You are not logged in.
Pages: 1
Hi team,
While extracting packet level information from captured logs, ENTRY TYPE for TX and RX_OFDM length field included (Length in bytes of MPDU; includes MAC header, payload and FCS ) in v1.2.0 of reference design.
i.e When payload=1400B was assigned in log_capture script. Logs recorded 1428B in length field.
--> Was same behavior was followed in v0.96 i.e setting payload=1400B captured 1428B in log information or only 1400B in length field?
Offline
We made a change between those two versions of the reference design:
- In the 1.2.0 version of the reference design, "payload" means "MAC payload", i.e. the "payload" is the bytes after the 802.11 MAC header and before the FCS bytes. You can see this definition here for the AP.
- In the 0.96 version of the reference design, "payload" means "PHY payload", i.e. the total bytes that the PHY transfers. You can see this definition here for the AP.
We felt that "MAC payload" was a better definition for "payload" in the context of the WLAN Exp framework. Feel free to modify this for your needs, but just note that this definition exists in the AP, STA, and IBSS projects (search for "wlan_create_ltg_frame").
Offline
Just to clarify- starting with v1.0 of the ref design the RX_OFDM and TX_LOW "length" fields mirror the LENGTH field in the underlying PHY packet's SIGNAL field. This value includes the MAC header (usually 24 bytes) and FCS (always 4 bytes). This definition is true for all packets, including LTG, Ethernet, and management. We think this definition is the most sensible for logging, as it provides a clear record of the duration of the packet, independent of the packet's type or source.
In previous releases the RX_OFDM and TX_LOW length fields did not include the MAC header or 4-byte FCS. This is the 28 byte difference you see when comparing data packet log entries across versions.
Offline
Thanks murpho for the clarification.
Offline
Pages: 1