The parameter "usePoke" is set to true in all agents of the system. The parameter "n" in messageNotification is set to the number of other agents in the system (here 5). The "senOut" boolean outputs are then connected to the "u" input of the messageNotification agent. The system is set up.
The conventional communication method uses a constant refresh-rate in the UDP inboxes of all agents, leading to excessive event generation, even when the agents are not active. This again leads to long simulation times. By using "usePoke", the UDP inboxes of the agents are only refreshed when another agent send out a message. The other agents are "poked" by the boolean signal and update their inboxes. In systems where the agents are idle in long periods of time, the "usePoke" technique can save substantial simulation time.
For more information you can also refer to MessageNotification model