Fusion Methods in Multiple Sensor Systems using Feedforward Sigmoid Neural Networks
Nageswara S. V. Rao · Intelligent Automation & Soft Computing · 1999
ABSTRACTConsider a system of N sensors (S1, S2, …, SN}, where the sensor Sj outputs Y(j) ϵ in response to input X ϵ , according to an unknown probability distribution Py(j)IX. A training n-sample (X1, Y1), (X2, Y2),…, (Xn, Yn) is given where Yi, = (Yi(1), Yi(2),…, Yi(N)) and Yi(j) is the output of Sj in response to input Xi ϵ . The problem is to choose a fusion function f: from a family , based on the sample, to minimize the expected square error where Y = (Y(1), Y(2)…, Y(N)). We consider to be the set of feedforward neural networks of sigmoid units with a single hidden layer and bounded weights. The computation of f* ϵ that exactly minimizes I(f) is not possible in general since the underlying distributions are unknown. Under the boundedness of X and Y, we show that for a sufficiently large sample, a neural network estimate fˆ can be obtained such that P[I(fˆ) — I(f*)>ϵ] 0, δ, 0<δ<1, and any distribution PY, X. Using various properties of the feedforward neural networks we obtain three diffe...