Algorithm 1. Learning progress of sensor vi
1	Initialise Q value for each available action arbitrarily;
2	for k = 0 to a predefined integer do;
3	calculate π;
4	for each available action a ∈ Ai do;
5	Qk+1(a) = Qk(a) + π (a)α1(∑a (a)π(a) − Qk(a));
6	end for
7	end for
8	aopti ← argMax(Q);
9	vi takes the action aopti;