YOU ARE AT:Network InfrastructureReader Forum: Front-haul compression for emerging C-RAN and small cell networks

Reader Forum: Front-haul compression for emerging C-RAN and small cell networks

Editor’s Note: Welcome to our weekly Reader Forum section. In an attempt to broaden our interac-tion with our readers we have created this forum for those with something meaningful to say to the wireless industry. We want to keep this as open as possible, but we maintain some editorial control to keep it free of commercials or attacks. Please send along submissions for this section to our editors at: [email protected].

The vision of small cell and centralized radio access network (C-RAN) has recently generated significant interests among the research community, the device suppliers and the original equipment manufacturers OEMs (Project C-RAN). The emerging architectures are expected to provide significant energy savings through a range of schemes, such as network resource sharing, traffic off-loading and interference management. The network architects can support a variety of network topologies based on the C-RAN concept. In particular, when deployed together with a macro- and micro-cellular network, the small cell architectures based on C-RAN are expected to provide significant capital expenditure and operating expense savings, which in turn, could be translated to savings for the end-users.

Link requirements and deployment scenarios

To enhance the network capacity and the quality of experience, both a traditional and the emerging architectures shall support the higher data rate requirements offered by the LTE and LTE-Advanced protocols. These protocols offer significantly higher spectral efficiency through a range of PHY and MAC layer techniques, such as carrier aggregation, MIMO, coordinated multi-point, interference cancellation, etc. In particular, LTE-A offers aggregation of up to five 20-megahertz LTE carriers. When combined with MIMO, the CA techniques may lead up to 100 megabits per second between the radio and the baseband units (i.e., the front-haul). Once the quantized I and Q samples are available, the receiver could apply a wide variety of interference cancellation, MIMO decoding, and CoMP algorithms to enhance the SNR in the network.

Table 1 illustrates an example throughput calculation for a three sector LTE-A system with five 20-megahertz carriers.

To transport data in the front haul, the operators may use the existing fiber or cable connections. Alternatively, the operators may rely on the emerging techniques, such as, front-haul over wireless links. The deployment decision is typically influenced by the infrastructure constraints. For example, in dense urban areas if it is difficult to deploy new fibers or requires zero foot-print solution, a wireless link might be more suitable. On the other hand, areas that already support fiber links, the operators may take advantage of the existing infrastructures.

Regardless of the deployments, the equipment suppliers must provide cost-effective power efficient solution so that the operators can gracefully migrate to higher capacity systems. Such efficient solutions typically result from the utilization of low-cost optical connectors, savings in number of the links and improvements in spectral efficiency in the front-haul network. One possible way to achieve these savings is through data compression. For example with two-to-one front-haul compression in C-RAN and small cell networks, it is possible to support a data rate of up to 4.9152 gigabits per second while staying with an optical connector that supports only 2.5 Gbps. With low data rate optical connectors and less number of links, it is possible to reduce both cost and energy consumption. Furthermore, with such configuration, it is easy to transport 15-bit I and 15-bit Q samples for up-to three sectors with 2×2 MIMO and two LTE component carriers (one 10 megahertz and one 20 megahertz) in a system. This in-turn enables opportunity to apply advanced interference cancellation and load management techniques based on the quantized I and Q samples that leads to system level cost and energy savings in the radio access network.

Figure 1 illustrates an example system architecture based on C-RAN and small cell with I2Q Compression.

Performance requirements

In addition to enhancing the spectral efficiency, the wireless protocols maintain certain signal quality (e.g., EVM) so that a particular QoE could be maintained in the network. On the other hand, depending on the modulation schemes, the EVM requirements may vary.

Table 2 summarizes the LTE-A EVM requirements for various modulation schemes. Similar requirements exist for other wireless protocols, such as WCDMA, GSM, etc.

Performance example

To provide a superior QoE, the compression technology used in the front-haul network should meet the EVM requirements specified by the wireless protocols. Furthermore, the chosen compression methods should keep enough margins for other modules in the signal chain so that the overall EVM performance could be achieved by the operators while achieving a higher throughput in the network.

Figure 2 illustrates a typical 3GPP E-TM3.1 downlink signal spectrum and the corresponding EVM performance with the I2Q data compression technology. In this example, a 20-megahertz LTE-A signal is compressed and de-compressed with an EVM of less than 1% RMS for a compression ratio of two-to-one. This leaves enough mar-gins for rest of the modules in the signal chain to meet the overall EVM requirements in the system.

Summary

A variety of compression techniques are being currently considered by a number of organizations and operators to gracefully migrate towards a system with larger capacity in traditional and emerging radio access networks. With superior EVM performance, it is possible to support data compression in wireless networks while keeping enough margins for other modules in the signal chain. Based on the available I and Q samples in a compression enabled solution, the system architects and the OEMs will have opportunity to optimize the system performance through a range of advanced signal processing and network resource sharing techniques in the emerging C-RAN and small cell networks.

ABOUT AUTHOR