Description
The Extended Typical Urban (ETU) model is a tapped-delay-line (TDL) statistical channel model that characterizes the multipath propagation effects experienced by radio signals in dense urban environments. Defined in 3GPP specifications such as TS 36.104 and TS 38.901, it is part of a suite of channel models used for base station (BS) and User Equipment (UE) performance testing. The model mathematically represents the radio channel as a finite impulse response (FIR) filter with a set of discrete taps. Each tap is defined by a relative delay, a complex amplitude (which varies over time to model fading), and a power level. The ETU model specifically features a root mean square (RMS) delay spread of 991 nanoseconds and a maximum excess delay of 5000 nanoseconds, representing a channel with significant time dispersion due to reflections from buildings and other large structures.
In operation, the ETU model is implemented in channel emulators or simulation software to assess receiver performance under challenging real-world conditions. For conformance testing, a UE or base station receiver is subjected to a standardized test signal that has been distorted by the ETU channel model, often combined with additive white Gaussian noise (AWGN). The device's ability to correctly demodulate and decode the signal is then measured, with metrics such as throughput, block error rate (BLER), or reference sensitivity power determining pass/fail criteria. The model uses a specific Doppler spectrum (typically the Classical or Jakes spectrum) to simulate the time-varying nature of the channel caused by relative motion between transmitter and receiver.
The ETU model's tap parameters are derived from extensive empirical measurement campaigns in urban areas. It includes nine distinct taps with specific delays and power levels. The power delay profile is exponential, meaning later arriving echoes generally have lower power. The fading on each tap is typically modeled as Rayleigh or Rician distributed, representing non-line-of-sight or partial line-of-sight conditions, respectively. This detailed statistical representation allows the model to accurately reproduce key channel impairments like frequency-selective fading and inter-symbol interference (ISI), which advanced receiver algorithms like equalizers and OFDM/OFDMA are designed to mitigate.
Within the 3GPP ecosystem, the ETU model is crucial for ensuring that devices meet minimum performance requirements, guaranteeing a consistent user experience across different networks and vendors. It is applied in test cases for reference sensitivity, adjacent channel selectivity, and blocking characteristics. With the evolution from LTE to NR, the basic principles of the TDL model remain, but the ETU parameters are integrated into a more flexible framework in TR 38.901, which supports a wider range of frequency bands and scenarios. Understanding the ETU model is essential for radio frequency (RF) and protocol test engineers, as it forms the bedrock of realistic physical layer performance validation.
Purpose & Motivation
The Extended Typical Urban (ETU) channel model was created to provide a standardized, realistic, and repeatable test condition for evaluating the performance of wideband cellular systems, starting with LTE, in challenging urban propagation environments. Prior to its standardization, performance testing often used simpler models (like AWGN or flat fading) or proprietary channel models, making it difficult to compare results between different vendors' equipment or to guarantee a minimum level of real-world performance. The ETU model solves this by defining a specific, severe multipath scenario that all devices must be tested against, ensuring they can handle the time dispersion and fading typical of dense urban macrocells.
The motivation for developing the ETU model stemmed from the transition to OFDMA-based systems like LTE, which are particularly sensitive to delay spread and frequency-selective fading. System designers needed a channel model that accurately reflected the delay spreads observed in real urban measurements—much larger than those in the earlier Typical Urban (TU) model used for GSM. The 'Extended' prefix indicates this increased delay spread. By creating a harsh but standardized test condition, 3GPP aimed to drive the development of robust receiver designs with effective equalization and channel estimation algorithms, ultimately ensuring that end-users experience reliable high-data-rate services even in complex radio environments.
Furthermore, the ETU model serves a critical regulatory and commercial purpose. It forms part of the basis for conformance testing standards, which are used by certification bodies to approve devices for market entry. This provides network operators with confidence that deployed devices will perform adequately on their networks. From a system design perspective, the model is also used in link-level and system-level simulations to estimate cell coverage, capacity, and the benefits of advanced receiver features, guiding network planning and feature development. Its continued use into the 5G NR era demonstrates its enduring value as a benchmark for urban macrocell performance.
Key Features
- Standardized 3GPP Tapped-Delay-Line (TDL) channel model
- Represents high delay spread urban macrocell environment (RMS delay spread 991 ns)
- Defined with 9 discrete taps with specific delays and power levels
- Ucludes time-varying fading modeled with a Doppler spectrum (e.g., Classical)
- Used for UE and base station RF performance/conformance testing
- Applicable for LTE and NR system evaluation and simulation
Evolution Across Releases
Introduced as a key performance test channel model for LTE User Equipment (UE) and evolved NodeB (eNB). Defined in TS 36.104 with a 9-tap power delay profile, an RMS delay spread of 991 ns, and a maximum excess delay of 5000 ns. It was established as a benchmark for testing receiver performance under severe frequency-selective fading conditions.
Defining Specifications
| Specification | Title |
|---|---|
| TS 21.905 | 3GPP TS 21.905 |
| TS 36.104 | 3GPP TR 36.104 |
| TS 36.116 | 3GPP TR 36.116 |
| TS 36.117 | 3GPP TR 36.117 |
| TS 36.141 | 3GPP TR 36.141 |
| TS 36.855 | 3GPP TR 36.855 |
| TS 36.878 | 3GPP TR 36.878 |
| TS 37.901 | 3GPP TR 37.901 |