Q1. Consider a second-order recurrent neural network with a single input signal enters the network through a tapped delay line of 2-dimensional vector: (un, Un-1), and two state neurons (X1, X2) with tan-sigmoid activation function. The network is fully connected with the following weights and biases values: bj = 0.2, b2 = 0.4, wlij = 0.3, W2ij = 0.6. The output of the net is driven from neuron X2. (i) Draw the architecture of the network with the required multipliers. (ii) Suppose the following sequential input signal is presented the network u(t) = 0.5, 0.8, and -0.3 for time steps t=0, t=1 and t=2 respectively, determine the activation of each neuron at each time step if the initial states are: Xi(0) = 0.7, x2(0) = 0.9