Figure 7.35 shows CDF plots of RSRQ levels for LTE 1800 ad LTE 800 measured
along the route marked in Figure 7.33. As indicated in the CDF plot, the CDF of LTE
800 includes the RSRQ level for both active set and detected set combined together.
The upper and lower extremes for LTE 800 is greater compared to LTE 1800. But the
much lower for LTE 800 compared to LTE 1800. This poor RSRQ is observed at the
Noise and interference scenario is similar for both of the frequency bands. But as the
coverage for the 1800 band closes down, the received signal level goes down as indicat-
ed in the RSRP analysis made in Section 7.2.2. With the RSRP level going down, the
difference of RSRP with the total received signal RSSI increases and thus the RSRQ
level goes down as indicated in Equation 5.1. LTE 800 measurement has been done
with both test sites active. Singerjärvi LTE 800 active and detected set data as men-
tioned previously has been filtered out by removing Kumpuvaara site data from the
measurement file. With the signal of Kumpuvaara existing in the measurement of LTE
800 as well, this would mean a considerable rise in RSSI level at the UE but with the
MEASUREMENTS AND ANALYSIS
68
deteriorated RSRP level of Singerjärvi LTE 800 site. This resulted in the poor RSRQ
levels at the cell edge for LTE 800.
The case explained above can be clarified with Figure 7.36. At a point where the Sing-
erjärvi LTE 800 is observed in the detected set list will obviously be served by Kumpu-
vaara LTE 800. Hence at that point, the RSSI level of the UE includes received signal or
RSRP of both Kumpuvaara and Singerjärvi. For example in above figure, let the meas-
ured RSRP for active set (from Kumpuvaara) is -90 dBm, RSRP for detected set (from
Singerjärvi) is -105 dBm and RSSI level be approximately -80 dBm.
Utilizing Equation 5.1, RSRQ of detected set would be -8 dB approx. But in case of
only site available as for LTE 1800, RSSI level at the same point will be lower for ex-
ample at -85 dBm. In that case, the RSRQ level with the single active set (assuming the
RSRP level be same i.e. -105 dBm) would be -3 dB approx. This is the reason why the
RSRQ values are poorer at the detected set region of Singerjärvi LTE 800. In the current
test network with just two sites, RSRQ might seem irrelevant. But with a loaded multi-
site network, triggers based on RSRP do not exactly give the interference scenario. In
such cases, adding RSRQ based triggers become handy.
Do'stlaringiz bilan baham: