A TDD carrier may have plenty of downlink capacity while a UE still waits for an uplink opportunity. Giving another complete slot to uplink reduces downlink time. SBFD offers another choice: gNB receives uplink on one frequency portion while it transmits downlink on another portion of the same carrier.

The basic picture is simple, but three details matter immediately. The full-duplex operation is at the gNB. UE remains half duplex. Also, an SBFD subband is not a replacement for the UE's BWP: both restrictions apply when UE transmits or receives.

This note uses Release 19 TS 38.300, TS 38.331, TS 38.213 and TS 38.214.  The relevant SBFD clauses and RRC definitions were reviewed, rather than the complete specifications. The resource grids and calculations are teaching examples, not captured network configurations.

• Executive Summary
• What is SBFD ?
  • TDD, FDD and full duplex
  • Why add an uplink subband ?
• How are time and frequency resources divided ?
  • SBFD symbols and subbands
  • The active-BWP intersection
• How does UE know the SBFD resources ?
  • Time-domain configuration
  • Frequency-domain configuration
  • Worked RIV and BWP example
• How does scheduling work ?
  • SBFD and non-SBFD occasions
  • Physical-layer checks
• Can UE perform random access in SBFD ?
• Why is interference difficult ?
  • Self-interference and cross-link interference
  • An illustrative suppression budget
• What should I check in a trace ?
• DCI and MAC CE for SBFD
  • DCI format 1_0 and the RACH occasion indicator
  • LTM command and the RO field
  • MAC CE for semi-persistent CLI measurements
• Technical Challenges
  • Receiver protection and self-interference cancellation
  • Neighboring cells and UE-to-UE interference
  • Guard bandwidth and the UL/DL resource tradeoff
  • Half-duplex UE scheduling and channel constraints
  • Hardware readiness and deployment validation
• RRC Parameters
  • TDD pattern and SBFD interval
  • SCS-specific carrier and subband allocation
  • Dedicated BWP settings
• Reference

Executive Summary

What is SBFD ?

SBFD means Subband Full Duplex. Within a configured SBFD symbol, the gNB uses different subbands for DL and UL. The transmissions overlap in time, but the scheduled frequency resources do not overlap. TS 38.300 clause 23.1 gives the overall behavior. [1]

TDD, FDD and full duplex

The word full duplex can be misleading if the frequency allocation is omitted. SBFD separates UL and DL within one TDD carrier. Same-frequency in-band full duplex would require simultaneous opposite-direction operation on overlapping frequencies, which is a different interference problem.

Why add an uplink subband ?

A DL-heavy pattern can leave long gaps between UL opportunities. SBFD can place UL resources inside some of those gaps. A UE may send its data sooner, or spread an uplink transfer across more time occasions. This is especially interesting when UE transmit power limits the link.

There is a resource cost. Allocating an UL subband reduces the DL frequency resources in those symbols. Guard regions also consume resources. So more UL opportunities do not mean that UL and DL throughput both increase automatically. Scheduling and interference determine whether the exchange is worthwhile. [5]

How are time and frequency resources divided ?

Figure 1 starts with a simple TDD pattern and then adds SBFD to four of its DL slots. The upper and lower blue regions remain DL. The orange region becomes UL at the same time. The grey strips represent illustrative frequency guards, not a mandated guard width.

Figure 1. SBFD adds UL frequency resources during selected DL symbols while retaining DL resources in the same symbols. The diagram is a teaching allocation, not a complete cell configuration.

SBFD symbols and subbands

An SBFD symbol is configured within a DL or flexible symbol of tdd-UL-DL-ConfigurationCommon. Symbols configured as UL by that common pattern remain non-SBFD. The SBFD interval is consecutive and is described by start/end slot and symbol indices. It is not necessarily a whole-slot interval. [3], clause 11.1

In Figure 1, all 14 symbols in slots 2 through 5 are selected. At 30 kHz reference SCS, each slot lasts 0.5 ms, so the illustrated interval lasts 2 ms. A real configuration must also satisfy channel-specific rules, including SSB-related restrictions. The diagram does not allocate SSB, PDCCH, reference signals or grants.

The active-BWP intersection

The subband tells UE which part of the carrier can be used in that direction. The active BWP imposes a second boundary. Except for the specified CLI-measurement case, usable RBs must belong to both sets. A wide BWP does not permit transmission outside the UL subband. [1]

UL frequency region = active UL BWP ∩ UL subband

DL frequency region = active DL BWP ∩ configured DL subband(s)

The intersection is still only an eligible region. UE needs the applicable scheduling or configured-transmission resources before transmitting. If the UL BWP does not overlap the UL subband, there are no eligible UL RBs for that combination.

How does UE know the SBFD resources ?

There are two separate questions: when SBFD applies, and where each subband is located. RRC carries both answers. Start with the cell-common pattern, then inspect the SCS-specific carrier allocation and the UE's dedicated BWP settings. The field names in this section were checked against TS 38.331 V19.4.0. [2], clause 6.3.2

Time-domain configuration

These four SBFD fields are extensions of TDD-UL-DL-Pattern, not independent entries at the top of SIB1. Reading the decoded parent structure matters. A start/end pair without its pattern and reference SCS is not enough to place the interval on a time axis.

Frequency-domain configuration

SCS-SpecificCarrier gives the carrier numerology, offsetToCarrier and carrier width. Its sbfd-Subband-Allocation-r19 field points to SBFD-Subband-Allocation-r19. That structure supplies the location-and-bandwidth values. They are encoded RIVs, not literal bandwidths in RBs.

For these subband RIVs, TS 38.213 clause 11.1 invokes the RIV interpretation with a reference size of 275. UE then adds offsetToCarrier to the decoded starting offset. Using the actual carrier width as the RIV reference size can produce a plausible-looking but wrong allocation. This rule concerns the subband configuration; do not apply it blindly to every PUSCH grant. [3]

Worked RIV and BWP example

Consider a teaching carrier with 100 RBs and offsetToCarrier = 10. Its CRBs run from 10 through 109 at the chosen SCS. Use 20 UL RBs starting at carrier-relative offset 40, with DL regions on either side. These decoded values illustrate the arithmetic; they are not a complete, validated RRC message.

For each length L in this example, L−1 ≤ floor(275/2), so the short-length RIV branch applies. Longer allocations use the other branch of the specification's RIV equation.

RIV = 275 × (L − 1) + S

Now assume the active UL BWP covers CRBs 30–89. Its intersection with the UL subband is CRBs 50–69. Those become BWP-relative PRBs 20–39. If the active DL BWP also covers CRBs 30–89, its usable DL regions are CRBs 30–45 and 74–89.

Keep the coordinate systems separate. The subband RIV gives a carrier-relative offset using the specified 275 reference. offsetToCarrier locates that region relative to Point A. The active BWP then supplies another origin for BWP-relative PRB indexing. A decoded value of 40 is not automatically PRB 40 in the active BWP.

How does scheduling work ?

The SBFD map establishes the resource structure. DCI and configured scheduling select transmissions within that structure. A grant does not give UE permission to ignore its BWP, the SBFD subbands, or simultaneous transmit/receive restrictions.

SBFD and non-SBFD occasions

A transmission or reception sequence can be restricted to one symbol type, or configured to use occasions of both types. The dedicated configuration controls this distinction. It does not turn UE into a full-duplex device.

TS 38.214 clause 6.1.2.2 gives the uplink frequency-allocation rules, including the type-1 case in 6.1.2.2.2. An allocation across symbol types may need a different starting-RB interpretation. Check the configured offset, grant type and DCI format together instead of reusing the non-SBFD PRB range unchanged. [4]

Physical-layer checks

TS 38.213 clause 11.1 keeps the common SBFD/non-SBFD classification from being changed by other information. Dynamic scheduling can select resources, but it does not freely redraw that common map. The clause also restricts UL transmission in SBFD symbols indicated for SSB presence in the active DL BWP. [3]

The receiver must also interpret reference signals in the correct symbol context. For example, PUSCH SRS-resource associations distinguish SBFD and non-SBFD resources in TS 38.214. A valid frequency allocation alone does not establish that the complete transmission is valid.

Can UE perform random access in SBFD ?

Yes, with the supported SBFD random-access configuration. Release 19 supports CBRA and CFRA using four-step RA on SBFD resources. The presence of an UL subband alone does not define a valid PRACH occasion. [1], clause 23.2

A cell selects one of these options. UE support for that option matters. The detailed occasion selection, association and validity rules belong to TS 38.213 clause 8 and the corresponding TS 38.331 configuration. Do not replace those rules with a simple test that the PRACH frequency lies inside the UL subband.

