Inferensys

Glossary

Minimization of Drive Tests (MDT)

A 3GPP standardized feature that leverages commercial user equipment to collect radio measurements and location data, replacing costly manual drive tests for network optimization and coverage verification.
Data engineer managing feature store on laptop, feature definitions visible, casual data engineering session.
3GPP STANDARDIZED FEATURE

What is Minimization of Drive Tests (MDT)?

A 3GPP standardized feature that leverages commercial user equipment to collect radio measurements and location data, replacing costly manual drive tests for network optimization and coverage verification.

Minimization of Drive Tests (MDT) is a 3GPP-defined automation feature that utilizes commercial User Equipment (UE) to gather radio measurement data and associated location information, effectively replacing the need for expensive, labor-intensive manual drive testing. The core mechanism involves the network configuring capable UEs to perform and report measurements such as Reference Signal Received Power (RSRP), Reference Signal Received Quality (RSRQ), and power headroom while in connected or idle mode.

MDT operates in two primary modes: Immediate MDT, where measurements are reported in real-time during an active connection, and Logged MDT, where a UE in idle mode stores measurements and reports them upon reconnection. This data is streamed to a Trace Collection Entity (TCE) via the Operations, Administration, and Maintenance (OAM) system, providing operators with a granular, geographically accurate view of coverage holes, interference, and mobility issues across the entire network footprint.

3GPP STANDARDIZED FEATURE

Key Features of MDT

Minimization of Drive Tests (MDT) leverages commercial UEs to collect radio measurements and location data, replacing costly manual drive tests for network optimization and coverage verification.

01

Immediate MDT

Immediate MDT collects measurements from UEs in RRC_CONNECTED state and reports them in real-time to the network. This mode is ideal for troubleshooting active sessions and capturing detailed radio conditions during data transfers.

  • Measurement triggers: periodic, event-based (e.g., A2 serving cell threshold)
  • Includes detailed L1/L3 RSRP/RSRQ and power headroom reports
  • Location information derived from GNSS or RF fingerprinting
  • Used for real-time coverage hole detection and handover boundary analysis
RRC_CONNECTED
UE State Required
02

Logged MDT

Logged MDT allows UEs in RRC_IDLE or RRC_INACTIVE state to record measurements according to a configuration received while connected. The UE stores data and reports it when it next establishes a connection.

  • Configuration includes logging duration, interval, and area scope (cell list or tracking area)
  • Records serving cell and intra-frequency neighbor measurements
  • Timestamps and location stamps each log entry
  • Critical for mapping idle-mode coverage and detecting sleeping cell issues
RRC_IDLE/INACTIVE
UE State Required
03

Management-Based MDT

In Management-Based MDT, the network operator selects specific UEs for measurement collection based on criteria such as IMSI, IMEI, or area scope. This is initiated from the Element Management System (EMS) or Network Management System (NMS).

  • Uses Trace Function activation via the Home Subscriber Server (HSS)
  • Supports both Immediate and Logged MDT modes
  • Enables targeted troubleshooting for VIP users or specific device models
  • Selection criteria: IMSI ranges, IMEI-SV ranges, or geographic area
Trace-Based
Activation Method
04

Signaling-Based MDT

Signaling-Based MDT activates measurement collection for a specific UE through control plane signaling, initiated by the Mobility Management Entity (MME). This targets an individual subscriber for detailed session analysis.

  • Activated via S1-AP Initial Context Setup or UE Context Modification procedures
  • Tied to a specific UE context, not a general area
  • Enables per-subscriber QoS verification and mobility analysis
  • Complements Management-Based MDT for granular, subscriber-level diagnostics
Per-UE
Granularity
05

Location Information Mechanisms

MDT enriches radio measurements with location data to geolocate coverage problems. Multiple positioning methods are supported, balancing accuracy against UE power consumption.

  • GNSS (GPS/GLONASS/Galileo): Highest accuracy, available only with clear sky view
  • RF Fingerprinting: Pattern-matches measured cell IDs and signal strengths against a predicted database
  • Enhanced Cell ID (E-CID): Uses timing advance and angle of arrival for improved cell-level positioning
  • Location data is privacy-controlled; operators must anonymize or obtain consent per local regulations
GNSS
Highest Accuracy Method
06

MDT Configuration Parameters

The network configures MDT behavior through a set of standardized parameters transmitted to the UE via RRC signaling. These parameters define what, when, and how measurements are collected.

  • LoggingInterval: Time between consecutive measurement samples (e.g., 1.28s, 2.56s)
  • LoggingDuration: Total time the UE should record measurements (e.g., 10min, 60min)
  • AreaConfiguration: Defines the geographic scope using cell IDs or tracking area codes
  • TraceReference: Unique identifier linking the MDT session to the management system
  • TargetMBSFN-AreaList: Optionally restricts logging to specific MBSFN areas for multicast optimization
RRC Reconfiguration
Delivery Method
COMPARATIVE ANALYSIS

MDT vs. Traditional Drive Tests

A technical comparison of 3GPP Minimization of Drive Tests (MDT) against conventional manual drive testing methodologies for radio network optimization.

FeatureMDTTraditional Drive Test

Data Collection Source

Commercial UEs in active mode

Dedicated test equipment and vehicles

Geographic Coverage

Indoor, outdoor, and private areas where subscribers operate

Limited to public roads and accessible outdoor locations

Collection Duration

Continuous, 24/7 during normal UE operation

Scheduled, time-limited measurement campaigns

Operational Expenditure

Minimal incremental cost; leverages existing infrastructure

High cost: vehicles, fuel, personnel, and specialized equipment

Location Accuracy

GNSS-based when available; RF fingerprinting otherwise

High-precision GNSS with post-processing

Measurement Density

Sparse per UE but aggregated across millions of devices

Dense along drive route but spatially constrained

User-Experience Correlation

Directly reflects actual subscriber perceived quality

Proxy measurement; may not represent real user conditions

3GPP Standardization

Standardized in Rel-10 (LTE) and enhanced in Rel-16 (NR)

Not standardized; vendor-proprietary tools and formats

MDT EXPLAINED

Frequently Asked Questions

Clear, technical answers to the most common questions about 3GPP's Minimization of Drive Tests feature, covering architecture, privacy, and operational impact.

Minimization of Drive Tests (MDT) is a 3GPP standardized feature (introduced in Release 10) that leverages commercial User Equipment (UE) to collect radio measurements and location data, effectively replacing costly manual drive tests for network optimization. It works by configuring selected UEs in the network to perform specific measurement logging tasks while in connected or idle mode. These UEs record parameters like Reference Signal Received Power (RSRP), Reference Signal Received Quality (RSRQ), and timing advance, stamping them with precise location coordinates from the device's GNSS. The collected logs are then reported back to the network via RRC signaling or a Trace Collection Entity (TCE), where a Trace Integration Function processes the data for analysis by Self-Organizing Network (SON) modules or operations teams. This provides a continuous, geographically distributed view of the user experience without the capital expenditure and carbon footprint of deploying dedicated drive-test vehicles.

Prasad Kumkar

About the author

Prasad Kumkar

CEO & MD, Inference Systems

Prasad Kumkar is the CEO & MD of Inference Systems and writes about AI systems architecture, LLM infrastructure, model serving, evaluation, and production deployment. Over 5+ years, he has worked across computer vision models, L5 autonomous vehicle systems, and LLM research, with a focus on taking complex AI ideas into real-world engineering systems.

His work and writing cover AI systems, large language models, AI agents, multimodal systems, autonomous systems, inference optimization, RAG, evaluation, and production AI engineering.