Isolating 5G NR Throughput in an NSA Deployment

Isolating 5G NR Throughput in an NSA Deployment

Is it possible to extract plots from LMA specifically for 5G NR throughput only?


Short Answer

Yes. LMA captures all data layers simultaneously — from the Application layer down to the Physical layer of every component carrier (CC). While NSA mode combines LTE and NR traffic at the upper layers, LMA allows you to peel back each layer independently and isolate NR-only throughput at both the protocol and physical levels.

 

Background: How 5G NSA (EN-DC) Works

In 5G Non-Standalone (NSA) mode — also known as E-UTRAN/NR Dual Connectivity (EN-DC) — the device maintains two simultaneous radio connections:

•        An LTE anchor (the Master Node, or MN) that handles control plane signaling and provides the primary data path.

•        A 5G NR secondary node (SN) that supplements the LTE anchor with additional downlink (and sometimes uplink) capacity.

 

At the top of the stack (Application layer), the user sees a single unified data stream. The operating system and application have no visibility into how the traffic is split between LTE and NR radios — they receive combined throughput as if from a single connection. As you move down the protocol stack, however, the two radio paths gradually separate. By the time you reach the Physical layer, each component carrier — whether LTE or NR — has its own independent throughput metrics.

 

This architecture is what makes NSA powerful for operators: they can boost peak data rates by aggregating LTE and NR capacity without waiting for a full standalone 5G core. It also means that measuring "5G NR throughput" requires careful attention to which layer of the stack you are reading from.

 

The LMA Data Layer Hierarchy

LMA records measurements at every layer of the 3GPP user-plane stack in real time. The table below maps each protocol layer to the KPI group in LMA, describes what is captured, and indicates whether the measurement reflects combined LTE+NR traffic or can be split by radio access technology.

 

Protocol Layer

KPI Group

What LMA Captures

Network Scope

Application

App Throughput

Total DL/UL bytes/sec delivered to the user application

LTE + NR (combined)

PDCP

PDCP Layer

Reassembled SDUs after header decompression; combined NR+LTE flow in EN-DC

LTE + NR

RLC

RLC Layer

Segmented/reassembled PDUs; first layer where NR and LTE traffic can be viewed separately

LTE  |  NR

MAC

MAC Layer

Transport block scheduling; per-CC allocation visible here

LTE  |  NR

Physical (PHY)

PDSCH / PUSCH

Actual air-interface throughput per component carrier (CC); includes modulation/coding scheme (MCS), MIMO layers, and resource block utilization

LTE CC(s)  |  NR CC(s)

 

The key insight is that the split between LTE and NR traffic first becomes visible at the RLC layer, and reaches full component-carrier granularity at the Physical layer. The sections below walk through each level with the corresponding LMA KPI screenshots.

 

Layer 1 — Application Throughput (Combined LTE + NR)

The Application layer reflects the total data rate delivered to the user's application — for example, the bytes-per-second seen by a speed-test client or a streaming video buffer. In NSA mode, this number represents the sum of everything the device is receiving across all active radio paths. It is the highest-level KPI and does not distinguish between LTE and NR contributions.

 

In LMA, the Application throughput KPI is found in the group shown below:

 

Figure 1: LMA Application Layer KPI — total DL/UL throughput delivered to the user application (LTE + NR combined).

Layer 2 — PDCP / RLC / MAC / PDSCH (Combined Network View)

Moving down from the application, LMA provides a group of KPIs that span PDCP, RLC, MAC, and the PDSCH transport channel. At this level the display still shows LTE and NR throughput combined — this mirrors the EN-DC bearer model, where a single PDCP entity manages packet flow across both the LTE Master Node and the NR Secondary Node.

 

This view is useful for:

•        Confirming total protocol-layer throughput against the application-layer reading.

•        Identifying overhead losses between the application and the air interface.

•        Verifying that EN-DC dual connectivity is active (both nodes contributing to the PDCP flow).

 

Figure 2: LMA PDCP / RLC / MAC / PDSCH KPI group — combined LTE and NR throughput across the full protocol stack.

Layer 3 — Isolating 5G NR Throughput

To answer the customer's specific request, LMA provides a dedicated KPI view that isolates NR throughput from the combined stream. This separation occurs at the RLC/MAC boundary, where the device maintains separate logical channels for the LTE Master Node and the NR Secondary Node.

 

Selecting the NR-specific KPI group produces a plot that shows only the 5G NR contribution to the total data flow — giving the clean NR-only picture the customer is looking for:

 

Figure 3: LMA NR-isolated throughput view — 5G NR DL/UL throughput separated from the LTE anchor contribution.

