Testing carrier aggregation in LTE-Advanced network infrastructure
If the high data rate targets are to be met, LTE-Advanced will require a channel bandwidth that is much wider than the 20 MHz currently specified for LTE. This will not be possible with just a single carrier in the limited spectrum bands available to most operators. Consequently, carrier aggregation--the ability to combine multiple carriers scattered around the spectrum--will be a key measure to achieve the wider effective bandwidth that will be required, typically up to 100 MHz. This means that multiple carriers composed of either contiguous or non-contiguous spectrum need to be added together to allow these wider channel bandwidths--and thus faster data rates--to be achieved.
Implementing carrier aggregation in a network will mean that operators and infrastructure vendors will require a test mobile equipped with carrier aggregation, ahead of real mobile terminals becoming available.
Evolution to LTE-Advanced
The aim of the 3GPP program for LTE-Advanced is to meet or exceed the requirements of IMT-Advanced within the time frame laid down by the International Telecommunications Union Radiocommunication Sector (ITU-R).
The key targets of IMT-Advanced are: 100 MHz bandwidth, a data rate of 1 Gbps in the downlink and 500 Mbps in the uplink, with 8x8 MIMO and 4x4 MIMO, respectively, in the downlink and uplink. C-plane latency will be a maximum of 50 ms, while U-plane latency will be less than or equal to 5 ms. Table 1 compares these targets with the specification for LTE Release 8 and for LTE-Advanced.
Table 1: 3GPP LTE-Advanced specification compared with LTE Release 8 and IMT-Advanced targets
The evolved standard will offer a higher average spectrum efficiency and cell edge user throughput than Release 8 LTE, with greater spectrum flexibility due to newly allocated bands. Self-organizing networking and deployment will be an integral part of LTE-A, because the network complexity will make manual optimization unfeasible. It is envisaged that there will be a smooth and low cost transition from LTE (Release 8) to LTE-A over the intervening period.
Furthermore, LTE-A will need to coexist with LTE, with a progressive development in infrastructure and gradual upgrades to terminals. Functionality will also need to be scalable.
As the spectrum is already crowded it is difficult for the regulatory bodies to allocate a non-fragmented part of the spectrum with 100 MHz bandwidth. Likewise the majority of the bands already assigned for LTE (see Tables 2(a) and (b)) are not broad enough on their own to provide the 100 MHz bandwidth specified for LTE-A. There is also an issue with legacy systems, where bandwidth is occupied by standards that pre-date LTE Release 8. Hence there is a need to combine the available spectrum bands in one of a number of prescribed ways, a technique collectively known as carrier aggregation.
Table 2(a) Band designations for LTE FDD:
Table 2(b) Band designations for LTE TDD:
Carrier aggregation is a means of flexible spectrum allocation in order to achieve wider bandwidth transmission. A complete system bandwidth of up to 100 MHz may consist of between two and five basic frequency blocks called component carriers (CC). At least some of the CCs are backward compatible with Release 8 LTE, and the aggregated bandwidth may be made up from either CCs from the same band (intra-band CA) or CCs from different bands (inter-band CA). LTE-A supports both contiguous and non-contiguous spectra for intra-band CA. Some examples are given in Figure 1.
Figure 1: Examples of carrier aggregation.
The first diagram in Figure 1 shows the case of contiguous intra-band carrier aggregation, where 100 MHz bandwidth is obtained by aggregating five component carriers from adjacent bands. The second diagram shows the non-contiguous intra-band carrier aggregation case. It can be seen that there is fragmented bandwidth in between the CCs. The final diagram shows inter-band carrier aggregation: the inter-band carrier aggregation is clearly non-contiguous as there is a fragmented bandwidth between the component carriers.
For frequency division duplexing (FDD), asymmetric bandwidth may be supported for uplink and downlink. Symmetric operation is defined as the case where there are equal numbers of CCs for the downlink and uplink, while asymmetric operation uses a larger number of CCs for the downlink than for the uplink. In time division duplexing (TDD), the uplink and downlink are always symmetric because they share the same carrier. A further consideration is intra-band symmetry, as shown in Figure 2, which relates to whether or not the aggregated carriers form a mirror image across the aggregate bandwidth.
Figure 2: Intra-band symmetry.
For LTE-A (3GPP Release 10), carrier aggregation is assumed to be symmetrical within the band, unless an exception is stated. The advantage of symmetry is that for a zero-IF receiver it avoids the data resource element (RE) overlapping at the DC point.
3GPP has specified a range of carrier aggregation scenarios for initial investigation for LTE-A, with architecture using up to three transceiver chains, which can operate anywhere in the range 300 MHz – 6 GHz. This poses some huge design problems for both eNodeBs and user equipment (UE). In the future all five of the CCs will be allowed to be non-contiguous, as shown in Figure 3, which further increases the number of transceiver chains.
Figure 3: Five non-contiguous component carriers.