following image we see the distribution of samples by carrier. As expected, Altice’s 2450 (850MHz) and 3750 (900MHz) carriers have higher sample densities around -70dBm, which explains the results in Fig. 4.

Fig. 5. RSRP distribution by operator and channel

The ability to provide LTE service on “low band” carriers favors coverage in indoor scenarios. This gives the Altice operator an advantage.

CINR comparison

When we look at the CINR, we also see that Altice is leading the way.

Fig. 6. Comparison of % of samples with CINR> 10 dB

This graph is similar to the one in Fig. 2. This makes sense: since all operators have acceptable RSRP levels (> 100dBm) within the route under study, the “Coverage Level” indicator is modeled by the CINR.

When we examine the data by carrier, we obtain the following density graph:

Fig. 7. CINR distribution by operator and channel

Note that Altice has one particular carrier (EARFCN 3750) with a noticeably better CINR than the others.

It is striking that this carrier is in the 900MHz band (which, together with 850, is more susceptible to interference since they have greater propagation). With the data collected we cannot precisely determine the reason for this difference. Some possible causes can be:

If we exclude Altice’s “atypical” carrier from the analysis, the graph in Fig. 6 looks like this:

Fig. 8. Comparison of % of samples with CINR> 10 dB – Excluding EARFCN 3750

It can be observed that the gap between operators reduces, however Altice continues with slightly better interference levels.

All three operators have more than 20% of the CINR samples below 10dB. This result is expected in a city environment, where there is a higher density of base stations, which increases the interference in the radio environment.

There are multiple reasons that can affect the level of interference in an LTE network. Among these we have:

Conclusions

From our study we can conclude the following

We have prepared a PDF reports with the detailed maps per operator and channel. To access this information and the raw data (in plain text) please email us to info@ekspresa.com or through our contact form.

During the coming weeks, the Ekspresa team will continue to carry out coverage measurements on emblematic roads of the city of Santo Domingo: Av. 27 de Febrero, Av. John F. Kennedy, Av. Washington and Las Américas freeway. We invite you to follow us on Linkedin to be notified when the next reports are available.