This is the primary deliverable for customers who need to report NR throughput independently. It can be exported directly from LMA as a plot or as tabular data for inclusion in a test report.

 

Layer 4 — Component Carrier Breakdown (Per-CC Throughput)

For deployments using carrier aggregation — multiple LTE CCs, multiple NR CCs, or a combination — LMA supports a further level of granularity: throughput broken down by individual component carrier. This is found in the Physical layer KPI group and provides the most detailed view of how the air interface is performing.

 

Per-CC analysis is particularly useful when:

•        The NR deployment uses multiple NR CCs (e.g., an NR anchor plus a supplemental downlink, or two NR CCs in carrier aggregation).

•        You need to assess the relative contribution of each band to the total throughput.

•        Troubleshooting uneven load distribution or unexpected CC deactivation events.

 

Figure 4: LMA per-component-carrier throughput breakdown — each LTE and NR CC shown as a separate trace for maximum granularity.

Layer 5 — Full Protocol Stack Summary

Once you have examined each layer independently, LMA can also present a combined summary view that overlays all layers — Application, PDCP, RLC, MAC, and PHY — on a common time axis. This panel is useful for:

•        Validating data-path integrity (each layer should track closely; large gaps indicate protocol overhead or retransmissions).

•        Confirming the relationship between physical-layer throughput and the application-layer experience.

•        Providing a comprehensive snapshot for reporting or customer review.

 

Figure 5: LMA full protocol stack summary — Application through PHY layers on a common time axis.

Real-World Example: Multi-CC NSA Download

The following screenshot shows a real LMA capture of a downlink session in an NSA configuration with multiple NR component carriers and a concurrent LTE anchor. The physical-layer traces for each CC are shown independently, allowing direct comparison of their individual throughput contributions and the additive total delivered to the user.

 

Figure 6: Real-world LMA capture — multi-CC NSA download showing physical throughput on multiple NR and LTE component carriers simultaneously.

In this type of plot you can clearly see which component carriers are active at any moment, how their individual throughput levels vary over time, and how they sum to produce the aggregate rate seen at the application layer.

 

 

Step-by-Step: Extracting NR-Only Plots from LMA

1.     Open your LMA project file captured during the AT&T NSA test session.

2.     Navigate to the Application Layer KPI group to confirm total combined throughput (Figure 1).

3.     Review the PDCP/RLC/MAC/PDSCH combined group to verify protocol-layer performance (Figure 2).

4.     Select the NR-isolated throughput KPI group to extract the 5G NR-only plots the customer has requested (Figure 3).

5.     If per-CC detail is needed, open the Physical layer CC breakdown view (Figure 4) and export individual CC traces.

6.     Use the full-stack summary panel (Figure 5) for a consolidated view suitable for executive or customer reporting.

 

All of these views are generated from the same LMA capture file — no re-testing is required. Each KPI group can be exported as an image, a CSV, or included directly in an auto-generated LMA report.

 

    • Related Articles

    • What is RMA and how is it used for Network Testing

      WirelessMetrix Product Description Remote Monitoring App (RMA) Application-Layer Test Controller for the LinkMaster Platform February 2026 — For distribution to partners and customers 1. Overview The Remote Monitoring App (RMA) is an Android ...

    • Wi-Fi Coverage Testing: Heat Maps vs. Point-Data Surveys

      WirelessMetrix Support | Frequently Asked Questions Wi-Fi Coverage Testing: Heat Maps vs. Point-Data Surveys Category: Wi-Fi / WLAN Testing | Product: DASView Walk-Test Software Question Can WirelessMetrix provide heat map drawings showing wireless ...

    • What does GOOD/BAD result mean during iPerf test?

      When running a iPerf test, you may come across the test results GOOD / BAD These results evaluate whether the throughput is above (GOOD) or below (BAD) a threshold set on the Plan Settings tab. In this example, the DL speeds are 100 Mbps, which is ...

    • RAT Indicator in LMA

      Question:  In LMA, is there a screen where it would show RAT information/changes for NR and LTE? Similar to what QXDM has (RAT indicator screen) Answer:  Yes, you can find this parameter under "Serving Network":

    • How to filter map data in LMA

      This is a video tutorial covering the same topic as this article. When analyzing tests in LMA, you may want to filter map data to only show certain values, such as when the signal quality is poor or where the throughput is above a certain threshold. ...