PDA

View Full Version : In LTE,Why handoff based on RSRQ is not very good?



amit3111
2011-10-05, 12:09 AM
Why LTE handver based on RSRQ is not very good?
RSRQ from the definition is similar to RX Qual or Ec/Io. This means it will give the ratio of useful energy to the total energy. Then why is the LTE fraternity moving away from using it as Ec/Io is used: To decide the best sector for the UE?
To fully understand let's see how RSRQ is measured. RSRQ is function of RSRP over the total Noise (RSSI). Now unlike UMTS/CDMA, in LTE RSRP can be measured only in two symbols per Time Slot. And because of the bursty nature of traffic in LTE (because it's a data network) the load can fluctuate a lot. This means the total noise from the site and surrounding sites added together can change a lot from one Time Slot to another. Another point to add is the total noise is measured over the entire bandwidth that will have several subcarriers (This is the denominator in the RSRQ equation) and the RSRP is the average power in one subcarrier.
Because of these reasons, the RSRQ can fluctuate wildly and very quickly and hence handoff based on it will cause unnecessarily high frequency of handovers.
Amit

s52d
2011-10-05, 01:15 AM
Why LTE handver based on RSRQ is not very good?
RSRQ from the definition is similar to RX Qual or Ec/Io. This means it will give the ratio of useful energy to the total energy. Then why is the LTE fraternity moving away from using it as Ec/Io is used: To decide the best sector for the UE?
To fully understand let's see how RSRQ is measured. RSRQ is function of RSRP over the total Noise (RSSI). Now unlike UMTS/CDMA, in LTE RSRP can be measured only in two symbols per Time Slot. And because of the bursty nature of traffic in LTE (because it's a data network) the load can fluctuate a lot. This means the total noise from the site and surrounding sites added together can change a lot from one Time Slot to another. Another point to add is the total noise is measured over the entire bandwidth that will have several subcarriers (This is the denominator in the RSRQ equation) and the RSRP is the average power in one subcarrier.
Because of these reasons, the RSRQ can fluctuate wildly and very quickly and hence handoff based on it will cause unnecessarily high frequency of handovers.
Amit

And same is valid for HSPA as well. We are using RSCP for years, and never looked back.
Very simply: in handover position, with several signals, the strongest one is probably the best one.
Of course, all other parameters have to be adjusted properly, and this is the reason why
several operators stay on Ec/Io based soft handovers.

On the other hand, S/N based H/O make sense for inter-frequency handover. A beet weeker
carrier on inerference-free band is better as heavily interferred strong one.
Not many networks actually use more then one carrier just now. CSL maybe?

BR
s52d