Soft Handover Report Range
Soft Handover Report Range
Definition
These parameters define the difference between the quality of a cell (currently it is
evaluated with PCPICH Ec/No) and the overall quality of the active set (if w=0, then it is
the quality of the best cell). The Report Range parameters for soft handover include
Report Range for event 1A and Report Range for event 1B.
Balance in setting
Settings of these parameters determine the size of the soft handover area and the soft
handover subscriber proportion.
In a CDMA system, it is required that the UE proportion in soft handover should be 30% to
40% so as to ensure smooth handover.
When the Report Range s are set at 5dB, the proportion of UEs in the soft handover state
(number of active set cells ≥ 2) is around 35%. It is recommended that this value be
slightly bigger in the early stage of deployment (5 to 7dB).
To save system resources, this figure can be gradually decreased with the growth of the
number of subscribers, but it must be bigger than 3dB.
The default configuration is 5dB. In addition, in special applications, different Report
Range values can be set for event 1A and event 1B to reduce the ping-pong effect and
change the soft handover proportion in some special applications.
For example, if the adjustment of the hysteresis values for events 1A and 1B is insufficient
for good control of the ping-pong effect, the Report Range for event 1B can be set larger
than that for event A to reduce the ping-pong effect.
However, the Report Range for events 1A and 1B should generally be kept consistent;
instead, the time-to-trigger setting, L3 filter coefficient and hysteresis value should used to
reduce the ping-pong effect.
Soft Handover Absolute Thresholds
Definition
These parameters correspond to the signal strength that satisfies the basic QoS
assurance. The soft handover absolute threshold parameters include absolute threshold
for event 1E and absolute threshold for event 1F.
Balance in setting
This value is the absolute threshold value used in the measurement reports of events 1E
and 1F in the soft handover algorithm, corresponding to the signal strength that satisfies
the basic QoS assurance.
This value affects the trigger of events 1E and 1F.
Because an absolute threshold is only a necessary condition, but not a sufficient one, for
access judgment, this value should be relative loose. With value settings in IS-95 and the
lower threshold of -20dB, -18dB is deemed to be a reasonable value.
A mobile station that is moving at the speed of 20km/h goes across the handover area in
averagely 20 to 30 seconds, while it takes only 5 to 6 seconds for a mobile station that is
moving at the speed of 100km/h to go across the handover area. When such factors as
hysteresis and trigger delay in event judgment are taken into account, the tacking time
needs to be further reduced.
Based on the analysis above, FilterCoef should be configured as follows: 5 as the default
setting for intra-frequency filter coefficient, and this parameter can be adjusted according
to the actual situation. In addition, for different cell coverage types, typical values are
recommended as follows:
a, if the cell covers urban area, the intra-frequency filter coefficient can be 7;
b, if the cell covers suburbs, the intra-frequency filter coefficient can be 6;
c, if the cell covers rural area, the intra-frequency filter coefficient can be 3.
If the number of UEs entering the handover area is the same as the number of UEs
leaving the handover area, there will be no influence on the actual soft handover
proportion. The bigger the hysteresis value is, the stronger the signal fluctuation
resistance capability will be, and thus the better the ping-pong effect will be suppressed,
but the slower the handover algorithm can react on signal changes.
Therefore, in the setting of this parameter, not only the radio environment (slow fading
characteristic) but also the actual handover distance and the UE movement speed should
be taken into due consideration.
The setting of this parameter can be adjusted within the range of 2 to 5dB. As events that
add cells to the active set, 1A and 1E are critical events. In order to ensure timely
handover, the hysteresis value for event 1A can be smaller, but not be too smaller, than
those for 1B, 1F, 1C and 1D; otherwise, the soft handover proportion will be affected.
In addition, In addition, hysteresis adjustment should generally be considered together
with the filter coefficient and time-to-trigger settings.
Balance in setting
The setting of the hysteresis value can effectively reduce the average handover times
and mis-handover times, and thus can prevent the occurrence of unwanted handover.
The bigger the hysteresis value is, the less the average handover timers will be.
However, the increase of the hysteresis value will bring more risks of call drop.
It is stipulated in TS 25.133V3.6.0 that the physical layer of intra-frequency
measurement updates the measurement result every 200ms. Therefore, a time-to-trigger
value below 200ms does not make any practical sense, and it should be as close as
possible to an integral multiple of 200ms.
In addition, mobile stations moving at different speeds respond differently to the time-to-
trigger value. The call drop rate is more sensitive to the time-to-trigger value when the
mobile station is in high-speed movement, while it is less sensitive when the mobile
station is in low-speed movement, and ping-pong handover and mis-handover are
suppressed to a certain extent.
Therefore, for cells where there are more high-speed moving mobile stations, this value
can be relatively small, while for cells where there are more low-speed moving mobile
stations, this value can be relatively big.
Different types of events have different requirements on the time-to-trigger setting:
events that add cells to the active set (event 1A and event 1E) generally require a small
time-to-trigger setting, while events that replace cells in the active set (event 1C and event
1D) generally require low ping-pong handover and mis-handover and do have produce
remarkable influence on the call drop rate. For the latter type of events, the time-to-trigger
setting can be properly big. For events that remove cells from the active set (event 1Band
event 1F), the time-to-trigger value is set mainly to reduce ping-pong handover; the initial
setting can be the same as that for event 1A and event 1E, and can be properly adjusted
based on the actual network statistics result.
WEIGHT
Definition
Weighted factor
Balance in setting
This parameter is used to determine the soft handover relative threshold based on the
measurement value of each cell in the active set.
The bigger this parameter is, the higher the relative threshold obtained under the same
condition will be.
When W=0, the determination of soft handover relative threshold is related to only the
best cell in the active set.