Scribd
Upload
1K views

LTE PRACH Optimization (ENodeB Cluster 2) Final Report 05052015

The document discusses an LTE PRACH optimization trial conducted on a cluster of 272 cells. It provides: 1) An overview of the random access procedure and how root sequence planning can cause conflicts. 2) Details of the optimization trial conducted from April 28th to May 4th to resolve root sequence conflicts. 3) Analysis of key performance indicators like access success rates, call drop rates and traffic throughput, showing mostly stable or improved trends after the optimization.

Uploaded by

Syed Kazmi
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
1K views

LTE PRACH Optimization (ENodeB Cluster 2) Final Report 05052015

The document discusses an LTE PRACH optimization trial conducted on a cluster of 272 cells. It provides: 1) An overview of the random access procedure and how root sequence planning can cause conflicts. 2) Details of the optimization trial conducted from April 28th to May 4th to resolve root sequence conflicts. 3) Analysis of key performance indicators like access success rates, call drop rates and traffic throughput, showing mostly stable or improved trends after the optimization.

Uploaded by

Syed Kazmi
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 16

Security Level:

LTE PRACH Optimization (eNodeB Cluster 2)


Preliminary Report 05/05/2015
www.huawei.com

HUAWEI TECHNOLOGIES CO., LTD.

Contents

Security Level:

Overview

Trial Cluster

KPIs

Conclusion

www.huawei.com

HUAWEI TECHNOLOGIES CO., LTD.


Page 2

How it works
LTE Random Access Procedure (Contention-based)
During random access, the eNodeB send random access
configuration to a UE. The UE sends the random access
preamble to the eNodeB to initiate a random access request.
The random access preamble is a burst, which consists of a
cyclic prefix (CP), a preamble sequence, and an extra part in
the time domain and six resource blocks in the frequency
domain.
Preamble sequences are generated from cyclic shifts of
Zadoff-Chu (ZC) sequences. The logical index of a ZC
sequence ranges from 0 to 837.
Each cell can be configured with 64 preamble sequences. If
64 preamble sequences cannot be generated from a single
ZC sequence, they are obtained from subsequent ZC
sequences.
Contention-based random access
procedure
HUAWEI TECHNOLOGIES CO., LTD.

The logical index of a ZC sequence and the value of a cyclic


shift are indicated by the prach-Configndex field in SIB2. The
number of cyclic shifts is determined by the eNodeB based
on theConfidential
cell type and cell radius.
Huawei

How it works
Root Sequence Planning(Low Speed Cell)
Preamble sequences are generated from cyclic shifts of Zadoff-Chu (ZC) sequences. The logical index of
a ZC sequence( format 0-3)ranges from 0 to 837. Ncs indicate the cyclic shifts.

configuration
0

configuration
8

13

59

15

10

76
93

value

value
46

18

11

22

12

119

26

13

167

32

14

279

38

15

419

HUAWEI TECHNOLOGIES CO., LTD.

Huawei Confidential

How it works
Influence Analysis of Root Sequence conflict
For example, if root sequence of cell A and cell B is like below, and cell as and cell B
have the overlapping coverage area, then conflict exists:

Cell A

Cell b

Root Sequence Index


3
4
7

10

11

Due to Root Sequence conflict between Cell A and Cell B, if UE send preamble to Cell A then Cell B also may
detected the preamble and send RAR message.
UE will access to cell A, the RAR of cell B is not useful, so it is the waste of system resource
HUAWEI TECHNOLOGIES CO., LTD.

Huawei Confidential

Activity Description
With the root sequence reuse check, 272 cells root sequence configuration is suggested to be optimized
for Cluster 2:
eNodeB Cluster cell list

LTE PRACH
OPTIMIZATION 272 Cells

MOP Execution
Week
18

Date
28-April
00:00H

HUAWEI TECHNOLOGIES CO., LTD.

NE

MOP

MOP Ex

eNodeB Cluster 2

LTE PRACH
Optimization

HW

Huawei Confidential

Page 6

KPI Monitoring
Table below show the comparison of general KPI between the 3rd week (7 days, 4/21~4/27) before the
MOP and after the MOP(7 days, 4/28~5/04)). It take the average result of 7 days in 1 week. Referring to
it, we can find all KPI keep normal trend
MOP

Type

Units

Accesibility

LTE
PRACH Optimization

KPI Name

Before

After

Difference

RRC Setup Success Rate

99.967

99.968

0.001

E-RAB Setup Success


Rate

99.960

99.953

-0.008

(eNodeB Cluster 2)

Retainability

Call Drop Rate

99.992

99.994

0.002

Mobility

Intra-frequency Handover
Out Success Rate

0.014

0.009

-0.004

CSFB

success rate of successful


end to end data CSFB

99.940

99.938

-0.002

Mbps

average downlink
throughput per cell

2503.857

2516.714

12.857

GB

DL Traffic Volume

16.227

16.136

-0.558

Max Active User

2060.344

2057.272

-3.072

Traffic

Load

HUAWEI TECHNOLOGIES CO., LTD.

Huawei Confidential

Page 7

KPI Monitoring-RRC SSR

RRC SSR keep normal trend which is greater than 99.9% all the time;

HUAWEI TECHNOLOGIES CO., LTD.

Huawei Confidential

Page 8

KPI Monitoring-ERAB SSR

ERAB SSR keep normal trend;

HUAWEI TECHNOLOGIES CO., LTD.

Huawei Confidential

Page 9

KPI Monitoring-Call Drop Rate

Call Drop Rate keep stable compared with before, no increase after the MOP;

HUAWEI TECHNOLOGIES CO., LTD.

Huawei Confidential

Page 10

KPI Monitoring- Intra-Freq HO SR

Intra-Freq HO SR keep stable;

HUAWEI TECHNOLOGIES CO., LTD.

Huawei Confidential

Page 11

KPI Monitoring-CSFB SR

CFSB Success Rate is in regular trend;

HUAWEI TECHNOLOGIES CO., LTD.

Huawei Confidential

Page 12

KPI Monitoring-Traffic User

Total traffic average user number didnt change much after the MOP;

HUAWEI TECHNOLOGIES CO., LTD.

Huawei Confidential

Page 13

KPI Monitoring-DL Traffic & Throughput

DL traffic and average DL throughput have no big difference;

HUAWEI TECHNOLOGIES CO., LTD.

Huawei Confidential

Page 14

KPI Monitoring-Random Access Attempt

After the MOP executed on 4/28, we can find that L.RA.GrpB.Att times decreased absolutely.
As the root sequence overlapping of optimized cells mainly happened in the Group Bs range.
When the confliction resolved, the wrong detect of random access decreased, then
L.RA.GrpB.Att times reduced.
HUAWEI TECHNOLOGIES CO., LTD.

Huawei Confidential

Page 15

Conclusion
There is no negative impact from LTE PRACH Optimization MOP. All KPI keep
normal trend ;
After the MOP executed, the wrong detect of random access times decreased
absolutely which means less random access response times, resulting in less
waste of system resource.

HUAWEI TECHNOLOGIES CO., LTD.

Huawei Confidential

Page 16

You might also like