Optimized Resource Allocation and RRH Attachment in Experimental SDN based Cloud-RAN

In this paper, we design and implement an SDN-based architecture Pgm-RAN which provides an effective representation of the radio network state at different network levels. Thanks to its programmability, Pgm-RAN enables the implementation of real-time RAN control algorithms in a modular fashion within a Cloud-RAN environment. To demonstrate the effectiveness of Pgm-RAN and assess its applicability, we design and implement a control plane application that ensures an enhanced joint Remote Radio Head assignment and physical radio resource allocation denoted by enhanced-DPS. To do so, we formulate the problem as Integer Linear Program while taking into consideration the mobile users requirements and the radio environment conditions. The problem is then resolved in a polynomial time leveraging linear relaxations and Branch & Cut algorithm. Both experimental and simulation results are provided to prove the effectiveness of our framework and to gauge the performance of enhanced-DPS. It is worth noting that our experimental platform relies on two main building blocks: i) Open Air Interface platform and ii) extended version of FlexRAN SDN controller.