The reviewed stage-2 description excludes Request for Other SI and early UL synchronization with an LTM candidate cell from SBFD RA. It also excludes RA associated with additional PCIs of one serving cell; RACH-based LTM cell switching is supported there only for the intra-DU case. These restrictions matter when deciding whether a particular trigger may use SBFD resources.

Why is interference difficult ?

Frequency separation does not make the transmitter and receiver independent. The gNB's DL signal is much stronger than the UL signal it wants to detect. Leakage, limited isolation and receiver dynamic range can therefore limit the benefit even when the scheduled bands do not overlap.

Figure 2 separates wanted links from unwanted coupling. Self-interference remains inside the serving gNB. Cross-link interference connects opposite-direction operations between nodes. A neighboring gNB and a nearby uplink UE create different victim-receiver problems.

Figure 2. Frequency-separated scheduling still leaves self-interference and cross-link interference to control. Interference strength depends on filtering, geometry, power and neighboring allocations.

Self-interference and cross-link interference

Digital cancellation happens after the RF chain and ADC. It cannot undo overload that has already destroyed the received signal. Some isolation or suppression is therefore needed before that point. Guard bandwidth helps filtering, but it trades away usable resources. There is no single guard width or cancellation number that follows from the definition of SBFD.

An illustrative suppression budget

Assume 46 dBm total DL output and a target residual self-interference of −101 dBm integrated over a 20 MHz UL receiver bandwidth. This gives an end-to-end budget of 147 dB from the stated transmitter reference to the receiver-band residual. It is not a requirement for one cancellation circuit.

Total required suppression in this example = 46 − (−101) = 147 dB

At 290 K and 5 dB receiver noise figure, the approximate noise power over 20 MHz is −174 + 10 log10(20 × 10⁶) + 5 = −96 dBm. A −101 dBm residual is 5 dB below that noise. Adding powers in linear units gives about 1.2 dB noise-plus-interference increase.

The 147 dB includes whatever combination of spectral rejection, isolation and cancellation connects those reference points. Different UL bandwidth, DL power, leakage spectrum or receiver tolerance changes the result. This calculation explains why suppression budgets can be large without treating one study's number as a universal deployment threshold.

What should I check in a trace ?

A useful debug sequence follows the configuration before the grant. Otherwise, a correct DCI decode may still be interpreted against the wrong symbol type, frequency origin or BWP.

1. Check UE SBFD support and the network-selected feature options. A general Release 19 label is not enough to prove every optional behavior.

2. Find the cell-common TDD configuration, reference SCS and SBFD start/end fields. Derive the symbol interval.

3. Find the UL and DL SCS-specific carrier entries. Decode each applicable subband RIV using the specified 275 reference.

4. Add the carrier offset and convert all resources to the same CRB coordinate system.

5. Intersect the subbands with the active UL/DL BWPs. Then apply the scheduled or configured channel resources.

6. Check Configuration 2, RB-offset, repetition and reference-signal settings before comparing SBFD and non-SBFD occasions.

7. For RACH, check the supported RACH option and valid occasion rules. For unexpected errors, separate RF interference from invalid resource configuration.

For the neighboring-cell interference intuition, see the TDD Cross-Link Interference simulator. That simulator uses whole-slot directions and does not implement SBFD subband allocation or same-gNB self-interference.

The CLI measurements themselves started in Release 16, before SBFD. SRS-RSRP and CLI-RSSI, the CLI measurement object and event I1 are described in the CLI and RIM page, together with remote interference management. That page also lists what Release 19 adds for SBFD, such as L1 CLI reporting.

DCI and MAC CE for SBFD

RRC establishes the SBFD time and frequency resources. DCI and MAC CE then control particular transmissions, random-access choices and measurements. So a trace needs both layers: the command tells UE what to do, while the configuration determines which resources that command can use.

DCI format 1_0 and the RACH occasion indicator

For a PDCCH order, DCI format 1_0 can contain a one-bit RACH occasion indicator. The bit is present when the applicable SBFD single or dual RACH configuration is provided; otherwise its length is zero. It selects the first or second PRACH occasions according to Table 7.3.1.2.1-6. If the Random Access Preamble index is all zeros, the field is reserved. [6], clause 7.3.1.2.1

This is the PDCCH-order interpretation of DCI 1_0. Do not look for this field in every ordinary PDSCH scheduling instance of that format. Also, the RACH occasion indicator is distinct from the PRACH association indicator and PRACH resource indicator; those fields have separate presence conditions and purposes.

