MobileInsight 5.0 release¶
July 1, 2020
we are thrilled to announce the release of MobileInsight v5.0.0. The main updates include
Migration to Python3: Starting from 5.0.0, all the MobileInsight codes will be written in Python 3. Please update your customized MobileInsight codes accordingly if they are still based on Python 2.
KPI analyzers: MobileInsight 5.0 now supports KPI analyzers, a new set of user-friendly cellular analytics APIs. KPI analyzers track the runtime Key Performance Indicators (KPIs) of the cellular network. They support the real-time KPI query on per-cell/user basis. Please read the tutorial of KPI analyzers and labelled dataset examples to learn how to use them,
MobileInsight v5.0.0 supports the following standardized KPIs (standardized in 3GPP TS32.450)
Accessibility:
RRC connection establishment success rate
Attach success rate
Dedicated bearer setup success rate
Service request success rate
Mobility:
Tracking area update success rate
Intra RAT handover success rate
Retainability:
Abnormal RRC connection release rate
Integrity:
IP throughput
MobileInsight v5.0.0 also supports the following experimental KPI analyzers:
Mobility:
Handover disruption time
Handover prediction
Handover head of line blocking
Data plane (L1/L2):
Downlink PDCP packet loss
Uplink PDCP packet loss
Block Error Ratio (BLER)
Network latency tracking and breakdown: MobileInsight v5.0.0 offers analyzers that can track and break down latency components for LTE uplink and downlink data packets.
The uplink latency analyzer can track the following latency components for each packet
UL scheduling latency
UL transmission latency
UL retransmission latency
DRX-doze latency
SR waiting latency
SR to grant latency
The downlink latency analyzer can track the following latency components for each packet
Statistics of retransmission
Statistics of t-reordering timer configuration
Statistics of Reordering + UL scheduling for NACK