RapidIO for cloud radio access, small-cell network

The wireless radio access network must progress significantly to meet the end user Quality-of-Experience (QoE) requirements associated with mobile data and video traffic. The changes in the access network in many cases are required by industry standards, which evolve to meet end-user requirements. For example, throughput and spectral efficiency in the wireless network is expected to increase by approximately 10x and 4x respectively between HSPA+ to LTE-Advanced (LTE-A) in the coming years. To improve QoE, the wireless protocols are also enhanced to reduce network latency. For example, LTE and LTE-A will support around 10 msec of latency as compared to 150 msec in GSM.

Although end users demand better QoE in terms of more bandwidth and faster responsiveness, the load on the network varies as the users move across different cells during different part of the day. To meet the variable demand, the 4G standards based on LTE and LTE-Advanced (LTE-A) incorporated features such as load indication and resource status reporting using the X2 interface between different eNodeBs at the base-station. Furthermore, quality-of-service (QoS) management parameters continue to evolve and are expected to be supported in addition to already existing low-latency hand over and interference information exchange over X2 interface between eNodeBs. This paper discusses how RapidIO supports low latency scalable solution in traditional, Cloud, and Small cell network while supporting various data flows and new requirements in the access network.

Radio access network data flows
In traditional and emerging RAN architectures, the base-stations include multiple processing cards that are connected using dedicated CPRI links to the radio units at the top of the tower. With fibre optic link, it is possible to deliver maximum power to the radio with minimum loss over a reasonable distance. This is particularly beneficial in emerging distributed RAN architecture where a large number of radio units are connected over fibre to the base band units. Each of the radio nodes, in this architecture, serves a set of users in a small or macro cell configuration.

In both traditional and distributed architectures, the RapidIO protocol connects multiple processing units (e.g., DSPs, SOCs, ASICs, FPGAs) in the channel or base band cards. The protocol ensures guaranteed delivery with lowest latency of around 100 nsec between any processing nodes.

In a typical RAN architecture, once the signal processing related to a particular radio interface is completed, the data is transported between the base band and the radio using CPRI protocol. In a distributed architecture, the CPRI protocol may not lead to a cost effective implementation of load management and interference control since by definition CPRI protocol does not provide a standardised low latency packet based switching capability that can be used to distribute traffic across multiple base band cards from the radios. RapidIO in this case is expected to provide best-in-class performance.

Two major flows in the access network include transmission from mobile to the base station (uplink flow) and reception at the mobile from the base station (downlink flow). Retransmission, in the form of Hybrid-Automatic-Repeat-Request (HARQ), might be required in case of transmission error as part of the LTE/LTE-A protocol. The HARQ timing of around 4 msec is actually much lower than the LTE/LTE-A round-trip latency. If the interconnect fabric between processing nodes support superior flow control and fault tolerance capability, it is further possible to minimise the number of HARQ retransmissions resulting into lower latency. This provides a differentiation point for OEMs and eventually leads to better QoE for the end user.