UE still derives the actual valid PRACH occasions from its configuration and TS 38.213 clause 8. The first/second distinction follows those rules, including the single-versus-dual RACH option. A selected occasion type alone does not identify a complete time/frequency allocation. [3]

LTM command and the RO field

The (Enhanced) LTM Cell Switch Command MAC CE can indicate the occasion type for contention-free random access. Its one-bit RO field selects first PRACH occasions with 0 and second PRACH occasions with 1. In the LTM command, the field is included with the contention-free RA resources when the C flag is set. [7], clauses 5.1.1 and 6.1.3.75

This choice applies to the RA procedure associated with the cell switch. It does not grant arbitrary UL transmission in any DL symbol. The target configuration and SBFD RA restrictions still apply, including the intra-DU limitation described in random access.

MAC CE for semi-persistent CLI measurements

The SP CLI Measurement Resource Set Activation/Deactivation MAC CE controls an RRC-configured measurement set. It identifies the serving cell, DL BWP and set, and carries an activation/deactivation indication. Activation can also supply TCI-state information for the measurement resources. [7], clauses 5.18.41 and 6.1.3.87

This supports cross-link interference measurement. Activating measurements does not activate SBFD itself. Keep the measurement state separate from the common SBFD map and the scheduler's resource assignments when reading a trace.

Technical Challenges

Drawing separate UL and DL subbands is straightforward. Making them work together in a deployed radio is harder. The gNB must receive a weak uplink while transmitting downlink, and the scheduler must turn the added UL opportunities into a useful gain. The following points are engineering implications of the resource and interference constraints discussed above, not additional normative requirements.

Receiver protection and self-interference cancellation

The first problem is the power difference between gNB transmission and reception. Frequency separation helps, but transmitter leakage can still enter the UL receiver. Antenna isolation, filtering and cancellation must work together across the intended power, bandwidth and antenna configurations.

Receiver protection must happen early enough in the RF chain. If the LNA or ADC overloads, later digital cancellation cannot reconstruct the lost signal. A cancellation result measured with one fixed coupling path also does not prove performance across changing beams or environmental conditions. Characterization needs to cover the operating range and the ability to track changes.

The suppression-budget example explains why a large total isolation number can arise. That total includes several mechanisms; it should not be read as the required performance of one canceller. Self-interference and inter-site interference are also identified as practical limitations in the reviewed research abstract [5].

Neighboring cells and UE-to-UE interference

Suppressing the serving gNB's own leakage does not remove interference from another gNB. A neighboring DL transmitter can impair UL reception, particularly when the cells use incompatible time/frequency allocations. Coordination therefore needs to consider the subbands as well as the slot directions.

UE pairing creates another constraint. A nearby UL UE can disturb a DL UE through spectral leakage and limited receiver rejection. Geometry, transmit power and frequency separation all matter. Scheduling a different UE pair or reducing UL power can help, but may reduce capacity or coverage elsewhere.

CLI measurements can inform these decisions, but the measurements consume resources and may become outdated as traffic or UE positions change. Activating the SP CLI measurement set provides observations; it does not by itself choose a safe scheduling allocation.

Guard bandwidth and the UL/DL resource tradeoff

Guard regions make filtering easier, but they remove usable bandwidth. A wider UL subband also leaves less frequency space for DL in the same symbols. The useful question is how much additional UL service the network gains for the DL resources and guard bandwidth it gives up.

More UL occasions can shorten waiting time without increasing peak UL bandwidth. This can help a power-limited UE, but the benefit depends on packet arrivals, load and interference. If little UL traffic is waiting, reserving extra UL resources may offer little return. A semi-static resource map also cannot follow every short traffic burst as freely as individual grants can.

Compare configurations using the same carrier bandwidth and realistic traffic. Count guard resources, retransmissions and DL throughput loss as well as UL throughput. A gain in one direction is not sufficient evidence that the overall allocation is better.

