Consider a binary hypothesis testing where the observation Y has the following PDFs
$$f(y|H_0) = \begin{cases}
A_0(a - |y|), & \text{for } |y| \le a, \\
0, & \text{for } |y| > a,
\end{cases} \text{ given } H_0,$$
$$f(y|H_1) = \begin{cases}
A_1(b - |y|), & \text{for } |y| \le b, \\
0, & \text{for } |y| > b,
\end{cases} \text{ given } H_1.$$
Let's assume $b > a > 0$.
a) Calculate the constants $A_0$ and $A_1$ in terms of parameters $a$ and $b$.
b) Assume uniform cost assignment $c_{ij} = 1 - \delta_{ij}$ and find the optimum decision rule based on Bayes
criterion.
c) Evaluate $P_D$ and $P_{FA}$ as a function of the detection threshold $\Lambda_0 = \frac{\pi_0}{\pi_1}$. Assume $b = 2a$ and sketch
the ROC curve.