179844550 2G Huawei Capacity Optimization Process Pptx

HUAWEI 2G CAPACITY OPTIMIZATION

For internal use 1 © Nokia Siemens Networks

Presentation / Author / Date

Contents 1. Capacity Optimisation Overview 2. CCCH Capacity Optimisation 3. SDCCH Capacity Optimisation 4. Traffic Capacity Optimisation 5. Transmission Capacity Optimisation 6. BSC Hardware



Document Information  Document Version: 1.0

 Issue Date: June 25, 2010  Author: Christos Kyriazopoulos  Document Owner: Ville Salomaa  SOFTWARE RELEASE: GBSS9.0  SCOPE:  BSS Capacity Monitoring  BSS Capacity Optimisation  Transmission Network Monitoring  Transmission Network Capacity Optimisation  CONVENTION:  Raw counters are marked in BLUE  Formulas are marked in GRAY  Parameters are marked in RED  MML commands are marked in GREEN






Network Elements Capacity Overview A Interface •Circuits configuration •A over IP

MS/Client parameters

HLR/AuC/EIR

MSC/VLR

•GSM/GPRS/E DGE capability and release

BSC

•Multislot support

BTS

•Boards

•TRX

•SW features

A Gs

BSC Um Abis

MS

Gb

BTS

Abis Interface

Gb Interface

Air Interface (Um)

•E1 configuration

•CCCH(PCH+AGCH)

•Abis over IP

•Gb link capacity

•SDCCH •CS Traffic (TCH)

•PS Traffic (PDCH)



SGSN

GGSN

PDN

General Methodology 1. Monitor blocking counters: - If blocking > 0 then take measures to relieve congestion

reactive optimisation

2. Monitor utilization counters: - If utilization of resources is above specific alarm threshold (e.g. > 80%) take measures to improve capacity proactive optimisation Blocking or Utilization issues must occur repeatedly before triggering capacity optimisation; check corresponding counters on same hours, same day of consecutive weeks: - Check duration of the problem - Check availability of current and adjacent network elements - Check patterns of behaviour (hours of occurrence, weekdays/weekends) - Check surroundings (theatres, concert halls, stadiums, shopping centres, etc.) - Check blocking/utilization of adjacent network elements (homogeneously spread or unbalanced) 3. Introduce solution: - Re-establish full availability - Increase support from existing NEs (coverage, tilts, azimuths, etc.) - Increase NEs and/or Interfaces capacity - Add NEs This document focuses on increasing NEs and Interfaces capacity






CCCH (1)  AGCH performance monitoring BLOCKING: - AGCH Blocking = ([L3188A:CELL_DEL_IND]/[Channel Requests (all reasons)])*{100} % - L3188A:CELL_DEL_IND: MSG DEL IND Messages Sent on Abis Interface, due to overload DL CCCH (when the IMM ASS CMD message sent from the BSC is deleted by the BTS because the downlink CCCH in the cell is overloaded, the BTS reports a DELETE IND message to the BSC) - Channel Requests (all reasons) = A300A:CELL_CH_REQ_MOC + A300C:CELL_CH_REQ_MTC + A300D:CELL_CH_REQ_ECALL + A300E:CELL_CH_REQ_CALL_REESTB + A300F:CELL_CH_REQ_LOC_UPDATE + A300H:CELL_CH_REQ_PACKET_CALL + A300I:CELL_CH_REQ_LMU_AND_RESERVED + A300K:CELL_CH_REQ_PROTOCOL_INCOMPATIBLE  PCH performance monitoring BLOCKING: - PCH Blocking = [L3188L:CELL_FCTRL_PAGING_MSG_DEL_PCH_QUE]/([Delivered Paging Messages for CS Service]+[Delivered Paging Messages for PS Service])*{100} % - Paging Overload Rate CS = ([MSG CCCH LOAD IND (PCH) Messages Sent on Abis Interface]/[Delivered Paging Messages for CS Service])*{100} % - Paging Overload Rate PS = ([PACKET CCCH LOAD IND Messages Sent on Abis Interface])/([Delivered Paging Messages for PS Service])*{100} % - L3188L:CELL_FCTRL_PAGING_MSG_DEL_PCH_QUE: Paging Messages Discarded from the PCH Queue (Note: when the cell is configured with the PCCCH, this measurement provides the number of discarded paging messages in only the CS domain) - A330:CELL_PAGES_CS: Delivered Paging Messages for CS Service - A331:CELL_PAGES_PS: Delivered Paging Messages for PS Service - L3188C:CELL_CCCH_LOAD_PCH_OVERLOADS: MSG CCCH LOAD IND (PCH) Messages Sent on Abis Interface (overload due to CS service) - L3188D:CELL_OVERLOAD_PS: PACKET CCCH LOAD IND Messages Sent on Abis Interface (overload due to PS service) For internal use 8 © Nokia Siemens Networks


CCCH (2)  PCH performance monitoring UTILISATION: - A343: CELL_QUE_OCCUPY_PK_VALUE: Peak PCH Paging Queue Usage (When the new paging messages are successfully added to the PCH queue after scheduling, the number of current PCH messages in the PCH queue is calculated, and then divided by the valid PCH queue length (calculated based on the current queuing ratio). Thus, the average paging queue usage is obtained. Then, record the peak paging queue usage within the granularity period of one minute.)



