LTE-Advanced (LTE-A) is an emerging mobile communications standard being developed by 3GPP. Specified as part of Release 10 of the 3GPP specifications, it is now approved for 4G IMT-Advanced. LTE-A leverages many existing LTE Release 8/9 parameters, while also incorporating a number of enhancements, including carrier aggregation, an enhanced multiple access scheme and MIMO transmission, multi-hop transmission, coordinated multipoint (CoMP) transmission/reception, and support for heterogeneous networks. These enhancements enable significant benefits, but they also create baseband and RF design challenges that further complicate the 4G physical layer (PHY) architecture development. Editors note: A companion article to this one "Defining the 4G PHY architecture design challenges" looks at the details of these enhancements. It can be found here.
Next-generation Electronic Design Automation (EDA) tools, with their array of new capabilities, offer a viable resolution to this dilemma. The trick is in understanding what these new capabilities are and how they can be used to overcome 4G challenges.
A number of EDA tools available on the market today can be used for LTE-based design; however, creating superior systems designs for the emerging LTE-A standard requires an entirely new set of functionality. Some of the key capabilities required include:
- an instrument-grade Standards Intellectual Property (IP) reference that can be used throughout the design process and enable algorithmic modeling that actually stays in touch with both RF and system-level performance;
- a modular top-down electronic system level (ESL) design approach that spans both baseband and RF domains;
- high-performance measurement and modeling techniques; and
- an open SW/HW platform, with single-vendor worldwide applications and support.
To better understand how these capabilities enable designers to overcome 4G challenges, consider the example of the SystemVue EDA environment, an EDA environment for ESL design that offers the four capabilities previously identified (Figure 1). Focused on the PHY of wireless communications systems, it 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. The W1918 LTE-Advanced library is an option to SystemVue that provides 100 reference models, coded sources and receivers, and testbenches for 3GPP Releases 8, 9 and 10.
Figure 1. SystemVue acts as a system-level design cockpit, unifying a range of cross-domain capabilities into one environment, including the baseband algorithm design, baseband hardware implementation, RF system architecture, RF hardware design, and system integration and verification. Challenge 1
: Gaining access to a working algorithmic reference that can be used throughout the design flow.
Algorithmic or IP references are a critical component of model-based design. Virtually everyone involved in the design process requires access to some level of algorithm reference, whether it’s an individual designer or different teams of engineers. The problem lies in finding a reference that can be used by the entire design team, throughout the entire design flow.Solution
: Answering this challenge requires an EDA solution that supports model-based design and is able to deliver IP to different teams of engineers throughout the design flow (Figure 1). In the case of SystemVue, because it provides a cross-domain PHY modeling framework for model-based design, it can be used to create a working PHY that integrates real-world baseband and RF using any combination of software, hardware, simulation, and measurements. The W1918 library provides the open, “Golden Reference” for that model-based design (Figure 2). The reference can be used throughout the design process and ensures that any algorithmic modeling is done with awareness of both RF and system-level performance.
Figure 2. SystemVue’s W1918 LTE-A baseband verification library provides engineers with the algorithmic references they need for model-based design throughout the entire design flow. Challenge
2: Ensuring flexible early verification and minimizing non-recurring engineering (NRE) cost.
As an emerging standard, LTE-A can be difficult to verify. Standard-compliant test benches (e.g., TS 36.101-104) must be configured that require scripting and reference IP, and incur NRE cost. In addition, many verification associated tests (e.g., throughput testing) require a completed, operational system with closed feedback loop. Solution
: Meeting this challenge requires pre-configured, standards-based testbenches, a complete working reference PHY to start the design process with, and support for specialized measurements like throughput, EVM and ACLR. It also requires highly-parameterized sources that can be used as open block diagram references for algorithm developers, or to provide customizable test vectors for download to test equipment for hardware verification.
SystemVue’s W1918 LTE-Advanced library implements a fully-coded downlink source with up to eight antennas and an LTE-A fully-coded UL source with up to 4 antennas. To assist in verifying their Release 10 algorithms, designers can replace some SystemVue components with their own algorithm components, either as MATLAB or C++ code.Challenge 3: Addressing the added stress on the RF design caused by carrier aggregation (CA).
CA is the mechanism by which LTE-A specifies spectrum allocations of up to 100 MHz. It allows the aggregation of contiguous and non-contiguous component carriers to provide the wider bandwidth. Unfortunately, the increased bandwidth drives the Peak-to-Average Power Ratio (PAPR) to extreme levels and exposes frequency-dependence and other analog degradations, which cross multiple component carriers. Non-contiguous CA, the multitude of possible RF bands and the number of MIMO layers further add to the RF design challenges. Solution
: Addressing these issues requires the ability to translate real RF limitations back up to system-level performance and correlate PHY simulations with measurements. It also demands utilization of Crest Factor Reduction (CFR) and Digital Pre-Distortion (DPD) to help linearize power amplifier design, two strategies which are essential to dealing with the high PAPR resulting from the increased bandwidth enabled by CA.
