Verification library for all current cellular standards including 5G NR, LTE-A, and LTE. Includes reference designs and measurements for each standard.


The W4522E PathWave 5G and Cellular Library includes:

  • Supports 5G NR physical layer models for 3GPP TS 38.211 and 3GPP TS 38.212
  • 3D MIMO channel based on 3GPP TR 38.901 V14.0.0
  • Over-the-Air (OTA) simulation for FR2 mmWave
  • Multi-antenna system architectures incorporating user antenna patterns
  • Artificial intelligence (AI) and machine learning (ML) in communications air-interface for 6G research

The PathWave 5G and Cellular Library provides trusted reference algorithmic modeling IP and a new, innovative simulation methodology that can be added to the PathWave System Design software platform. Consisting of signal processing building blocks, subsystems, reference multi-antenna system modeling examples, and infrastructure components, the library allows system architects to execute realistic technical research and easily evaluate your 5G communication system design. This library also includes 4G technologies for LTE and LTE-Advanced for analysis of coexistence and non-standalone operations of cellular systems.

The 5G and cellular library includes the following new AI and ML example workspaces:

  • NR_CSI_Feedback_Codebook: Demonstrates link-level simulation of CSI feedback to enhance throughput, while also saving CSI feedback information without compression (perfect) for AI-based CSI feedback training.
  • NR_CSI_Feedback_NN: Shows a pretrained Neural Network model used for CSI feedback.
  • NR_CSI_Feedback_TransNet: Illustrates the use of a pretrained TransNet model for CSI feedback compression.
  • AI_ChannelEstimation: Evaluates the performance of an NR receiver model utilizing a pretrained neural network model for channel estimation, comparing it with conventional MMSE channel estimation.
  • Generate Training Data: Generates training data for a neural network model to learn from MMSE channel estimation patterns, aiming to replace conventional MMSE channel estimation in the NR receive model.