User's Guide

- 17 of 42 –
neo.
cortec
.
receives it. Note that the Scheduled Data transmissions of Node B is received by both
Node A and Node C, but the routing protocol ensures that the payload is only received
by Node C. Once the payload package has reached Node C, and if the transmission
type is “Acknowledged”, an acknowledge will be routed back to the originator in the
same way as the payload data was routed – this is shown by the green line in the
illustration.
When the payload travels from node to node through the network, a local handshake
between the sending and receiving node is made; upon successful receive of the
payload package, a local Acknowledge is send back to the sending node. This ensures
that a package is not removed from the outbound queue of the sending node until it
has received confirmation that the receiving node has received the payload package,
and the CRC was successful.
If the local acknowledgement is not received within a timeout window, the payload
package will be retried next time the Scheduled Data transmission is due.
3.3.4.6 Latency
Since the payload is being transmitted with the scheduled data transmissions, the end-
to-end latency is directly coupled to the Scheduled Data rate.
As can be seen from Figure 8, the payload package waits at each node on the route.
The wait time is the time from the package was received as part of a Scheduled Data
from a neighbouring node, until it is time for the node to transmit its own Scheduled
Data. Depending on the offset between the Scheduled Data Events from the two nodes,
the wait time can be anything between 0 and the Scheduled Data Period. In a real
network, where there typically are multiple hops between the originator and the
destination, the wait time at each node will in average be one half of the Scheduled
Data Period.
Generalising this, enables us to calculate the average end-to-end latency like this:
𝑡 = 0.5 ∗ 𝑇
()*
∗ 𝑛
Where T
SCD
is the Scheduled Data period, and n is the number of hops between source
and destination.
3.3.4.7 Payload size & throughput
The NeoMesh network is optimized for applications where the amount of data being
transmitted is relatively low. Wireless Sensors, where the measurements may only be a
few bytes is a good example.