CCCH (3) Capacity optimisation methodology 1. If blocking occurs on CCCH (AGCH or PCH) check the following parameters: - BSAGBLKSRES: Blocks Reserved for AGCH. Value range: 0-7. Recommended: 2 - BSPAMFRAMS: Multi-Frames in a Cycle on the Paging CH. Value range: 2-9. Value depends on paging load. Increase value when paging load increases. Value should be kept as small as possible. - PAGTIMES: Paging Times. Value range: 1-8 (For the BTS, this parameter is used to determine paging retransmissions. This parameter and the number of paging times configured in the MSC determine the number of paging retransmissions.) 2. Check if Flow Control feature is enabled. Recommendation is that Flow Control is always enabled.  Control of the Arrival Rate of Paging Messages on the A Interface (MSC-BSC): The control of the arrival rate of paging messages on the Abis interface (from the A interface) is performed when the parameter STARTPGARRIVALCTRL=YES. The BSC monitors the number of paging messages over A interface in real time. If the number of paging messages exceeds PGMAXMSGNUMINPERIOD within a PGSTATPERIOD, the BSC discards the subsequent paging messages coming from A interface within this period. - STARTPGARRIVALCTRL: Paging Arrival Control. Recommended value: YES - PGMAXMSGNUMINPERIOD: Max Paging Message Num in Period. Recommended value: 660 - PGSTATPERIOD: Paging Statistical Period. Recommended value: 1000ms  Flow Control on LAPD links (BSC-BTS): • Flow control depending on the length of I frame queue of the LAPD links: If the rate at which the BSC sends messages on the LAPD links is higher than the rate at which the messages on the Abis interface are sent to the BTS, DL messages are accumulated in the I frame queue or are even discarded. - FCSTHD: Flow Control Start Threshold. Recommended value: 90% - FCETHD: Flow Control End Threshold. Recommended value: 80%



CCCH (4) Capacity optimisation methodology  Flow Control on LAPD links (continued): • Flow control depending on the flow control level and the service type of the paging messages: This function works only when the BSC is connected to HUAWEI core network. The BSC controls the flow on the LAPD links based on the CPU usage. The BSC obtains the CPU usage of the boards once every second and compares it with the specified threshold to determine the flow control level.

Related parameters: - P11: CPU Usage for Critical Paging FC. Recommended value: 90% - P12: CPU Usage for Major Paging FC. Recommended value: 85% - P13: CPU Usage for Minor Paging FC. Recommended value: 80% - P14: CPU Usage for Slight Paging FC. Recommended value: 75% For internal use 11 © Nokia Siemens Networks


CCCH (5) Capacity optimisation methodology 3. Check volume of PS pagings (A331:CELL_PAGES_PS: Delivered Paging Messages for PS Service). If too high then check if PCCCH is configured. If PCCCH is configured then packet pages can be transmitted through PPCH, thus reducing PCH load. CS pages can also be transmitted through packet control channels (PACCH or PPCH). For this to work, Gs interface needs to be configured between SGSN-MSC. Also Network Mode of Operation should be set to 1. - NMO: Network Operation Mode 4. Check Location Area: re-size might be required (make smaller). 5. Consider splitting cells in the paging overload area. This will grow CCCH capacity. 6. Add CCCH capacity (Extended BCCH).






SDCCH (1)  SDCCH Performance Monitoring BLOCKING: - SDCCH Congestion Rate (Overflow) = ([Failed SDCCH Seizures due to Busy SDCCH]/[SDCCH Seizure Requests])*{100} % UTILISATION: - SDCCH Utilisation = ([R3550M:CELL_SIG_CH_TRAF_SD]/[CR3020:CELL_CH_AVAIL_NUM_SD_AVR])*{100} % - R3550M:CELL_SIG_CH_TRAF_SD: Traffic Volume on Signalling Channels (SDCCH) (Erlangs) - CR3020:CELL_CH_AVAIL_NUM_SD_AVR: Mean Number of Available Channels (SDCCH) AVAILABILITY: - SDCCH Availability = ([Mean Number of Available Channels (SDCCH)]/[Mean Number of Dynamically Configured Channels (SDCCH)])*{100} % OTHER: - Immediate Assignment Success Rate = ([Call Setup Indications (Circuit Service)]/[Channel Requests (Circuit Service)])*{100} %



SDCCH (2) Capacity optimisation methodology 1. Check cell signalling statistics: - if SDCCH Congestion Rate (Overflow) > 0.5 % then cell is an object for capacity optimisation 2. Check SDCCH usage. If SDCCH is congested try to identify the reason of high usage. Check SMSs, LAUs, Call Setups. SDCCH cause distribution will influence the possible solution. • Check Call Setups: - CELL_ESTB_IND_MOC_NONSMS_SD: Number of Call Setup Indications for MOC on SDCCH - CELL_ESTB_IND_MTC_SD: Number of Call Setup Indications for MTC on SDCCH • Check amount of SMS. Check and verify with Core engineers SMS Center parameterization. - A3030B: CELL_ESTB_IND_MOC_SMS_SD: Number of Call Setup Indications for SMS on SDCCH - CA3340: CELL_Pt_to_Pt_SMS_SD: Number of Point-to-Point Short Messages on SDCCH (includes UL + DL) • Check LAU/RAU requests: - A300F: CELL_CH_REQ_LOC_UPDATE: Number of Channel Requests for Location Update - A3030F: CELL_ESTB_IND_LOC_UPDATE_SD: Number of Call Setup Indications on SDCCH for Location Update 3. If high SCDDH usage is due to LAU then: • Check if the problem is caused by roamers that do not have access to the network, thus causing big amount of failed LAUs/RAUs. • Check if cell is in LA border: if yes, then we can increase CRH parameter value - CRH: Cell Reselect Hysteresis Parameters (Cell reselection hysteresis. This is one of the parameters used for deciding whether to reselect cells in different location areas.) • Check LA border planning. Verify LA borders by checking HO statistics between cells in LA border: - H380:CELLCELL_INCELL_HO_REQ: Incoming Inter-Cell Handover Requests between 2 cells • Check the value of T3212; if too low, increase - T3212: T3212 (This parameter specifies the length of the timer for periodic location update). Recommended value: as high as possible, usually 4h. • Check whether moving LA borders (if possible to move) could help relieving the congestion. • Check the pattern of LAU requests. Check hours and duration of high number of such requests. Check whether the problem is constant throughout the day or it occurs only during 1 hour for example. If the problem occurs only on specific hour of day check if it is worth acting to solve it (costs vs. benefits). For internal use 15 © Nokia Siemens Networks


