Christian Fager

Han Zhou, Haojie Chang, David Widen et al.

This paper presents a deep learning-assisted methodology for the inverse synthesis of a compact, wideband inverted Doherty power amplifier (PA). Convolutional neural networks (CNNs) and genetic algorithms (GAs) are jointly employed to generate pixelated Doherty combiner networks that integrate load modulation, impedance matching, power combining, and phase compensation into a single structure. As a proof of concept, we design and fabricate a GaN HEMT Doherty PA with a pixelated output combiner. The prototype achieves a measured peak drain efficiency of 51%-63% and a 6-dB back-off efficiency of 48%-54% over 1.9-2.5 GHz. Within the same frequency range, the measured output power is 44+/-0.3 dBm. Furthermore, with digital predistortion (DPD) applied, the prototype circuit demonstrates an adjacent channel leakage ratio (ACLR) better than -53.2 dBc.

Han Zhou, Haojie Chang, David Widen et al.

The output combiner of a Doherty power amplifier (PA) integrates load modulation, impedance matching, and phase compensation within a single network, making its design and synthesis highly challenging. In this paper, we propose a three-port Doherty combiner design methodology that combines deep convolutional neural networks (CNNs), pixelated layout representations, and genetic algorithms (GA) with dual-state impedance synthesis to address both peak and back-off power conditions. As a proof of concept, two GaN HEMT Doherty PA prototypes incorporating three-port pixelated combiners are designed and fabricated. Both prototypes achieve a measured saturated output power exceeding 44.2 dBm with peak drain efficiency above 71.2% within 2.6-2.8 GHz. Furthermore, a drain efficiency as high as 64% is measured at the 6-dB back-off level. After applying digital predistortion, each prototype achieves an adjacent channel leakage ratio (ACLR) better than -51.3 dBc.

Han Zhou, Richard Bannister, Caspar Pierce et al.

Traditional microwave filter design typically relies on iterative parameter tuning and predefined topologies, which limits design space and increases development time. This study uses a deep learning approach combining convolutional neural networks with genetic algorithms to automate pixelated microwave filter synthesis. To validate the approach experimentally, both S-parameter and spatial electric-field measurements were analyzed. The synthesized low-pass filter demonstrated excellent agreement between simulated and measured performance, achieving a 7 GHz passband with over 20 dB suppression beyond 9.5 GHz. Electro-optical measurements, for the first time, revealed electric field patterns that resemble coupled transmission-lines or stub structures, providing insight into the emergent characteristics of AI-generated designs.

Alireza Bordbar, Lise Aabel, Christian Häger et al.

We consider the problem of pilot-aided, uplink channel estimation in a distributed massive multiple-input multiple-output (MIMO) architecture, in which the access points are connected to a central processing unit via fiber-optical fronthaul links, carrying a two-level-quantized version of the received analog radio-frequency signal. We adapt to this architecture the deep-learning-based channel-estimation algorithm recently proposed by Nguyen et al. (2023), and explore its robustness to the additional signal distortions (beyond 1-bit quantization) introduced in the considered architecture by the automatic gain controllers (AGCs) and by the comparators. These components are used at the access points to generate the two-level analog waveform from the received signal. Via simulation results, we illustrate that the proposed channel-estimation method outperforms significantly the Bussgang linear minimum mean-square error channel estimator, and it is robust against the additional impairments introduced by the AGCs and the comparators.

Ali Soltani Tehrani, Haiying Cao, Thomas Eriksson et al.

This paper presents a new black-box technique for modeling long term memory effects in radio frequency power amplifiers. The proposed technique extends commonly used behavioral models by utilizing parameters that dynamically change depending on a long term memory effect while keeping the original model structure intact. This enables us to accurately track and model transient changes in power amplifier characteristics that vary slowly and are induced by the input signal. Identification of long term memory effects is discussed and an iterative identification algorithm for the model parameters is proposed. The model is experimentally tested on a 100 Watt Doherty power amplifier with a 4 MHz Gaussian noise signal that has a step--like change in the amplitude, representative of a realistic communication signal with bursty behavior and a 20 MHz 3GPP LTE test data. Results of behavioral modeling show a 2-2.5 dB and 5-6 dB improvement in average and peak NMSE modeling performance respectively, which shows the suitability of the technique to model bursty signals.

Ali Soltani Tehrani, Jessica Chani, Thomas Eriksson et al.

This paper presents an investigation into parameter adaptation in behavioral model--based digital predistortion for radio frequency power amplifiers. A novel measurement setup framework that emulates real--time adaptation in transmitters is developed that allows evaluation of different parameters, configurations and adaptation algorithms. This setup relieves the need for full feedback loops for parameter adaptation while providing the flexibility needed in the design process of parameter adaptation. Issues such as convergence speed, sensitivity to quantization noise in the feedback loop and predistortion performance are investigated for some different parameter update algorithms using the proposed measurement setup. The approach presented in this paper allows the possibility to analyze different aspects of digital predistortion adaptation algorithms, and is an important enabling step for further research on parameter adaptation before the real--time hardware is implemented for use.