1: \begin{abstract}

2: This paper presents novel numerical approaches to finding the secrecy

3: capacity of the multiple-input multiple-output (MIMO) wiretap channel

4: subject to multiple linear transmit covariance constraints, including

5: sum power constraint, per antenna power constraints and interference

6: power constraint. An analytical solution to this problem is not known

7: and existing numerical solutions suffer from slow convergence rate

8: and/or high per-iteration complexity. Deriving computationally efficient

9: solutions to the secrecy capacity problem is challenging since the

10: secrecy rate is expressed as a difference of convex functions (DC)

11: of the transmit covariance matrix, for which its convexity is only

12: known for some special cases. In this paper we propose two low-complexity

13: methods to compute the secrecy capacity along with a convex reformulation

14: for degraded channels. In the first method we capitalize on the accelerated

15: DC algorithm which requires solving a sequence of convex subproblems,

16: for which we propose an efficient iterative algorithm where each iteration

17: admits a closed-form solution. In the second method, we rely on the

18: concave-convex equivalent reformulation of the secrecy capacity problem

19: which allows us to derive the so-called partial best response algorithm

20: to obtain an optimal solution. Notably, each iteration of the second

21: method can also be done in closed form. The simulation results demonstrate

22: a faster convergence rate of our methods compared to other known solutions.

23: We carry out extensive numerical experiments to evaluate the impact

24: of various parameters on the achieved secrecy capacity.

25: 

26: \textit{Index Terms} - MIMO, wiretap channel, secrecy capacity, sum

27: power constraint, per antenna power constraint, convex-concave.

28: \end{abstract}