SDCCH (3) Capacity optimisation methodology 4. Check if TCH Immediate Assignment is allowed: - IMMASSEN: TCH Immediate Assignment (Whether to allow immediate TCH assignment. If this parameter is set to YES, the BSC can assign a TCH immediately when there is no available SDCCH for a channel request.) Note: It is not recommended to activate this in congested LA borders. 5. Activate SDCCH dynamic conversion feature: Dynamic SDCCH conversion can be triggered if the SDCCH resource is insufficient or the SDCCH allocation fails during the channel assignment - SDDYN: SDCCH Dynamic Allocation Allowed (Whether to allow SDCCH dynamic allocation, that is, whether to allow dynamic conversion between TCHs and SDCCHs.) - IDLESDTHRES: Idle SDCCH Threshold N1 (When the number of idle SDCCH channels in a cell is smaller than this parameter, the system searches for available TCHs and transforms them into SDCCH channels) - CELLMAXSD: Cell SDCCH Channel Maximum (Maximum number of SDCCHs in the cell. Before converting a TCH into an SDCCH, the BSC compares the number of SDCCHs after the conversion in the cell with "Cell SDCCH Channel Maximum". If the number of SDCCHs after the conversion in the cell exceeds this parameter, the BSC does not convert the TCH into an SDCCH.) 6. Add SDCCH/8 channel 7. Add TRX

Note: Huawei recommended SDCCH configuration in a cell is: - 1 SDCCH/8 per 2 TRXs for FR - 1 SDCCH/8 per 1 TRX for HR






Traffic (1)  TCH Performance Monitoring BLOCKING: - TCH Congestion Rate (Overflow) = (([Failed TCH Seizures due to Busy TCH (Signaling Channel)]+[Failed TCH Seizures due to Busy TCH (Traffic Channel)]+[Failed TCH Seizures in TCH Handovers due to Busy TCH (Traffic Channel)])/([TCH Seizure Requests (Signaling Channel)]+[TCH Seizure Requests (Traffic Channel)]+[TCH Seizure Requests in TCH Handovers (Traffic Channel)]))*{100} % UTILISATION: - TCH Utilisation = (([Mean Number of Busy TCHs (TCHF) (900/850 Cell)]+[Mean Number of Busy TCHs (TCHF) (1800/1900 Cell)]+[Mean Number of Busy TCHs (TCHH) (900/850 Cell)]+[Mean Number of Busy TCHs (TCHH) (1800/1900 Cell)])/([Mean Number of Dynamically Configured Channels (TCHF) (900/850 Cell)]+[Mean Number of Dynamically Configured Channels (TCHF) (1800/1900 Cell)]+[Mean Number of Dynamically Configured Channels (TCHH) (900/850 Cell)]+[Mean Number of Dynamically Configured Channels (TCHH) (1800/1900 Cell)]))*{100} % - TCH FR Utilisation = (([Mean Number of Busy TCHs (TCHF) (900/850 Cell)]+[Mean Number of Busy TCHs (TCHF) (1800/1900 Cell)])/([Mean Number of Dynamically Configured Channels (TCHF) (900/850 Cell)]+[Mean Number of Dynamically Configured Channels (TCHF) (1800/1900 Cell)]))*{100} % - TCH HR Utilisation = (([Mean Number of Busy TCHs (TCHH) (900/850 Cell)]+[Mean Number of Busy TCHs (TCHH) (1800/1900 Cell)])/([Mean Number of Dynamically Configured Channels (TCHH) (900/850 Cell)]+[Mean Number of Dynamically Configured Channels (TCHH) (1800/1900 Cell)]))*{100} % AVAILABILITY: - TCH Availability = ([Mean Number of Available Channels (TCH)]/[Mean Number of Dynamically Configured Channels (TCH)])*{100} %



Traffic (2) Capacity optimisation methodology 1. Check cell congestion statistics: - if TCH Congestion Rate (Overflow) > 1.5 % then cell is an object for capacity optimisation 2. Check cell traffic channel availability in order to verify that congestion is not due to availability issue. - if TCH Availability < 100%, cell congestion is due to availability issue: check cell alarms 3. Check availability of neighboring sites. If neighboring cells are unavailable this will cause big amount of HOs directed to our current cell thus leading to congestion. 4. Check cell traffic channel configuration. Check if all HR resources are in use before TCH congestion occurs. - verify that HR is enabled; following parameters can be used to cause earlier HR channel selection according to cell load: - TCHRATEMODIFY: TCH Rate Modify (When this parameter is set to YES and the BSC policy is used, the BSC preferentially selects full-rate or half-rate channels based on the internal load) - TCHBUSYTHRES: TCH Traffic Busy Threshold (If the current channel seizure ratio reaches or exceeds this value, the half-rate TCH is assigned preferentially; otherwise, the full-rate TCH is assigned preferentially) - TCHAJFLAG: TCH Rate Adjust Allow (Whether to allow the cell to dynamically change a channel from full rate to half rate or from half rate to full rate) - ALLOWHALFRATEUSERPERC: Ratio of TCHH (Maximum allowed ratio of the number of half rate channels to the total number of channels in a cell); verify that value = 100 - In case AMR is supported by the operator, verify that is enabled. Following parameters can be used to cause earlier AMR HR channel selection according to cell load (note: it would make sense first to check whether there is a considerable amount of AMR HR capable phones in the network): - AMRTCHHPRIORALLOW: AMR TCH/H Prior Allowed (Whether to enable the BSC to assign AMR half rate channels preferentially according to the channel types allowed by the MSC and the current TCH seizure ratio of the cell) - AMRTCHHPRIORLOAD: AMR TCH/H Prior Cell Load Threshold (Load threshold for assigning half rate channels preferentially. If the current TCH seizure ratio of the cell is greater than this threshold, AMR half rate channels are assigned preferentially.) - ALLOWAMRHALFRATEUSERPERC: Ratio of AMR-HR (Maximum allowed ratio of the number of AMR half rate channels to the total number of channels in a cell); verify that value = 100 For internal use 19 © Nokia Siemens Networks


Traffic (3) Capacity optimisation methodology 4.

