System Description (VoIP)



High-Speed Packet Access, a 3GPP standardized evolution of Wideband Code Division Multiple Access (WCDMA), has become a huge success as the world’s leading third-generation mobile standard. HSPA consists of HSUPA and HSDPA which enhance WCDMA in the uplink and the downlink separately. Several key techniques such as HARQ, fast base station-controlled scheduling, and the shorter frame size were introduced into HSPA targeting the higher data rates with more efficient spectrum usage and lower transmission delay (see Figure 14.1).
In HSDPA, high-speed downlink shared channel (HS-DSCH) is associated with a 2 ms frame size, in which the users share at most 15 fixed spreading factor 16 (SF 16) highspeed physical downlink shared channels (HS-PDSCHs) by code multiplexing. In this way, radio resource can be utilized in a more efficient manner than earlier dedicated resource allocation in WCDMA. Furthermore, HARQ in HSPA can provide physical layer (L1) retransmissions and soft combining, which reduce the amount of higher layer
WCDMA/HSPA key performance indicators.
FIGURE 14.1 WCDMA/HSPA key performance indicators.
ARQ transmissions and frame selections (i.e., hard combining). This also improves the efficiency of the Iub interface between base station and radio network controller (RNC) significantly.
Rather than the conventional dedicated transport channel (DCH), the enhanced DCH (E-DCH) mapping on the enhanced dedicated physical data channel (E-DPDCH) is used in HSUPA to further improve the uplink data rate up to 5.76 Mbps. The 2 ms TTI frame length introduced into HSUPA can further reduce the transmission delay in the air interface: the minimum Round Trip Time (RTT) is decreased from about 150 ms in WCDMA to 50 ms in HSPA. Besides, base station controlled fast scheduling can select the user in a good channel condition for the transmission reaching the higher throughput due to the multiuser diversity gain. This is impossible in WCDMA system, where the reaction of the RNC-controlled scheduling is too slow to follow the fast changing channel conditions. This is because of the transmission delay between UE and the RNC via the base station.


To ensure competitiveness of 3GPP radio access technologies beyond HSDPA and HSUPA, a LTE of the 3GPP radio access was initiated at the end of 2004. As a result, the Evolved Universal Terrestrial Radio Access Network (E-UTRAN) is now being specified as a part of the 3GPP Release 8. Important aspects of E-UTRAN include reduced latency (below 30 ms), higher user data rates (DL and UL peak data rates up to 100 Mbps and 50 Mbps, respectively), improved system capacity and coverage, as well as reduced CAPEX and OPEX for the operators [9].
In order to fulfill those requirements, 3GPP agreed on a simplified radio architecture. All user plane functionalities for the radio access were grouped under one entity, the evolved Node B (eNB), instead of being spread over several network elements as traditionally in
GERAN (BTS/BSC) and UTRAN (Node B/RNC) [8]. Figure 14.2 depicts the resulting radio
architecture where:
• The E-UTRAN consists of eNBs, providing the E-UTRA user plane (PDCP, RLC, MAC) and control plane protocol terminations (RRC) towards the UE and hosting all radio functions such as Radio Resource Management (RRM), dynamic allocation of resources to UEs in both uplink and downlink (scheduling), IP header compression and encryption of user data stream, measurement and measurement reporting configuration for mobility and scheduling.
• The eNBs may be interconnected by means of the X2 interface.
• The eNBs are also connected by means of the S1 interface to the Evolved Packet Core (EPC), which resides in MME/SAE gateway.
Figure 14.3 shows the E-UTRA frame structure, which consists of 10 sub-frames, each containing 14 symbol blocks. In E-UTRAN system, one sub-frame (1 TTI) of length 1.0 ms is regarded as the minimum time allocation unit. In frequency domain, a minimum allocation unit is a Physical Resource Block (PRB), which consists of 12 subcarriers (each subcar-rier is 15kHz). In DL, OFDMA scheme is designed to allow signal generation by a 2048-point FFT. This enables scaling of system bandwidth to the needs of an operator. At least the
bandwidths of {6, 12, 25, 50, 100} PRBs in the range of 1.4MHz to 20MHz are available
yielding a bandwidth efficiency of about 0.9. In DL, user data is carried by downlink shared channel (DL-SCH), which is mapped to physical downlink shared channel (PDSCH).
E-UTRAN architecture.
FIGURE 14.2 E-UTRAN architecture.
The UL technology is based on Single-Carrier FDMA (SC-FDMA), which provides excellent performance facilitated by the intra-cell orthogonality in the frequency domain comparable to that of an OFDMA multicarrier transmission, while still preserving a single carrier waveform and hence the benefits of a lower Peak-to-Average Power Ratio (PAPR) than OFDMA. In UL direction, user data is carried by uplink shared channel (UL-SCH), which is mapped to physical uplink shared channel (PUSCH).
Frame structure in E-UTRAN system.
FIGURE 14.3 Frame structure in E-UTRAN system.
With all the radio protocol layers and scheduling located in the eNB, an efficient allocation of resources to UEs with minimum latency and protocol overhead becomes possible. This is especially important for RT services such as VoIP, for which, high capacity requirements were set: at least 200 users per cell should be supported for spectrum allocations up to 5 MHz, and at least 400 users for higher spectrum allocations [9].

Next post:

Previous post: