11.213GPPBase Station System (BSS) equipment specificationRadio aspectsRelease 1999TS
The alternative experimental methodology indicated in this subclause could be used to verify, or possibly modify, the needed number of samples indicated in the previous subclauses. This would be most useful in the case of multipath propagation conditions where the statistics are very complicated and can not easily fit into simple formulae.
The approach is indicated in the following, and should be carried out once and for all for each multipath propagation condition, for each logical channel and for each type of error event according to subclause A.5 to assess the needed number of samples. Then, it can be used ever after for any BSS.
1) Record the number of error events for Np periods of Ns samples (events) under static propagation conditions for an average system requirement Ps and for a given type of error event on a given logical channel.
2) Estimate the error ratio ERi = Mi/Ns for each of the Np periods.
3) Consider the average ER of all ERi and estimate the expected value E(ER) and the variance Var(ER):
The test requirement Pt is found as follows:
Pt = Ps + G s (Eq 9a)
Pt = K Ps ‑ B s (Eq 9b)
4) The test shall be repeated with a different number of samples Ns until the test requirement Pt differs with less than +/‑ 5 % in (EQ 9a) and (Eq 9b).
The initial number of samples Ns should be as indicated in annex 3 and the number of tests Np should be 100.
The average system requirement Ps should be around 1.0 E‑2. It can then be assumed that the needed number of samples Ns is inversely proportional to Ps.