Load balancing between cells: certain features can be activated to manage the traffic sharing between cells - Load HO: enable load HO algorithm - LoadHoEn: Load Handover Support (This parameter specifies whether a traffic load-sharing handover is enabled. The load handover helps to reduce cell congestion, improve success rate of channel assignment, and balance the traffic load among cells, thus improving the network performance.) - TRIGTHRES: Load HO Threshold (The load handover is triggered when the traffic load in a cell is greater than the value of this parameter) - LoadAccThres: Load handover Load Accept Threshold: (If the load of a cell is lower than the value of this parameter, the cell can admit the users handed over from other cells with higher load.) - Directed Retry: enable directed retry due to load - DIRECTRYEN: Directed Retry (Whether to enable a directed retry. The directed retry is to hand over an MS to a neighbouring cell in the same procedure as the handover.) - ASSLOADJUDGEEN: Assignment Cell Load Judge Enable (Activate directed retry due to heavy load in the cell) - CDRTTRYFBDTHRES: Cell Directed Retry Forbidden Threshold (if the current cell load is greater than or equal to the value of this parameter, the BSC allocates a channel to an MS through the process of directed retry) - DTLOADTHRED: Directed Retry Load Access Threshold (Only a cell whose load is lower than or equal to this threshold can be selected as a candidate target cell for directed retry)




Load balancing between cells: - Concentric Cells: check relative parameters so as to implement optimal traffic sharing between underlaid-overlaid cells. Main parameters to be checked are shown below: - HRIUOLDRATESELALLOW: Load of UL-OL Cells Rate Select Allowed (Whether to enable the BSC to assign half or full rate channels to MSs according to the channel seizure ratio in the overlaid and underlaid subcells) - TCHTRICBUSYOVERLAYTHR: TCH Traffic Busy Overlay Threshold (If the channel seizure ratio of overlaid subcell is greater than the value of this parameter, half-rate channels are assigned. Otherwise, full-rate channels are assigned) - TCHTRIBUSYUNDERLAYTHR: TCH Traffic Busy Underlay Threshold: (If the channel seizure ratio in the underlaid subcell exceeds the value of this parameter, half-rate channels are assigned. Otherwise full rate channels are assigned.) - ATCBHOEN: Concentric Circles ATCB HO Allowed (Whether to enable the ATCB handover algorithm for the concentric cell. According to the neighbour cell signal, the ATCB handover algorithm determines the coverage of the overlaid subcell and balances the load between the overlaid subcell, underlaid subcell, and neighbour cell. Therefore, the algorithm helps to decrease the interference, to improve the conversation quality, and to achieve aggressive frequency reuse in the overlaid subcell.) - Enhanced Dual Band Network: check relative parameters so as to implement optimal traffic sharing between underlaidoverlaid cells. Main parameters to be checked are shown below: - OUTGENOVERLDTHRED: UL Subcell General Overload Threshold (When the load of the underlay subcell is higher than this parameter, some of the calls in the underlay subcell will be switched to the overlay subcell, and channels in the overlay subcell will be preferentially assigned to calls initiated in the underlay subcell as well.) - OUTLOWLOADTHRED: UL Subcell Lower Load Threshold (When the load of the underlay subcell is lower than this parameter, some of the calls in the overlay subcell will be switched to the underlay subcell, and channels in the underlay subcell will be preferentially assigned to channel requests initiated in the overlay subcell as well.)




Check if additional capacity related features can be activated in the network: - BCCH Dense Frequency Multiplexing: enables the BCCHs to reuse frequencies more tightly to free more frequencies for nonBCCH TRXs, thus increasing the system capacity. - TIGHTBCCHSWITCH: TIGHT BCCH Switch (Whether to enable the BCCH aggressive frequency reuse algorithm) - Interference Based Channel Allocation (IBCA): On a network where the frequency resources are insufficient, the same frequency is repeatedly used in neighboring cells. In this case, severe co-channel interference and adjacent-channel interference exist on the network, and such interference cannot be eliminated even if the frequency hopping (FH) technology is applied. When the number of calls on such a network exceeds a certain limit, the mutual interference between calls will decrease the speech quality to such a level that the C/I ratio required by a call is not guaranteed. In this case, even if there is an idle channel on this network, the idle channel cannot be assigned to a call because of the severe interference. As a result, the utilization of the frequency resources is restricted, and the network capacity is thus decreased. To alleviate the interference on the network, the Interference Based Channel Allocation (IBCA) algorithm is introduced. The IBCA algorithm requires the BSC to estimate the C/I ratio of the new call in every channel assignment procedure; it also requires the BSC to estimate the interference caused to the established calls on the network when an idle channel is assigned to a new call. In this way, the optimal channel, that is, the one that meets the C/I ratio requirement of the new call and causes the least interference to the established calls after being occupied, is assigned to the new call to alleviate the interference and ensure the full use of the frequency resources. - IBCAALLOWED: IBCA Allowed (Whether to enable the IBCA algorithm) - Flex MAIO: BSC dynamically adjusts the MAIO according to the current interference level of a channel when assigning an MAIO to the channel (note that the BSC assigns an MAIO to only a channel under activation). In this way, the BSC assigns the MAIO with the minimum interference to the channel, and the channel experiences the minimum interference in the BTS. - FLEXMAIO: Start Flex MAIO Switch (Whether to enable the function of Flex Mobile Allocation Index Offset) - Precondition for Flex MAIO use is to have RF hopping enabled (BB FH does not support Flex MAIO). Also, only CS service supports Flex MAIO (PS service does not).




If congestion is still present although the previous described fine tuning and features activation, then: - Check Interference in the network (C/I); check frequency plan - Check coverage: maybe network layout should be changed in traffic hot spots - We can use TA distribution in order to identify traffic distribution among cells. In some cases overshooting can be detected, so we can check the possibility to reduce service area of the overshooting cell. Before doing so, we need, of course, to make sure that there is clear dominance in the area that we are going to shrink serving cell’s coverage. - Implement physical network changes where necessary and feasible: tilt, azimuth, antenna type, etc. - Add TRX - Long term monitoring (e.g. one month) can be used to identify whether we have constant growth in traffic in a site and area close by. If traffic increases in area level and we have already high HR/AMR HR utilization then there are not too many other options than implement a new site. - Add Site






Transmission - Abis Interface (1)  Abis Transmission Modes The following transmission modes can be used for Abis interface: • Abis over TDM In TDM-based networking mode, the BSC and the base station communicate with each other through the SDH/PDH network, and TDM transmission is applied to the Abis interface. - Fix Abis: the timeslot resources on the Abis interface correspond to the TCH resources on the Um interfaces based on configurations, and the mapping is 1:1. - Flex Abis: Flex Abis indicates that different services in the same BTS share Abis interface timeslots. An Abis TS is assigned only when an Um channel is occupied. In cases different BTSs share Abis interface timeslots, Abis interface timeslots are assigned as required. It is advisable to enable Flex Abis in the following cases: - Transmission resources are limited, and the rent for them is very high, for example, in satellite transmission mode. - The traffic volume in actual situation is far lower than that in the channel resource planning. - The cells that share Abis interface transmission resources have different traffic peak hours. - The proportion of PS service users in the cell is high. • Abis over IP In IP-based networking mode, the BSC and the base station communicate with each other through the IP/SDH/PDH network, and layer 3 of the protocol stack of Abis interface uses the IP protocol.



