GERAN evolution in 3GPP R7

On its latest GERAN plenary meeting in Brussels in January, 3GPP updated the R7 feasibility study TR 45.912 to version 0.4.0. It is an encribly interesting read.

3GPP R7 has the following performance objectives.

• The spectrum efficiency/capacity in an interference limited scenario should be 50% better (measured in kbps/MHz/cell for data and Erl/MHz/cell for speech)
• Peak data rates should increase with 100% in downlink and uplink
• Speech and data coverage should improve in a noise limited scenario, by means of a downlink sensitivity increase of 3 dB
• Service availability should improve (when cells are planned for speech) in the sense that the mean bit rate should increase by 50% at cell edges (for uplink and downlink)
• Reduced latency: a round trip time less than 450 milliseconds at initial access (no TBF assigned), and a round trip time less than 100 milliseconds after initial access (trip times from the mobile station up to the Gi interface and back are included in this RTT, which is stated for non-ideal radio conditions on the radio interface) — great for delay-sensitive applications, such as VoIP

Moreover, performance improvements should be balanced, meaning relatively similar uplink and downlink improvements on bit rates, coverage, capacity and latency. Throughput improvements should be supported by available round trip time e.g. RTT-bit rate-product should not increase over typical TCP window size. Furthermore, it’s clear that peak bit rate or improvements in ideal conditions are not the primary optimisation goal. Instead, typical performance in non-ideal conditions in a real network is tackled (one way to improve average performance is of course to improve peak performance).
Impact on BTS, BSC and CN hardware will be avoided, and of course legacy MSs can coexist with new MSs sharing the same radio resources.

There’s a whole bunch of new features being studied to realize these objectives.

• MS receiver diversity
• Dual-carrier and multi-carrier
• New modulation schemes
• Dual symbol rate
• Latency enhancements (improved ACK/NACK reporting, variable sized radio blocks, and two different proposals for reduced TTI — Transmission Time Interval)
• New burst structures and new slot formats
• Adaptation between MS receiver diversity and dual-carrier
• Turbo codes
• Enhancements to resource allocation
• Power control in frequency hopping

3GPP plans to finish the feasibility study by June at their plenary Lisbon meeting (#30). I’ll be posting more details in the weeks and months to come. The full R7 specs are due by the end of the year.