Half-duplex UE scheduling and channel constraints

SBFD gives the gNB simultaneous transmit and receive operation; UE remains half duplex. The scheduler must arrange different UEs or different occasions for opposite directions while respecting each UE's active BWP and configured subbands. [1], clause 23.1

A frequency allocation can be inside the UL subband and still be unusable. Control, reference signals, HARQ feedback and random access have their own timing and validity rules. SSB-related restrictions and the distinction between SBFD and non-SBFD occasions also need checking. [3]

Configuration 2 adds another interpretation step when transmissions span the two occasion types. The scheduler and UE must agree on the applicable RB-offset and reference-signal rules. See SBFD and non-SBFD occasions. These dependencies make a simple throughput-only scheduler model insufficient for validating a real implementation.

Hardware readiness and deployment validation

An RRC-capable implementation does not prove that an existing radio unit can transmit and receive concurrently with adequate isolation. Deployment assessment must include the RF paths, filtering, receiver dynamic range and antenna arrangement. The required hardware changes depend on the radio design.

Validation should progress from conducted RF measurements to multi-cell operation. First measure residual leakage and UL receiver degradation with DL transmission active. Then add neighboring-cell traffic and representative UE locations. Finally, check latency, throughput and retransmissions in both directions, including the power-limited UL cases that motivate SBFD.

RRC Parameters

Following are the ASN.1 definitions behind the SBFD resource map and dedicated BWP behavior. Start with the time pattern, then read the carrier subbands and UE-specific settings. These are selected definitions from TS 38.331 clause 6.3.2, not a complete RRC message or every SBFD-related IE.

TDD pattern and SBFD interval

The common configuration supplies the reference SCS and pattern instances. Each pattern carries its own SBFD boundaries. The extension brackets belong to ASN.1; they do not indicate repeated time intervals.

Following is based on 38.331 v19.4.0 (Release 19)

TDD-UL-DL-ConfigCommon ::=          SEQUENCE {
    referenceSubcarrierSpacing          SubcarrierSpacing,
    pattern1                            TDD-UL-DL-Pattern,
    pattern2                            TDD-UL-DL-Pattern                                                       OPTIONAL, -- Need R
    ...
}

TDD-UL-DL-Pattern ::=               SEQUENCE {
    dl-UL-TransmissionPeriodicity       ENUMERATED {ms0p5, ms0p625, ms1, ms1p25, ms2, ms2p5, ms5, ms10},
    nrofDownlinkSlots                   INTEGER (0..maxNrofSlots),
    nrofDownlinkSymbols                 INTEGER (0..maxNrofSymbols-1),
    nrofUplinkSlots                     INTEGER (0..maxNrofSlots),
    nrofUplinkSymbols                   INTEGER (0..maxNrofSymbols-1),
    ...,
    [[
    dl-UL-TransmissionPeriodicity-v1530     ENUMERATED {ms3, ms4}                                               OPTIONAL -- Need R
    ]],
    [[
    sbfd-StartingSlotIndex-r19          INTEGER (0..maxNrofSlots-1)                                             OPTIONAL, -- Need R
    sbfd-StartingSymbolIndex-r19        INTEGER (0..maxNrofSymbols-1)                                           OPTIONAL, -- Need R
    sbfd-EndingSlotIndex-r19            INTEGER (0..maxNrofSlots-1)                                             OPTIONAL, -- Need R
    sbfd-EndingSymbolIndex-r19          INTEGER (0..maxNrofSymbols-1)                                           OPTIONAL  -- Need R
    ]]
}

Interpret the four SBFD boundary fields using referenceSubcarrierSpacing and the applicable pattern period. See time-domain configuration for their relationship.

SCS-specific carrier and subband allocation

The carrier entry supplies the frequency origin and SCS. Its Release 19 extension adds the subband allocation. Each location-and-bandwidth integer contains an RIV, which must be decoded before drawing the RB range.