Transmission - Abis Interface (2)  Abis over TDM – Performance Monitoring  Fix Abis: There is 1:1 mapping between Um interface channels and Abis timeslots. So blocking on Abis is not possible. UTILISATION: - L1151A: BS_ABIS_USED_TS_AVG: Average Number of Busy Timeslots per Abis Port - L1151B: BS_ABIS_USED_TS_MAX: Maximum Number of Busy Timeslots per Abis Port - LST ABISTS: An OM command that queries the number of service timeslots configured on the Abis interface per Abis port. Utilisation can be found by dividing the “Average Number of Busy Timeslots per Abis Port” by the number of “Configured TS per Abis port”.  Flex Abis: In dynamic allocation mode, the Abis interface transmission resources form a pool. These resources are allocated to the Um interface channels only when the channels are occupied. This way it is possible to map more Um channels than configured timeslots on Abis. Which might create congestion during busy hours. BLOCKING: - A312F: CELL_ASS_FAIL_NO_IDLE_ABIS: Number of TCH Assignment Failures due to no available Abis resources - H322M: CELL_INTRABSC_INCELL_HO_FAIL_NO_IDLE_ABIS: Number of incoming internal inter-cell handover failures because no Abis CS resource is available in the target cell when the Abis dynamic allocation is enabled - H342L: CELL_INTERBSC_INCELL_HO_FAIL_NO_IDLE_ABIS: Number of failed incoming external inter-cell handovers because no Abis circuit resource is available in the target cell when the Abis dynamic allocation is enabled - RR2751: SITE_8K_ASS_FAIL_RATE: Congestion Ratio of Dynamic Assign Abis Resource (8K) - RR2752: SITE_16K_ASS_FAIL_RATE: Congestion Ratio of Dynamic Assign Abis Resource (16K); this counter includes CS and PS service.



Transmission - Abis Interface (3)  Abis over TDM – Performance Monitoring  Flex Abis: In dynamic allocation mode, the Abis interface transmission resources form a pool. These resources are allocated to the Um interface channels only when the channels are occupied. This way it is possible to map more Um channels than configured timeslots on Abis. Which might create congestion during busy hours. UTILISATION: - R2720:SITE_ASS_CS_8K: Dynamic Assign Abis Resource (8K_CS) (This counter is measured when the 8 kbit/s Abis resources are assigned to the voice channel for the CS service) - R2721: SITE_ASS_CS_16K: Dynamic Assign Abis Resource (16K_CS) (This counter is measured when the 16 kbit/s Abis resources are assigned to the voice channel for the CS service) - R2741: SITE_FLEX_TS_NUM: FlexAbis TSs (This counter is used to measure the number of available FlexAbis timeslots of the site.) - R2742: SITE_FAULT_FLEX_TS_NUM: Fault Flex TSs (This counter is used to measure the number of fault FlexAbis timeslots of the site.)



Transmission - Abis Interface (4)  Abis over IP – Performance Monitoring  BSC6000 BLOCKING: - Congestion Rate = ([L01057:BSC_LGCPORT_TXDROPPACKETS]/([L01057:BSC_LGCPORT_TXDROPPACKETS]+[L01055:BSC_LGCP ORT_TXPACKETS]))*{100} - L01057:BSC_LGCPORT_TXDROPPACKETS: Number of Packets Discarded on a Logical Port due to congestion - L01055:BSC_LGCPORT_TXPACKETS: Number of Packets Sent on a Logical Port UTILISATION: - L01061:BSC_LGCPORT_MEAN_TXRATE: Average Throughput of Packets Sent on a Logical Port - BANDWIDTH: Bandwidth of the Logical Port [32kbps] (ADD LOGICALPORT)  BSC6900 BLOCKING: - VS.ANI.IP.FailResAllocForBwLimit: Number of Failed Resource Allocations Due to Insufficient Bandwidth on the IP Transport Adjacent Node - VS.IPPATH.Fwd.Cong: Number of Forward Congestions on the IP Path - VS.IPPATH.Fwd.Cong.Dur: Duration of Forward Congestion on the IP Path - VS.IPPATH.Bwd.Cong: Number of Backward Congestions on the IP Path - VS.IPPATH.Bwd.Cong.Dur: Duration of Backward Congestion on the IP PATH UTILISATION: - OS.ANI.IP.AllocedFwd: IP Path Forward Bandwidth Allocated to IP Transport Adjacent Node - OS.ANI.IP.AllocedBwd: IP Path Backward Bandwidth Allocated to IP Transport Adjacent Node - TXBW: Forward Bandwidth: Transmit bandwidth of IP path (ADD IPPATH) - RXBW: Backward Bandwidth: Receive bandwidth of IP path (ADD IPPATH) For internal use 28 © Nokia Siemens Networks


Transmission - Abis Interface (5)  Abis over IP – IPPM (IP link Performance Monitoring) The quality of an Abis IP link can be monitored through the IPPM function. Enable this through MML command: ADD IPPM (BSC6000), ACT IPPM (BSC6900). The IPPM is a method to check the IP transmission quality. In this method, the forward monitoring (FM) and backward reporting (BR) messages are used to check the transmission quality of IP paths. - The monitor periodically sends FM messages to indicate number of outgoing packets, number of bytes, and sending time. - The peer responds with BR messages after receiving the FM message to report number of received packets, number of received bytes, the receiving time of PM message and the sending time of BR response. - The sender calculates packet loss rate, transmission delay and jitter according to the BR response from the receiver. BSC6000: - L01033:IPPM_FORWARD_DROPMEANS: Average IPPM Forward Packet Loss Rate - L01034:IPPM_FORWARD_PEAK_DROPRATES: Maximum IPPM Forward Packet Loss Rate - L01035:IPPM_RTT_MEANS: Average RTT Delay of an IPPM Link - L01032:IPPM_MAXRTTDELAY: Maximum RTT Delay of an IPPM Link BSC6900: - VS.IPPM.Forword.DropMeans: Average Forward Packet Loss Rate of IPPM - VS.IPPM.Forword.Peak.DropRates: Peak Forward Packet Loss Rate of IPPM - VS.IPPM.Rtt.Means: Average RTT Delay of IPPM - VS.IPPM.MaxRttDelay: Maximum RTT Delay of IPPM - DSP IPPMLNK: An OM command that queries the information on the received and transmitted packets of the IP PM link



