Lossless compression technique for the fronthaul of LTE/LTE-advanced cloud-RAN architectures

Some of the most advanced functionalities in LTE and LTE-Advanced wireless networks rely upon some kind of collaborative processing between cells, as happens in coordinated scheduling, Cooperative Multi-Point (CoMP), or enhanced Inter-Cell Interference Coordination (eICIC), among other techniques. In some of these functionalities the required amount of information exchange between cells is so high that centralized processing scenarios represent a more viable alternative, whereby central nodes perform baseband processing tasks and remote radio heads are connected to them via high-capacity fiber links (usually known as fronthaul links). The high fiber cost incurred by these so-called Cloud-RAN architectures is the main drawback for practical deployments, and compression techniques are therefore needed at the fronthaul network. The present paper explores a lossless compression technique for LTE and LTE-Advanced wireless networks where actual compression ratios depend upon the resources occupancy, thereby allowing for statistical multiplexing in the aggregation network which translates into significant cost reductions. Compression ratios as high as 30:1 can be achieved in lightly loaded LTE 2×2 MIMO cells while values around 6:1 are typically obtained for 50% cell loads.