High-Speed Downlink Packet Access(HSDPA) is an
enhanced3G(third-generation)mobile telephonycommunications
protocolin theHigh-Speed Packet Access(HSPA) family, also dubbed
3.5G, 3G+, or Turbo 3G, which allows networks based onUniversal
Mobile Telecommunications System(UMTS) to have higher data-transfer
speeds and capacity. As of 2013HSDPA deployments can support
down-link speeds of up to 42.3Mbit/s.HSPA+offers further speed
increases, providing speeds of up to 337.5Mbit/s with Release 11 of
the3GPPstandards
CapacityIt is the maximum throughput that the RBS can deliver to
one cell. The capacity is shared by all HSDPA users.
System CapacityIt is the average capacity per cell for a cluster
of cells. For system capacity calculation it is assumed that the
load is homogenously distributed and HSDPA is deployed in all
cells.
Dedicated Channel TrafficDCH traffic is defined as the traffic
carried by dedicated transport channels such as speech, PS or CS
radio bearers i.e. on channels other than HSDPA.
RBS LoadIt is the percentage of the maximum available RBS power
that is used in the downlink.
Power MarginPower margin saves a part of the RBS power to cater
for power variations, due to the dynamic UE behavior when users
move and experience varying channels conditions.For HSDPA it is
assumed that no power margin is needed and RBS may use 100% of the
available power in a system with HSDPA.
Shared Channel TransmissionShared channel transmission means
that a set of radio resources are dynamically shared among multiple
users.The sharing is done in time and code domain
Fast Radio Dependent SchedulingScheduling is the function that
determines which UE to transmit to at a given time instant.Three
scheduling algorithms are implemented.1. Proportional Fair
Scheduling2. Round Robin Scheduling3. Maximum Channel Quality
IndicatorProportional Fair SchedulingThe algorithm uses information
about fading peaks to prioritize users with good radio conditionsIt
also takes delay into account promoting users that have not been
given any data for a long timeRound Robin SchedulingThe algorithm
gives every user same amount of radio resources (TTI).The algorithm
is fair for all users from a resource point of view but bit rate
varies.
Max CQI(Channel Quality Indicator)UE sends CQI in the UL to aid
rate adaptation and schedulingThe algorithm maximizes system
throughput by prioritizing users with good radio channelsThe CQI
report estimates the number of bits that can be transmitted to the
UE using a certain assumed power with a block error rate of 10%
High-order ModulationHS-DSCH uses 16 QAM if the UE category
permit.This allows twice as high data rates to be transmitted as
compared to QPSK2 ms TTITransmission Time Interval for HSDPA is
short when compared to R99It is 2 ms for HS-DSCH for R99 it is
10-40 ms
Fast Link AdaptationAs opposed to R99 RBs, HS-DSCH is
transmitted with constant power within the TTI.Transmission rate is
controlled by adaptive channel coding.Data rate depends on radio
conditions (CQI)Fast Hybrid ARQ with soft combiningIn hybrid
automatic repeat request scheme, the received blocks that cannot be
decoded are buffered and soft combined with later received
transmissions of same information bits. Hybrid ARQ protocol
terminates in Node B which means short RTT (typically 12 ms
HSDPA PowerThe RBS power available for HSDPA is determined
dynamically, depending on R99 power usageAt least 25% of the
average power can be used for HSDPA
HSDPA Channel Structure
New Physical and Transport channels are introduced in HSDPA:
Transport ChannelHigh Speed Downlink Shared Channel
(HS-DSCH)
Physical ChannelsHigh Speed Physical Downlink Shared Channel
(HS-PDSCH)High Speed Shared Control Channels (HS-SCCH)High Speed
Dedicated Physical Control Channel (HS-DPCH)Associated Dedicated
Channel (A-DCH)
HSDPA accessibility rate degraded. Hi experts,What's the
possible reasons of HSDPA accessibility rate degraded (NSN
Technologie). The PS384 is used 20% in the infected NodeB,then i
suspect that the basic proble m is the CE congestion in the uplink
.In fact,the PS384 uplink bearer uses about 12 to 16 CE according
to the system module used. I'll change the parameter
HSDPAMaxBitrateUL from 6 (384kbps) to 4 (128kbps). This parameter
defines the maximum user bit rate allowed in a cell for HSDPA UL
DCH return channel. It is applicable to the PS domain NRT RAB.
For the HSDPA Accessibility please first check the RSRAN 073
Report for the infe cted NodeB and see where you are having the
maximum failures which is contributi ng max to HSDPA Accessibiility
degradation. Then Below are the reasons you have to look into. 1)
For Setup Fail BTS--Check CEs at BH 2) For setup fail UL Return
Channel--check UL power congestion and reduce the UL bit rate
accordingly. 3) For Setup Fail Iub Transport-- check Iub Resource
Congestion 4) For Setup Fail UE-- Check RAB Reconfiguration failure
Rate 5) For Setup Fail RNC-- Check RNC Unit Load and Alarms on
units 6) For Setup Fail too many Users-- Check simultaneous HSDPA
Users So first thing to do is to find actually which out of these 6
reasons is contrib uting for your accessibility degradation so that
you can work on that. Hope will be helpful. the HSDPA =downlink and
PS384 should be R99! the HSUPA=uplink and should be not use 384k,
so what you saying is DCH/HS, so if you have high RTWP (UL RSSI)
than try to lower the DCH rate to 128 or 64 Actually if the RAB
used is DL=HS/UL=DCH you will need HSDPAMaxBitrateUL =384 ot
herwise don't expect to have high bit rates on HSDPA DL (like 14
Mbpsec or more when you have MIMO or DC-HSDPA)). That is because
whatever application is used b y your mobile(TCP/IP usually) on
your test laptop it will need an acknowledgemen t sent in uplink(
to UTRAN network). That acknowledgement is using a DCH channel (
R99) that needs to have a specific bandwidth. You can try a new
value of HSDPAMaxBitrateUL =384 Kbpsec but have a look at your
HSDPA throughput and see if it is not limited by this change. After
implementing the HSDPAMaxBitrateUl=128Kbpsec,I found that HSDPA
throughput has deteriorated even more.Then I'll return to the first
configuration and try an alternative. Usually synchronization
affects RRC accessibility and you need to maintain a cer tain
number of consecutive sync burst to establish/keep your connection.
You nee d to check your sync timers, sometimes if it is set to high
value you lose your connection more frequently otherwise you should
have a look on the sync server/s olution. It is also a good idea to
check Sync alarms on node-B/RNC level. In cas e you have server
issue, then you are supposed to see more alarms at node-b/RNC
level. Moreover, you are supposed to see RAB drop due to abnormal
release/sync i ssue.
As Dan mention RRC failure will not effect HS accessibility or
HS/EUL Throughput , it will shows poor PS or CS accessibility, and
again if the failure due to syn ch it most be problem in Air
interface DL or UL. it could be UE accessing the ce ll from too
far, cell has high UL RSSI, or there is pilot population in DL. IF
y ou sure that cell s footprint is controlled and UL RSSI is good
you need to modify some timers as Dan suggest, I can t recommend
values as I m not NSN expert
While CE congestion is one of the common causes of accessibility
problems in mos t countries and vendors we should not conclude
until we see the actual CE utiliz ation...i suggest pulling CE util
stats first on a per hour basis to really veri fy if CE util is
reaching capacity/license limit....also there are counters indi
cating if accessibility issues are due to HW limitation...