7/17/2019 Sami Handoff Algorithms Parameter Estimations for the Handoff Algorithms http://slidepdf.com/reader/full/sami-handoff-algorithms-parameter-estimations-for-the-handoff-algorithms 1/25 [email protected]S-72.333 Post-Graduate course in Radio Communications Handover algorithms and parameter estimations 05-03-02 Sami Vesala S-72.333 Post-Graduate Course in Radio Communications
25
Embed
Sami Handoff Algorithms Parameter Estimations for the Handoff Algorithms
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
7/17/2019 Sami Handoff Algorithms Parameter Estimations for the Handoff Algorithms
• There exists basically three (3) different ways to control handovers incellular networks. Based on these the handover algorithms arecharacterized to:
Network controlledMobile assisted –
Network controlled Mobile controlled
Used widely in 1st
generation systems
Link quality is only
monitored by serving
BTS and surrounding
ones
Handover decisions are
made centrally in
”MSC” Typically support only
intercell handovers
and have massive
delays (several secs..)
Used widely in 2nd gen.
systems (GSM).
Both serving BTS and
mobile measure link quality.
Surrounding BTS
measurements obtained
only by the mobile and
reported to the network.
Handover decisions made by serving BTS and MSC.
Both inter- and intracell
handovers supported and
typically 1-2 sec delays.
Used for example in
DECT
Both serving BTS and
mobile measure link quality Surrounding BTS
measurements obtained
only by the mobile and
reported to the network
Link measurements at the
serving BTS are relayed tothe mobile.
Mobile makes the handover
decision
Very low delays
7/17/2019 Sami Handoff Algorithms Parameter Estimations for the Handoff Algorithms
algorithm• If the handover threshold comparison indicates that a handover due to uplink or
downlink quality or level is required, the radio link properties of the adjacent cellsare evaluated in order to find a target cell. For example next equations can be
used in case of uplink or downlink quality or level (for quality, parameters withlevel are replaced):
• The averaged measurement values of signal quality or level are comparedagainst the set threshold values to trigger the handover.
E.g. AV_RXQUAL_UL/DL_HO vs HoThresholdsQualUL
• Parameters such as total number of averages to be taken into account and anumber of averages which have to be worse/better than threshold value are
often used.
7/17/2019 Sami Handoff Algorithms Parameter Estimations for the Handoff Algorithms
• The umbrella handovers are utilised in order to make handovers between
different network layers or cell sizes possible when needed. For example
between GSM 900/1800 or macro/micro cells.
• The measurement results of the adjacent cell must satisfy the equations
before the umbrella handover is possible.
• In order to be effective, the umbrella HO algorithm should take into account also thepower class of the mobile. For example so that selected cells are macrocells for vehicle
mobiles and microcells for handhelds.
Example
algorithm
7/17/2019 Sami Handoff Algorithms Parameter Estimations for the Handoff Algorithms
Velocity adaptive handover algorithms• A velocity adaptive handover algorithm must adapt the temporal window
over which the mean signal strength estimates are taken by either keeping thesampling period constant and adjusting the number of samples per
window, or vice versa.• In principle, high-speed mobile should use shorter average window size,
and low-speed MS should use longer average window size. Therefore, allaveraging-processes should have two sets of window parameters, one set forhigh speed mobiles and one set for low speed mobiles.
• By applying various window size, fast-moving MSs have shorter window sizeand they may handover to target cell faster. For a slow-moving MS, a longerwindow size is applied in order to prevent it from unnecessary oscillation.
• Velocity adaptive handovers are used to increase the capacity of a cellularnetwork, areas of high traffic density may be covered with a multi-layer
network consisting of different sized cells. If a high speed mobile is located insuch an area it should be located in a macrocell to decrease the amount of handovers. In other words a high speed mobile locating in a microcell wouldresult in increased signalling load and potentially high amount of dropped callsin the network.
7/17/2019 Sami Handoff Algorithms Parameter Estimations for the Handoff Algorithms
• In CDMA based systems the mobiles use the pilot signals from different base stations toinitiate and complete handovers. Each pilot signal is measured in terms of Ec/I0 (energyper chip/ interference)
• Soft handover is utilised, where the mobile can be connected to several base stationssimultaneously (in practise ≤ 3)
• Parameters to estimate in soft handover:
– Window-add
– Window-drop
– Window-replace
– Time-to-trigger
– active set size (AS)
– Individual offsets
)1( η −⋅⋅
⋅=
W N
R E
I
E
O
b
O
C
SHO algorithm performance can
be monitored by e.g. :
• Rate of AS updates
• Mean size of AS
• Average signal quality
7/17/2019 Sami Handoff Algorithms Parameter Estimations for the Handoff Algorithms