S42/Central Software User Guide
1VV0301318 Rev. 3
– 2018-08-31
Reproduction forbidden without written authorization from Telit Communications S.p.A.- All Rights Reserved.
Page 60 of 66
•
Supervision Timeout
This is the maximum amount of time between two successful connection events. If this
amount of time passes without a successful connection event, the device is to consider the
connection lost, and return to an unconnected state. This parameter value is represented in
units of 10ms. The supervision timeout value can range from a minimum of 10 (100ms) to
3200 (32.0s). In addition, the timeout must be larger than the effective connection interval
(explained below).
The S42/Central calculates this timeout value as followed:
(Slave latency value + 1) x 2 x Connection interval time x 2
Limitations of calculated “Connection supervision timeout”:
calculated value:
+/- 10ms
minimum value:
>= 100ms
maximum value:
<= 32s
10.2.
Optimize the Connection Interval from Slave by using
the Slave Latency
The “effective connection interval” is equal to the amount of time between two connection
events, assuming that the slave skips the maximum number of possible events if slave latency
is allowed (the effective connection interval is equal to the actual connection interval if slave
latency is set to zero). It can be calculated using the formula:
Effective Connection Interval = (Connection Interval) * ( 1 + (Slave Latency) )
Take the following example:
Connection Interval:
80 (80 * 1.25ms =100ms)
Slave Latency:
4
Effective Connection Interval: (100ms) * ( 1 + 4 ) = 500ms
This tells us that in a situation in which no data is being sent from the slave to the master, the
slave will only transmit during a connection event once every 500ms.