Following is based on 38.331 v19.4.0 (Release 19)

SCS-SpecificCarrier ::=             SEQUENCE {
    offsetToCarrier                     INTEGER (0..2199),
    subcarrierSpacing                   SubcarrierSpacing,
    carrierBandwidth                    INTEGER (1..maxNrofPhysicalResourceBlocks),
    ...,
    [[
    txDirectCurrentLocation         INTEGER (0..4095)                                       OPTIONAL            -- Need S
    ]],
    [[
    sbfd-Subband-Allocation-r19     SBFD-Subband-Allocation-r19                             OPTIONAL            -- Need R
    ]]
}

SBFD-Subband-Allocation-r19 ::=             SEQUENCE {
    ul-subbandlocationAndBandwidth-r19          INTEGER (0..37949)                          OPTIONAL,           -- Need R
    firstDL-subbandlocationAndBandwidth-r19     INTEGER (0..37949)                          OPTIONAL,           -- Need R
    secondDL-subbandlocationAndBandwidth-r19    INTEGER (0..37949)                          OPTIONAL,           -- Need R
    ...
}

The UL allocation belongs in the UL carrier entry; the DL allocations belong in the DL carrier entry. Apply the field descriptions and PHY restrictions as well as ASN.1. The worked example separates RIV decoding, carrier offset and BWP intersection.

Dedicated BWP settings

The following excerpts retain the Release 19 extension groups containing Configuration 2. Earlier fields and groups are explicitly omitted. These show the owning IEs and field types; they are not complete, compilable ASN.1 definitions.

Following is based on 38.331 v19.4.0 (Release 19)

BWP-DownlinkDedicated ::= SEQUENCE {
    -- Original root fields omitted in this excerpt.
    ...,
    -- Earlier extension addition groups omitted in this excerpt.
    [[
    sbfd-Config2-Reception-r19          ENUMERATED {enabled}                                              OPTIONAL,   -- Need S
    pathlossOffsetPRACH-DCI-1-0-r19     ENUMERATED { enabled }                                            OPTIONAL,   -- Need R
    prachAssociationDCI-1-0-r19         ENUMERATED { enabled }                                            OPTIONAL    -- Need R
    ]]
}

Following is based on 38.331 v19.4.0 (Release 19)

BWP-UplinkDedicated ::= SEQUENCE {
    -- Original root fields omitted in this excerpt.
    ...,
    -- Earlier extension addition groups omitted in this excerpt.
    [[
    sbfd-Config2-Transmission-r19       ENUMERATED {enabled}                                                    OPTIONAL,   -- Need S
    sbfd-Config2-PUSCH-RB-Offset-r19    INTEGER(0..maxNrofPhysicalResourceBlocks-1)                             OPTIONAL,   -- Need R
    ul-Muting-NonSBFD-Symbol-r19        ENUMERATED {enabled}                                                    OPTIONAL,   -- Need S
    twoTA-Without-MultiDCI-MultiTRP-r19  ENUMERATED {enabled}                                                   OPTIONAL   -- Need R
    ]]
}

The Configuration 2 flags enable the specified operation across SBFD and non-SBFD occasions. The PUSCH RB offset is a separate allocation parameter. See SBFD and non-SBFD occasions for the associated PHY references.

The other fields remain visible because they share these extension groups. Their presence does not make them prerequisites for Configuration 2. The -- Need S and -- Need R comments are specification presence-handling annotations, preserved from the source.

Reference

The first four references define the Release 19 behavior discussed here. The research abstract supports the motivation and interference discussion; its scenario-specific throughput and suppression results are not treated as specification requirements.

1. TS 38.300 V19.4.0 : NR and NG-RAN Overall description; Stage-2. Clauses 23.1 and 23.2.
2. TS 38.331 V19.4.0 : NR RRC protocol specification. Clause 6.3.2 definitions and descriptions for TDD-UL-DL-Pattern, SCS-SpecificCarrier, SBFD-Subband-Allocation and dedicated BWP configuration.
3. TS 38.213 V19.5.0 : NR Physical layer procedures for control. Clauses 11.1 and 8.
4. TS 38.214 V19.5.0 : NR Physical layer procedures for data. SBFD-related PUSCH resource and SRS association rules, including 6.1.2.2.
5. Research abstract, 30 March 2023 : Nokia Bell Labs, Modeling and System-Level Performance Evaluation of Sub-Band Full Duplexing for 5G-Advanced. Abstract reviewed; full paper not used for numerical claims.

6. TS 38.212 V19.4.0 : Multiplexing and channel coding. Reviewed the DCI format 1_0 PDCCH-order fields in clause 7.3.1.2.1.

7. TS 38.321 V19.4.0 : MAC protocol specification. Reviewed RO selection in clause 5.1.1, LTM command fields in 6.1.3.75, and SP CLI measurement control in 5.18.41 and 6.1.3.87.