Spread-Out Overlapping Sources by Independent Component Analysis for Location Positioning

Television channel spectrum sensing using independent component analysis

“…The steps for recovering sources signals are as follows [12]. First, centring the data to make its mean zero; next, data whitening to obtain the vector z ; define the number of independent components m , to be estimated; choose randomly the initial values for the w i , i = 1, …, m , each of unit norm; and then maximise the output negentropy of y i = w i x using (7) The next step is to organise the columns vectors w i in W = [ w 1 , w 2 ,.…”

“…The advantages of applying the ICA lies in its frequency and bandwidth independence, good performance in presence of low signal‐to‐noise ratio (SNR) (see Table 1) and low computational complexity [11]. Therefore, and with the advantage of not requiring prior information about the sources occupying the monitored spectrum [12], the ICA may be used to overcome the issues of previously spectrum detection methods.…”

ICA based DoA estimation with minimum input data

“…In this paper, a technique to estimate DoA using ICA is presented. The method is able not only to locate the source in 3‐D spatial position and special conditions, but also to use a reduced number of sensors, 9 distinguish close frequency signals, and detect overlapping sources 10 . In addition, this paper proposes to use the technique to correlate each DoA to its respective source.…”