SystemVue meets this need by implementing three LTE-A deployment scenarios to characterize spectrum, CCDF and PAPR (Figure 3). To model/correct for power amplifier nonlinearities and memory effects, SystemVue’s W1716 digital pre-distortion application kit can be used. It works with test equipment and RF circuit co-simulation to quickly assess the “correctability” of a power amplifier. This enables designers to model the “dirty” power amplifier for inclusion in Layer 1 link-level architecture studies.
Figure 3. In this carrier aggregation example, SystemVue is used to make LTE-A CA signals. For this design, four LTE-A downlink sources are used to generate the baseband signal for each component carrier, then the signal is aggregated in the RF.Challenge 4
: Adequately addressing MIMO and channel considerations.
Verifying LTE-A is a complicated process, one that’s made even more so by the multiplicity of MIMO and having to consider the MIMO channel during simulation. Solution
: In this case, virtual design tool techniques (e.g., simulation links from a bottom-up EDA flow) offer one way to bridge the gaps, providing a surprising level of accuracy at the algorithm/architecture stage of a design. SystemVue has the ability to simulate multi-channel MIMO scenarios that may be less convenient to configure as actual hardware measurements, thus providing a simulation-based alternative for early algorithmic and functional verification. The capability is available with the W1715 MIMO Channel Builder, an optional SystemVue block set that provides both WINNER II and LTE-Advanced MIMO channel models (up to 8x8 MIMO) for predictive BER/FER and throughput fading simulation of LTE, LTE-A or 802.16m systems. With this solution, the effects of imperfect antenna arrays and correlated MIMO fading and propagation can be compared directly across both simulation and at-speed hardware faders. Furthermore, the use of comparable reference algorithms throughout the design process helps ease the transition to verification.Challenge 5
: Dealing with the significant increase in verification that’s needed for LTE-A.
Many factors work to increase the verification task, including the various baseband PHY operating modes, RF spectral allocations/bands and analog control settings; new semiconductor processes, battery and environmental conditions; and the need for scripting, regressions, IP exchange, and testbenches across domains. Solution
: Dealing with this challenge requires next-generation EDA tools featuring a host of new capabilities geared toward simplifying and speeding verification (Figure 4). SystemVue, for example, supports both Fast Circuit Envelope (FCE) verification modeling and native RF system modeling. For RFIC verification, FCE behavioral models are simply exported from an RFIC circuit tool. SystemVue then provides a reliable system-level model in seconds that accounts for things like power- and frequency-dependence, nonlinear memory effects and frequency translation.
Figure 4. In SystemVue, FCE behavioral models run native at the system level in seconds, without needing EDA licenses.
As with most emerging technologies, design challenges for LTE-A abound, particularly when it comes to the PHY architecture development. Adequately addressing these challenges is essential for any engineer looking to create superior systems designs for the emerging LTE-A standard. Next-generation EDA tools like SystemVue provide the capabilities needed to address these challenges and in turn, successfully develop and deploy LTE-A designs.About the Authors
Daren McClearnon is a product marketing manager for system-level design products within Agilent’s EEsof EDA organization. He has held engineering and management roles in the EDA industry since 1985, including field applications and field sales, customer education and product marketing for a variety of RF and system-level products. He graduated with a BSEE from Case Western Reserve University in Cleveland, OH.
Wu Huan is a system engineer with Agilent EEsof, responsible for the development of digital signal processing algorithms for emerging wireless applications, primarily 4G and MIMO. He began his career with Agilent in 2008 developing the industry’s first ETSI compliant v8.9 LTE PHY DSP library and is currently working on the new 4G LTE Advanced PHY DSP library and Digital Pre-Distortion tool kits for Agilent SystemVue. Huan earned his master's degree in electrical engineering from Beijing University of Posts and Telecommunications.
If you found this article to be of interest, visit EDA Designline
where you will find the latest and greatest design, technology, product, and news articles with regard to all aspects of Electronic Design Automation (EDA).
Also, you can obtain a highlights update delivered directly to your inbox by signing up for the EDA Designline weekly newsletter – just Click Here
to request this newsletter using the Manage Newsletters tab (if you aren't already a member you'll be asked to register, but it's free and painless so don't let that stop you [grin]).