AI and ML for 6G networks

AI and ML for 6G networks

Artificial intelligence in wireless communication

Today, we are living in an age of weak AI, a category defined by the following five key features:

1. Logical reasoning, e.g., AlphaGo
2. Perception, e.g., face recognition
3. Knowledge representation, e.g., IBM’s Watson for Oncology
4. Language processing, e.g., Apple’s Siri, Amazon’s Alexa
5. Planning and navigation, e.g. self-driving cars

Strong AI allows machines to develop capabilities that are equal to or surpass human intelligence (e.g., intelligent robots). Another relevant factor is machine learning (ML) as a subcategory of AI. It is, for instance, used to build systems that learn from data sets rather than from programmed instructions, thus leading to a learning process based on artificial multi-layer neural networks. Now, imagine a future wireless network that comes with an AI-native air interface, making radios capable of learning from the environment and from each other based on trained neural networks.

Neural networks are in turn a subcategory of machine learning and relevant in wireless communication – as the following three examples of neural networks show:

1. Recurrent neural network (RNN): output from the previous step serves as the input for the current step (e.g. text processing). RNNs are useful for time series prediction (“memory effects”) and linearizing analog RF Frontends as well as antenna subsystems through digital pre- and post-distortion algorithms based on ML models.

2. Convolutional neural network (CNN): feed-forward neural networks with up to 30 layers. A CNN processes structured arrays of data (e.g. originally designed for image processing) and is currently one option to realize a neural receiver.

3. Concept of an autoencoder: a special type of artificial neural network assisting with learning efficient data coding in an unsupervised manner. It aims to train the network to ignore insignificant data. Autoencoders, for example in the form of transformers, are currently being investigated to compress channel state information feedback, which is gathered from measurements in the downlink and sent back in the uplink direction.

6G artificial intelligence and machine learning

Even though artificial intelligence is one of the ten main 6G research areas, it is not a standalone research area. It still plays into all of the other areas however, such as cell-free massive MIMO, full-duplex communication and intelligent reflecting surfaces. The performance of every single example can be enhanced by data-driven, trained systems in 6G networks, increasing energy efficiency and therefore also sustainability at the same time. Using trained machine learning models for signal processing tasks like channel estimation, equalization and demapping will further optimize the air interface compared to present-day 4G LTE and 5G NR networks.

Rohde & Schwarz supports research activities across Europe, Asia and the US and works as a partner in projects like the 6G-Access, Network of Networks, Automation & Simplification (6G-ANNA) lighthouse project. This project aims to develop a design for 6G that includes end-to-end architecture and simplifies the interaction between humans, technology, and the environment using new sensors and algorithms to detect human movements.

Your AI challenge for 6G networks

Establishing an AI-native air interface for 6G networks means replacing blocks in the signal processing chain on the physical layer with trained machine learning models. The first step in this process is to replace individual processing blocks but ultimately combine tasks that logically belong together in one trained machine learning model. Such tasks are channel estimation, channel equalization and demapping. These tasks are combined and replaced with one single trained ML model known as a neural receiver.

However, the signal processing for the 6G air interfaces is just one area where the use of ML may provide an advantage. Another area is the linearization of power amplifiers or the entire RF Frontend used in today’s mobile devices and base stations. AI or ML can be applied to 6G for the air interface and RF Frontend during several different phases:

Phase 1: Initially, ML may replace today’s deterministic software-algorithm-based linearization models for power amplifiers with ML. Research already started in this field in 2020 and is primarily driven by universities. Key industry players also have conducted studies on this subject however. This process is also set to be applied to the entire RF Frontend (= antenna system and transceiver).

Data accessibility is a clear challenge when it comes to artificial intelligence for 6G. This is because access to data sets is required to train a neural network. The RF Frontend is typically designed by one vendor. This means that all necessary data for training neural networks is in the hands of one single vendor – making it easier to realize this phase.

Phase 2: This phase focuses on receiver aspects, applying the concept of a neural receiver, by replacing signal processing blocks such as channel estimation, channel equalization and demapping with a trained ML model.

Phase 3: This is where end-to-end (E2E) optimization comes in. ML is used to jointly optimize TX, RX, and baseband processing. The ultimate goal during this phase is to adapt the transmission to the underlying application (voice call, web browsing, XR, etc.) and deployment scenario of the impact of the transmission channel with the ML designs being a part of 6G PHY / MAC itself. A first step towards E2E learning is the replacement of the modulation mapper with a custom, learned constellation that perfectly adapts to the imperfections of the transmitter, receiver and the impact of the wireless channel. Custom modulations allow a pilotless transmission and thus further improves the performance of the overall system.

Towards an AI-native air interface for 6G

Such highly adaptive physical layer implementations require extensive verification prior to deployment in the field. This verification requires models operating reliably – even under the rare conditions observed in the field. However, trained AI/ ML models are only as good as the training data they have been trained with. This is where AI/ML model lifecycle management (e.g. model training, selection, exchange, activation and monitoring) comes in, as frequent collaboration between the user devices and base station/network is expected. Testing and measurement must verify smooth interoperability between the components provided by different vendors.

6G and AI or ML: Our solutions and benefits

How can test and measurement solutions provide greater insights and help improve your ML-based DPD model?

Test and measurement solutions can be used to create reference models based on a classical approach using iterative methods e. g. the R&S®SMW200A vector signal generator which helps to characterize underlying hardware or the R&S®FSW signal and spectrum analyzer which allows for the sample-by-sample correction of amplitude and phase iteratively for given waveform, also known as direct DPD. Such procedures provide a good baseline.

