Opnex’s approach relies on leveraging the theoretical results on systems control and optimization of the last decades that give rise to a novel spirit for network control and protocol architecting. The fundamental idea is to view the network as a continually evolving dynamic system that needs to be controlled and optimized. The control methodology relies on appropriately selecting and monitoring a set of quantities that constitute the network state or sufficient statistics of it. These quantities are incorporated in a parametric optimization problem that captures network dynamics. The solution of the optimization problem is then fed back to adjust network control parameters. With the appropriate selection of the optimization problem to be solved, the full network transport capacity can be achieved. These techniques inherently cut across communication layers and exploit available information in the network (e.g. the back-pressure control coupled with max-weight scheduling).
The above perspective shows a clear trend to rethink and redesign the wireless network protocol stack by incorporating control and optimization theory principles. While these two fields are very mature, and state-of-the-art results have exhibited the feasibility of such approaches, they have not led to final solutions. If translated to realistic protocols, these trends foretell deep impact in a wide range of protocols and techniques such as spectrum utilization, resource sharing, channel access, energy management, congestion control, topology control and interference mitigation.