LAUSR

In order to increase the network capacity and fulfill the Quality of Service requirements in multi-user wireless networks, interference alignment (IA) has been proposed. Subsequently, ergodic IA was introduced as a promising scheme to achieve a degrees-of-freedom metric equal to (1/2), at the expense of additional delay. In this letter, we study the delay-rate trade-off in ergodic IA and exactly compute the best delay exponent of JAP-B, the state-of-the-art ergodic IA scheme. Moreover, we derive lower and upper bounds for the best delay exponent and show the much better tightness of the proposed bounds compared with those available in the state of the art.