Design Articles
Introducing LTE-Advanced (Part 4 of 4)
Contributed by Agilent Technologies, Inc
Design and Test Challenges
 As an evolution of LTE, LTE-Advanced and Release 10 will pose many challenges to engineers. The LTE standard is new and quite complex, with multiple channel bandwidths, different transmission schemes for the downlink and uplink, both frequency and time domain duplexing (FDD and TDD) transmission modes, and
use of MIMO antenna techniques. LTE and LTE-Advanced will have to co-exist with 2G and 3G cellular systems for some time, so interworking necessities and potential interference remain important issues. In typical difficult radio environments, LTE sets the bar for performance targets very high, and LTE-Advanced raises it even higher.
Carrier aggregation
Although not considered a problem for the base station, carrier aggregation will undoubtedly pose major difficulties for the UE, which must handle multiple simultaneous transceivers. The addition of simultaneous non-contiguous transmitters creates a highly challenging radio environment in terms of spur management
and self-blocking. Simultaneous transmit or receive with mandatory MIMO support will add significantly to the challenge of antenna design.
The exact impact of carrier aggregation on the specifications depends on the reference UE architecture, and several are still under discussion. Until this discussion is concluded, the performance requirements for carrier aggregation remain to be decided.
Creating carrier aggregation signals
To illustrate the concepts of carrier aggregation some examples are provided here using Agilent’s SystemVue design software, which can be used for high level system design and verification.
Various options exist for implementing carrier aggregation in the transmitter architecture depending primarily upon the frequency separation, which heavily influences where the component carriers are combined:
- at digital baseband
- in analog waveforms before the RF mixer
- after the RF mixer but before the power amplifier (PA)
- after the PA
Figure 15. Possible UE transmitter architectures for various carrier aggregation scenarios
(36.912 V9.3.0 2010-06 Fig. 11.3.2.1-1)
All of the transmitter architectures illustrated in Figure 15 can be implemented easily in Agilent SystemVue software. Figure 16 shows a quick implementation of LTE Advanced sources with carrier aggregation.
Figure 16. Example of intra-band carrier aggregation in Agilent SystemVue
Figure 16 is an example of intra-band contiguous carrier aggregation. The structure assumes that each component carrier is processed by an independent signal chain. This structure could also be applied to non-contiguous carrier aggregation for both intra-band and inter-band.
Figure 17 shows the spectrum of two 20 MHz component carriers chosen from Band 7 (2600 MHz) are aggregated with the center frequency spacing set to 20.1 MHz (a multiple of the required 300 kHz). Figure 18 shows the constellation of the physical channels and physical signals in the first component carrier (2630 MHz).
Figure 17. Carrier aggregation spectrum of two adjacent component carriers
Figure 18. Constellation of the first component carrier
In Figure 19, four adjacent 20MHz component carriers chosen from 3.5 GHz are aggregated with the adjacent center frequency spacing set to 20.1 MHz.
Figure 19. Carrier aggregation spectrum of four component carriers
Enhanced uplink multiple access
The introduction of clustered SC-FDMA in the uplink allows frequency selective scheduling within a component carrier for better link performance. Also, the PUCCH and PUSCH can be scheduled together to reduce latency. However, clustered SC-FDMA increases PAPR by a significant amount, adding to transmitter linearity issues. Simultaneous PUCCH and PUSCH also increase PAPR. Both features create multi-carrier signals within the channel bandwidth and increase the opportunity for in-channel and adjacent channel spur generation. Test tools
will need to be enhanced with capability for signal generation and analysis of in-channel multicarrier signals in LTE-Advanced power amplifiers.
Figure 20 shows an example of spur generation caused by simultaneous transmission of two PUCCH signals at the channel edge.
Figure 20. Comparison of spurs generated by two adjacent vs. two channel edge RB [22]
The blue trace shows the spurs generated by two adjacent RB at the channel edge. The red trace shows the increased spurs caused by moving one of the RB to the other edge of the channel to simulate the effect of simultaneous PUCCH. Note that in some places the spurs rise by around 40 dB, which would require either a substantial improvement in power amplifier (PA) linearity or a reduction in the maximum operating level. Until issues relating to spurs are concluded, the extent to which enhanced uplink RF performance requirements will be included in Release 10 remains to be decided.
Designing an enhanced uplink signal
Figure 4 showed a block diagram for clustered SC-FDMA in LTE-Advanced. The implementation of this uplink transmission scheme using Agilent SystemVue models is shown in Figure 21. The input and output of each model can be observed.
Figure 21. Implementation of clustered SC-FDMA in Agilent SystemVue
Enhanced multiple antenna transmission
Higher order MIMO will increase the need for simultaneous transceivers in a manner similar to carrier aggregation. However, MIMO has an additional challenge in that the number of antennas will multiply, and the MIMO antennas will have to be de-correlated. It will be especially difficult to design multiband, MIMO antennas with good de-correlation to operate in the small space of an LTE-Advanced UE. Conducted testing of higher order MIMO terminals will no longer be usable for predicting actual radiated performance in an operational network. A study item in Release 10 of the 3GPP standard is looking at MIMO over the air (OTA) testing that could be extended to the higher order MIMO defined for LTE-Advanced.