Rohde & Schwarz also previously showcased an AI/ML-based neural receiver setup with custom constellations at the Brooklyn 6G Summit 2023. This setup uses an R&S®SMW200A vector signal generator to emulate a single user applying 2x4 MIMO transmission scheme. The signal generator is also used to add fading and noise to the transmission, emulating a real-world scenario. The signal is then captured with the R&S MSR4 multi-purpose satellite receiver by using its four receive channels, digitized and streamed to a server. This server hosts the R&S server-based testing framework that includes R&S®VSE vector signal explorer micro-services. Here the synchronization to the signal is performed, along with Fast Fourier Transform (FFT) and cyclic prefix removal, before this pre-processed data is processed by a neural receiver designed by Nvidia, using their SIONNATM framework.

Want to discuss your specific AI/ML test cases with our experts?

6G test solutions for AI and ML applications

6G AI and ML related news

Featured AI and ML content

Towards 6G: AI-native interface – neural receiver

Video: Towards 6G: AI-native interface – neural receiver

Watch this video to learn about achievable performance gains when using trained AI/ML models compared to traditional signal processing.

Watch Video

Artificial Intelligence (AI)

Video: Artificial Intelligence (AI) and Machine Learning (ML)

Machine learning (ML) has been very successful in image and video recognition as well as natural language processing. Watch this video and learn how researchers are interested in applying machine learning to signal processing in 6G.

Watch Video

Webinar: Will AI/ML revolutionize 6G?

Watch this webinar to get a summary of ongoing fundamental research for an AI-native air interface in the future 6G wireless communication standard.

Register to watch

Towards 6G: AI/ML-based neural receiver

Video: Towards 6G - AI/ML-based neural receiver

Watch this video for a joint demonstration of a neural receiver approach, using a trained machine learning model for signal processing tasks such as channel estimation, channel equalization, and demapping.

Watch Video

The role of AI/ML in future wireless communication

Webinar: The role of AI/ML in future wireless communication

Watch this webinar for an insight into how AI/ML is used in today’s 5G mobile networks, the current status of the fundamental research for an AI-native air interface in 6G and more.

Register to watch

Think Six - Is it time for wireless communication to get smart(er) with AI/ML? Part 1.

Video: Is it time for wireless communication to get smart(er) with AI/ML? Part 1

This video introduces the background theory and terminology of AI and ML and the potential role of Artificial Intelligence in mobile radio test.

Watch Video

#Think Six - Is it time for wireless communication to get smart(er) with AI/ML? Part 2.

Video: Is it time for wireless communication to get smart(er) with AI/ML? Part 2

This video looks at 3GPP Release 18 (5G-Advanced), in which industry experts start to determine whether machine learning shall bring a significant benefit, by initially studying three specific use cases.

Watch Video

#Think Six - Is it time for wireless communication to get smart(er) with AI/ML? Part 3.

Video: Is it time for wireless communication to get smart(er) with AI/ML? Part 3

How might an AI based air interface be realized? This video looks into the future to see whether an air interface for a 6G standard may be realized.

Watch Video

#ThinkSix - Validating a Machine-Learning Based Neural Receiver with 5G NR Multiple MIMO Signals

Video: Validating a Machine-Learning Based Neural Receiver with 5G NR Multiple MIMO Signals

Using Machine Learning (ML) powered by Artificial Intelligence for signal processing tasks in wireless communication really is beginning to move from theory to practice. Watch this video for a demonstration of validating the performance of a self-training neural receiver.

Watch Video

6G Artificial intelligence and machine learning FAQs

What is artificial intelligence in 6G?

Within the framework of 6G technology, AI not only plays an enabling role but will also be key to future networks. It is a method for optimizing networks and the designs of new waveforms that plays into many 6G research areas. Additionally, it can enable connected intelligence such as distributed learning.

What is machine learning in 6G?

Machine learning makes 6G network radios that are able to learn from each other and the environment, allowing a fully intelligent wireless network and management to be realized.

What are the advantages of an AI-native air interface?

Implementing a 6G AI interface leads to performance improvements as radios can learn dynamically, setting up waveforms and signals that use the available spectrum efficiently. This in turn also optimizes energy efficiency. The use of AI-native air interfaces also automates adaptation to the service needs with customized transmission schemes. It moreover enables air interfaces to adapt to any target platform.

What makes an AI-native interface different from conventional implementations?

Conventional algorithms (e.g. channel estimation) are developed manually and optimized by engineers based on well-conceived mathematical models such as wireless channel propagation properties. However, these models are only an approximation of reality. In contrast, AI/ML models learn from data. They can learn properties and algorithms without a developer describing or programming them explicitly. When trained on real-data, AI/ML models can accurately learn physical properties and often outperform algorithms implemented manually that are based on simplified mathematical models.

Is AI/ML in the air interface only relevant for 6G? What about 5G?

6G AI or ML air interfaces will not be implemented overnight. Rather, there will be an ongoing transition from 5G to 6G. In this respect, standardization has a lot to learn on how to specify a smooth interaction of AI or ML models in both user devices and base station. 3GPP 5G NR Release-18 (first 5G-Advanced release) kicks this off and studies AI-native air interface based on three use cases, namely CSI-RS feedback compression, beam management and positioning.


Sign up for our newsletter

Stay on top of the latest trends in wireless communications testing

Request information

Do you have questions or need additional information? Simply fill out this form and we will get right back to you.

Marketing permission

Your request has been sent successfully. We will contact you shortly.
An error is occurred, please try it again later.