Transmission - Abis Interface (6) Capacity optimisation methodology  Abis over TDM 1. In case of Fix Abis, if Utilisation meets the set thresholds (e.g. 80%) then check Fix Abis configuration: - MPMODE: Multiplexing Mode: Specifies the multiplexing mode of signalling timeslots over Abis. Values: 1:1, 2:1, 3:1, 4:1, 5:1, 6:1, 16K. (Certain rules apply for each multiplexing mode, check parameter description first.) Increase of MPMODE, e.g. from 1:1 to 4:1 can free additional Abis TS for TCHs. Modes 5:1 and 6:1 can only be used in Flex Abis. - FIXAPALT and FIXABISPRILDTHRED (BSC6000): these two parameters decide the load threshold on Fix Abis above which HR channels are assigned; consider reduction of the parameters in order to assign HR channels earlier. - FLEXABISMODE: Flex Abis Mode: Specifies the working mode of Abis. Values: FIX_ABIS, FLEX_ABIS, SEMI_ABIS (SemiSolid indicates that the Fix Abis is used in the BTS. All its upper-level BTSs use Flex Abis.) - If the highest Multiplexing Mode is used and still Utilisation is high, check the transmission plan: re-arrange chains and sites if possible. - If Utilisation is constantly very high (close to 100%), a change in Abis transmission mode from Fix to Flex will not brink significant benefits; it will probably result in congestion in Flex Abis mode. - Add E1. 2. In case Flex Abis is in use, if blocking counters are not null or utilisation counters are high, check Flex Abis configuration: - MPMODE: check if there is room to increase the value to 6:1, thus freeing some additional TSs for TCHs. - FLEXAPALT and FLEXABISPRILDTHRED (BSC6000): these two parameters decide the load threshold on Flex Abis above which HR channels are assigned; consider reduction of the parameters in order to assign HR channels earlier. - TDMCONGTH (BSC6900): Congestion remain ratio. If the ratio of available TDM bandwidth is less than or equal to this value, congestion control (LDR - see next slide for brief description) is triggered. Usual value 15%. - Check the transmission plan: re-arrange chains and sites if possible. - In case that only a portion of the E1 timeslots are flex TSs, investigate the possibility to increase the percentage of flex TSs in use before adding a new E1. - Add E1. For internal use 30 © Nokia Siemens Networks


Transmission - Abis Interface (7) Capacity optimisation methodology  Abis over IP 1. In case of Abis over IP, if blocking counters are not null: check congestion trigger and clear thresholds in the logical port (BSC6000): - LPNCONGBW: Congestion Bandwidth Threshold [%], usual value 85%. Above this value LDR actions are triggered. - LPNCONGCLRBW: Congestion Clear Bandwidth Threshold [%], usual value 75%. Below this value LDR stops. check congestion trigger and clear thresholds in the IP Path (BSC6900): - FWDCONGBW / BWDCONGBW: Forward / Backward congestion threshold, usual value 85%. - FWDCONGCLRBW / BWDCONGCLRBW: Forward / Backward congestion clear threshold, usual value 75%. LDR (Load Reshuffling algorithm) performs some actions in order to relieve congestion. These actions are: 1. PS service rate decrease 2. Allocate half-rate channels to newly-accessed MSs requesting CS services 3. Limit AMR rate for MSs in CS services 4. Handover of MSs in CS services from the full-rate channel to the half-rate channel The order of actions to be performed is controlled through the MML command: SET LDR 2. Check if ABIS_MUX feature is enabled. ABIS_MUX improves IP transmission efficiency: - ABISMUXFLAG (BSC6000): Abis MUX Global Enable Switch; value=ENABLE - MUXTYPE (BSC6900): IP MUX Type; value=ABISMUX 3. Increase Abis bandwidth if nothing of the above improves the situation.



Transmission - A Interface (1)  A Transmission Modes The following transmission modes can be used for A interface: • A over TDM A over TDM indicates that the TDM transmission is used on the A interface. In TDM-based networking mode, the BSC6900 and the MSC/MGW communicate with each other through the SDH/PDH network. • A over IP A over IP indicates that layer 3 of the A interface protocol stack uses the IP protocol. In IP-based networking mode, the BSC6900 and the MSC/MGW communicate with each other through the IP network.



Transmission - A Interface (2)  A over TDM – Performance Monitoring BLOCKING: - A3050: CELL_CM_SERV_REJ_CONG: Number of rejections of Service Requests due to congestion - A3129E: CELL_ASS_FAIL_NO_CIC: Number of ASS FAIL messages sent by the BSC to the MSC when the BSC receives an ASS REQ message that carries an unavailable A interface CIC - H342E:CELL_INTERBSC_INCELL_HO_FAIL_NO_CIC: Number of failed incoming external inter-cell handovers because the specified CIC on the A interface is unavailable, after the BSC receives the HO REQ from the MSC - H362E:CELL_INTERRAN_INCELL_HO_FAIL_NO_CIC: Number of failed incoming inter-RAT inter-cell handovers because the specified CIC on the A interface is unavailable, after the BSC receives the HO REQ from the MSC UTILISATION: - A Utilisation = ([AL0055:BS_A_INTF_BUSY_CIC_AVG]/([AL0050:BS_A_INTF_UNINST_CIC_AVG]+[AL0051:BS_A_INTF_FAIL_CIC_AVG]+ [AL0052:BS_A_INTF_MTN_CIC_AVG]+[AL0053:BS_A_INTF_CONG_CIC_AVG]+[AL0054:BS_A_INTF_IDLE_CIC_AVG]+[AL 0055:BS_A_INTF_BUSY_CIC_AVG]+[AL0089:BS_A_INTF_PEER_UNINST_CIC_AVG]))*{100} Mean number of circuits in busy state over mean number of circuits in all states in A interface in a granularity period. The possible states of an A interface circuit are: - Uninstalled - Faulty - Maintenance - Blocked - Idle - Busy - Uninstalled peer circuit - AL00n:BS_A_INTFACE_BUSY_TSn_SUMAVR: Average busy time of timeslot n during granularity period, 01 <= n <= 31 (TS 0 for synchronisation is not measured) - The MML command LST AE1T1 can be used to display all circuits in A interface. For internal use 33 © Nokia Siemens Networks


