You are here

Link Quality and Throughput

6 posts / 0 new
Last post
AC2OG
AC2OG's picture
Link Quality and Throughput

v.3.22.8.0 is offering the possibility to evaluate the quality of the link with "Neighbor Status", but still with "Mesh Status".
Why there is a discrepancy between the quality of the link at the same time and also the TxMbps measured in Mesh Status and with iperfspeed ?
Attached some screenshots 
w6bi
w6bi's picture
TxMbps
TxMbps isn't directly related to link quality.  From the Help page in the AREDN code:

Transmitted Mbps (TxMbps) is calculated with the formula (TxMbps = rate x EWMA) where rate is the 802.11 data rate in use by the transmitter and EWMA is the Exponentially Weighted Moving Average or the current time-weighted chance that a packet at this rate will reach the remote station. If no traffic is being routed to the neighbor, this value may be '0' until data is available to measure and determine the optimal rate.
Orv W6BI
 
AC2OG
AC2OG's picture
LQ
Thanks, Orv. Understood the point for TxMbps, but why the LQ is 100% on the Mesh Status and 87% on the Neighbor Status ?
nc8q
nc8q's picture
LQ is 100% on the Mesh Status and 87% on the Neighbor Status

The node is hearing the neighbor 100 % of the time and the neighbor is hearing your node 87% of the time.

IIRC, the node broadcasts that are used to determine LQ is sent at the lowest modulation scheme and at maximum power.
Thus LQ is not a good metric to indicate throughput.

 

AB7PA
The metrics on those two

The metrics on those two pages are calculated differently.

  • The LQ/NLQ metrics on the Mesh Status page are pulled directly from the OLSR table in the kernel which keeps track of packet loss.  From olsr.org:
    "If 7 of 10 packets that the neighbor sent went through, then the probability for a successful packet transmission from this neighbor is 7/10 = 0.7 = 70%. This probability is what we call the Link Quality."
  • The Quality metric on the Neighbor Status page is calculated based on packet retransmission rate.  It is the moving average of (total sent packets) divided by (total sent packets plus retransmissions). If the node had to send every packet twice for it to be successfully received, the link quality would be 50%.

The quality metrics on those two displays will probably not match perfectly, but they should be fairly close to indicate the relative quality of the link.

AC2OG
AC2OG's picture
The metrics
Thanks, Steve for the clarifications!
73 de Leo I3RKE/AC2OG

Theme by Danetsoft and Danang Probo Sayekti inspired by Maksimer