The potential reception gains from MIMO systems are a function of the number of antennas. Although the theoretical potential of such systems can be simulated, practical considerations make commercial deployment more challenging. At the base station, compact 4x antenna systems are already in use. Increasing
this to 8x to maximize the potential for spatial multiplexing and beamsteering may require the use of tower-mounted remote radio heads (RRH) to avoid the need to run 8 sets of expensive and lossy cables up the tower. The increased power consumption of MIMO systems is also a factor that cannot be overlooked. There is a trade-off between the number of antennas per sector and the number of sectors per cell. In some circumstances it may be preferable to use a six sector cell with four antennas per sector rather than a three-sector cell with eight
antennas per sector.
At the UE, the main issue with higher order MIMO is the physical space required for the antennas. Laptop data-only systems clearly have an advantage over handheld devices in terms of size, power handling, and throughput requirements. In addition, it is very hard in a small device to achieve the necessary spatial separation of the antennas in order to exploit the spatial beamforming in the channel. A common solution to this is to use cross-polarization rather than spatial separation to reduce the correlation between antennas.
Designing enhanced MIMO systems
Figure 22 is an example of an 8x4 LTE-Advanced system designed in Agilent SystemVue. It is an extrapolation of the existing closed-loop spatial multiplexing measurement defined for Release 8 in 36.101 8.2.1.4. The precoding matrix indicator (PMI) is fed back from the receiver to the transmitter and the throughput is calculated from the UE ACK/NACK reports. Different channel models can be used to cover the range of IMT-Advanced operating environments.
Figure 22. Example of how a DL closed loop spatial multiplexing measurement for Release 8
(36.101 8.2.1.4) could be expanded to 8x4 for LTE-Advanced
More advanced testing of spatial multiplexing performance in realistic conditions can be carried out by including UE CQI reports, which enable the use of adaptive modulation and coding (AMC) on the downlink.
Relaying
From the UE perspective, relaying is completely transparent so the design challenge is all on the network side. For the system to work, the link budget from the RN to the macro eNB must be good, which implies line-of-site positioning. The main operational challenge in getting relaying to work will be in the management of the UE. The UE must be instructed to hand over to a RN that is within range and release the RN when the UE goes out of range. If this process is not well managed, the performance of the cell could actually go down, not up as intended. Managing multi-hop relaying for coverage—for example, in a valley with no cabled backhaul—should be an easier task as no UE is involved.
Summary
These are just a few of the challenges that LTE-Advanced and Release 10 will present wireless design and test engineers. As the 4G specifications are published and the certification process moves ahead, so too will test vendors have to increase the capability of their products and invent ingenious new ways to verify the performance of the evolving 4G systems.
Outlook for LTE-Advanced Deployment
Industry-supported field trials are already demonstrating the viability of many of the technical concepts in LTE-Advanced, and 3GPP’s submission to the ITU included a self-evaluation of its proposals concluding that LTE-Advanced meets all 4G requirements for being officially certified as 4G. Nevertheless, the timing of LTE-Advanced deployment is difficult to predict and will be dependent on industry demand and the success of today’s Release 8 and 9 LTE rollouts.
From a standardization perspective LTE-Advanced is about two years behind LTE. However, the deployment of LTE-Advanced may be more than two years behind LTE for many reasons. These include the fact that LTE itself will have a slow rollout due to limited spectrum availability and the continued development and success of 2G and 3G systems. In addition, LTE-Advanced represents a big increase in system and device complexity, and it will take time for the industry to respond.
Design and Test Tools for LTE-Advanced Developers
As the leader in design and test products for LTE and wireless communications, Agilent will provide the tools needed to gain insight into complex LTE technology implementations.
Agilent SystemVue provides early R&D exploration of LTE Advanced features, facilitating the algorithm design and product development of systems based on this emerging new standard. SystemVue is Agilent’s electronic design automation (EDA) environment for electronic system level design, focused on the physical layer (PHY) of wireless communication systems. SystemVue enables system architects and algorithm developers to combine signal processing innovations with accurate RF system modeling, interaction with test equipment, and algorithm-level reference IP and applications.
For the 3GPP LTE design community, SystemVue provides math, C++, and graphical algorithmic modeling interfaces, dedicated “golden reference” blocksets for LTE Release 8 (compiled or source code IP), digital pre-distortion, physical 8x8 MIMO channel modeling and fading, and soon LTE-Advanced. With links from concept to hardware generation to test, SystemVue accelerates architectural exploration and model-based design of LTE Advanced Layer 1 systems, also linking to enterprise design flows and reducing overall verification effort.
SystemVue is a valuable, complementary environment that provides insight into expected hardware performance well before hardware is physically available, and for transitioning a project from initial inquiry into the standards to product development by cross-domain RF and baseband product teams focused on achieving next-generation system performance.
Agilent’s full range of LTE design and test products also includes baseband emulators, signal analyzers, sources, base station emulators, power meters and sensors, logic analyzers, scopes, signal creation software, and much more. For transmitter and receiver testing, the Agilent X-Series signal analyzers and generators with the existing LTE software can create and analyze LTE-Advanced component carriers (CCs), which are compatible with Release 8.
As LTE-Advanced is defined in Release 10 and beyond, Agilent products will be ready to take on the latest test requirements with powerful, standards compliant enhancements and features.
References
[22] ftp://ftp.3gpp.org/tsg_ran/WG4_Radio/TSGR4_54/Documents/R4-100427.zip
|