Transmission - A Interface (3)  A over IP – Performance Monitoring  BSC6000 BLOCKING: - Congestion Rate = ([L01057:BSC_LGCPORT_TXDROPPACKETS]/([L01057:BSC_LGCPORT_TXDROPPACKETS]+[L01055:BSC_LGCP ORT_TXPACKETS]))*{100} - L01057:BSC_LGCPORT_TXDROPPACKETS: Number of Packets Discarded on a Logical Port due to congestion - L01055:BSC_LGCPORT_TXPACKETS: Number of Packets Sent on a Logical Port UTILISATION: - L01061:BSC_LGCPORT_MEAN_TXRATE: Average Throughput of Packets Sent on a Logical Port (Kbps) - BANDWIDTH: Bandwidth of the Logical Port [32kbps] (ADD LOGICALPORT)  BSC6900 BLOCKING: - VS.ANI.IP.FailResAllocForBwLimit: Number of Failed Resource Allocations Due to Insufficient Bandwidth on the IP Transport Adjacent Node - VS.IPPATH.Fwd.Cong: Number of Forward Congestions on the IP Path - VS.IPPATH.Fwd.Cong.Dur: Duration of Forward Congestion on the IP Path - VS.IPPATH.Bwd.Cong: Number of Backward Congestions on the IP Path - VS.IPPATH.Bwd.Cong.Dur: Duration of Backward Congestion on the IP PATH UTILISATION: - OS.ANI.IP.AllocedFwd: IP Path Forward Bandwidth Allocated to IP Transport Adjacent Node - OS.ANI.IP.AllocedBwd: IP Path Backward Bandwidth Allocated to IP Transport Adjacent Node - TXBW: Forward Bandwidth: Transmit bandwidth of IP path (ADD IPPATH) - RXBW: Backward Bandwidth: Receive bandwidth of IP path (ADD IPPATH) For internal use 34 © Nokia Siemens Networks


Transmission - A Interface (4) Capacity optimisation methodology  A over TDM In case of A over TDM, if blocking counters are not null, or utilisation is high (e.g. above 80%): - Check state of circuits: DSP ACIC. Note number of circuits that are not idle or busy (e.g. faulty, uninstalled, etc.) and take proper actions to repair them (e.g. reset a circuit), thus freeing more A resources. - Add E1.  A over IP 1. In case of A over IP, if congestion appears or utilisation is high, check congestion trigger and clear thresholds in the logical port (BSC6000): - LPNCONGBW: Congestion Bandwidth Threshold [%], usual value 85%. - LPNCONGCLRBW: Congestion Clear Bandwidth Threshold [%], usual value 75%. check congestion trigger and clear thresholds in the IP Path (BSC6900): - FWDCONGBW / BWDCONGBW: Forward / Backward congestion threshold, usual value 85%. - FWDCONGCLRBW / BWDCONGCLRBW: Forward / Backward congestion clear threshold, usual value 75%. 2. Check if it is possible to activate the following features to increase bandwidth efficiency (BSC6900): - UDP multiplexing: enables multiple RTP packets to be multiplexed in one UDP packet, thus reducing the overhead of the UDP/IP/L2/L1 header and increasing the transmission efficiency. Caution: UDP_MUX improves the transmission efficiency but increases the transmission delay. Relative parameter: MUXTYPE (value=UDPMUX) - Header compression: used to reduce the frame header overhead on PPP links, thus improving bandwidth efficiency. Relative parameters: ACFC (value=Enable), PFC (value=Enable), IPHC (value=UDP/IP_HC or RTP/UDP/IP_HC). 3. Increase A interface bandwidth






BSC6000 - Boards 

CPU Usage Measurements are used to measure the CPU usage of BSC boards and the GBAM.

BLOCKING: - R9703:CPU_USG_MAX: Maximum CPU Usage UTILIZATION: - AR9702:CPU_USG_AVG: Mean CPU Usage



BSC6900 (R11 boards) - DPU Board BLOCKING: - VS.DPU.CPULOAD.MAX: Maximum CPU Usage of the DPU. - VS.DPU.CPULOAD.OVER: Rate of the period in which the CPU Usage of the DPU exceeds the Alarm Threshold. UTILIZATION: - VS.DPU.CPULOAD.MEAN: Average CPU Usage of the DPU. - VS.DPU.CPULOAD.LESS: Rate of the period in which the CPU Usage of the DPU is lower than the Alarm Threshold. FUNCTIONS: - DPUc: The DPUc board processes GSM voice services and data services. • Provides the speech format conversion and data forwarding functions • Encodes and decodes voice services • Provides the Tandem Free Operation (TFO) function • Provides the voice enhancement function • Detects voice faults automatically - DPUd: The DPUd board processes GSM PS services. • Processes the PS services on up to 1,024 simultaneously active PDCHs where signals are coded in MCS9 • Processes packet links • Detects packet faults automatically



BSC6900 (R11 boards) - GCU Board BLOCKING: - VS.GCU.CPULOAD.MAX: Maximum CPU Usage of the GCU. - VS.GCU.CPULOAD.OVER: Rate of the period in which the CPU Usage of the GCU exceeds the Alarm Threshold. UTILIZATION: - VS.GCU.CPULOAD.MEAN: Average CPU Usage of the GCU. - VS.GCU.CPULOAD.LESS: Rate of the period in which the CPU Usage of the GCU is lower than the Alarm Threshold. FUNCTIONS: The GCUa board provides the synchronization clock signals for the system. • Traces, generates, and maintains the synchronization clock • The standby GCUa board traces the clock phase of the active GCUa board. This ensures the smooth output of the clock phase in the case of active/standby switchover.



BSC6900 (R11 boards) - INT Boards BLOCKING: - VS.INT.CPULOAD.MAX: Maximum CPU Usage of the INT. - VS.INT.CPULOAD.OVER: Rate of the period in which the CPU Usage of the INT exceeds the Alarm Threshold. UTILIZATION: - VS.INT.CPULOAD.MEAN: Average CPU Usage of the INT. - VS.INT.CPULOAD.LESS: Rate of the period in which the CPU Usage of the INT is lower than the Alarm Threshold. FUNCTIONS: The INT (Interface) boards provide connectivity to BSC over Abis, Ater, A, Pb, Gb interfaces. INT board can be: • EIUa board: provides E1/T1 transmission over A, Abis, Ater, and Pb interfaces. Provides 32 E1 ports. • FG2a board: provides IP over Ethernet transmission for Abis, A and Gb interfaces. Provides 8 FE and 2 GE ports. • FG2c board: provides IP over Ethernet transmission for Abis, A and Gb interfaces. Provides 12 FE and 4 GE ports. • GOUa board: provides optical channels of IP over Ethernet for Abis and A interfaces. Provides 2 optical GE ports. • GOUc board: provides optical channels of IP over Ethernet for Abis, A and Gb interfaces. Provides 4 optical GE ports. • OIUa board: provides STM-1 transmission over Abis, Ater, A and Pb interfaces. Provides 1 STM-1 port for TDM transmission. • PEUa board: provides E1/T1 transmission for Abis and Gb interfaces. Provides 32 channels of IP over PPP/MLPPP over E1/T1. • POUa board: provides IP over channelized STM-1/OC-3 transmission for Abis, Ater, A, Pb, Gb interfaces. Provides 4 optical channelized STM-1/OC-3 ports for IP transmission.



BSC6900 (R11 boards) - SCU Board BLOCKING: - VS.SCU.CPULOAD.MAX: Maximum CPU Usage of the SCU. - VS.SCU.CPULOAD.OVER: Rate of the period in which the CPU Usage of the SCU exceeds the Alarm Threshold. UTILIZATION: - VS.SCU.CPULOAD.MEAN: Average CPU Usage of the SCU. - VS.SCU.CPULOAD.LESS: Rate of the period in which the CPU Usage of the SCU is lower than the Alarm Threshold. FUNCTIONS: The SCUa board provides the maintenance management and GE switching platform for the subrack in which it is located. It performs the following functions: • Provides the maintenance management function • Provides configuration and maintenance of a subrack or of the entire BSC6900 • Monitors the power supply, fans, and environment of the cabinet • Supports the port trunking function • Supports the active/standby switchover • Enables inter-subrack connections • Provides a total switching capacity of 60 Gbit/s • Distributes clock signals and RFN signals for the BSC6900



BSC6900 (R11 boards) - TNU Board BLOCKING: - VS.TNU.CPULOAD.MAX: Maximum CPU Usage of the SCU. - VS. TNU.CPULOAD.OVER: Rate of the period in which the CPU Usage of the SCU exceeds the Alarm Threshold. UTILIZATION: - VS. TNU.CPULOAD.MEAN: Average CPU Usage of the SCU. - VS. TNU.CPULOAD.LESS: Rate of the period in which the CPU Usage of the SCU is lower than the Alarm Threshold. FUNCTIONS: The TNUa board provides the TDM switching and serves as the switching center for the CS services of the entire system. It performs the following functions: • Provides 128 kbit/s x 128 kbit/s TDM switching • Allocates the TDM network resources



BSC6900 (R11 boards) - XPU Board BLOCKING: - VS.XPU.CPULOAD.MAX: Maximum CPU Usage of the XPU. - VS.XPU.CPULOAD.OVER: Rate of the period in which the CPU Usage of the XPU exceeds the Alarm Threshold. UTILIZATION: - VS.XPU.CPULOAD.MEAN: Average CPU Usage of the XPU. - VS.XPU.CPULOAD.LESS: Rate of the period in which the CPU Usage of the XPU is lower than the Alarm Threshold. FUNCTIONS: Loaded with different software, the XPU board is functionally divided into main control XPU board and non-main control XPU board. The main control XPU board is used to manage the GSM user plane resources, control plane resources, and transmission resources in the system and process the GSM services on the control plane. The non-main control XPU board is used to process the GSM services on the control plane. • Managing the user plane resources; managing the load sharing of the user plane resources between subracks • Maintaining the load of the control plane within the subrack; exchanging the load information on the control planes between subracks • Providing functions such as the logical main control function of the BSC6900, the IMSI-RNTI maintenance and query, and the IMSI-CNid maintenance and query • Forwarding the RRC connection request message to implement the sharing of user plane resources and sharing of control plane resources in the BSC6900 • Processing upper-layer signalling over the A, Um, Abis and Ater interfaces • Processing transport layer signalling • Allocating and managing the various resources that are necessary for service setup, and establishing signalling and service • connections • Processing RFN signalling - XPUa: Main control XPUa has 4 logical subsystems. Non-main control XPUa has 4 logical subsystems. - XPUb: Main control XPUa has 8 logical subsystems. Non-main control XPUa has 8 logical subsystems. For internal use 43 © Nokia Siemens Networks


Transcoder There are 2 configuration modes of Transcoders in Huawei BSS: 1. BM/TC Separated: Transcoder is a separate subrack (GTCS) which can be located either on the BSC side or on the MSC side. Ater interface should be configured. 2. BM/TC Combined: Transcoder functions are performed by specific boards installed in BM subracks (GMPS/GEPS). Ater interface does not need to be configured. UTILIZATION: - AR0755:BS_RES_BUSY_TC_AVG: Mean Number of Busy TC Resources - AR0754:BS_RES_IDLE_TC_AVG: Mean Number of Idle TC Resources - AR0751:BS_RES_FAIL_TC_AVG: Mean Number of Faulty TC Resources



THANK YOU



179844550 2G Huawei Capacity Optimization Process Pptx

Recommend Documents