6G Distributed MIMO Testbed for FR3
(2026) EITM01 20261Department of Electrical and Information Technology
- Abstract
- Research into distributed multiple-input multiple-output (D-MIMO) systems has gained increasing attention as a candidate technology for future 6G networks. However, evaluating such systems under realistic implementation constraints requires real-time testbeds that can capture effects such as synchronization, processing latency, hardware resource usage, and radio-frequency impairments.
This thesis further develops the Lund University Large Intelligent Surface (LuLIS) testbed. The testbed is a scalable, field programmable gate array (FPGA) based testbed with 16 distributed nodes. The work consists of two main contributions. First, a new distributed digital baseband architecture is developed to increase the supported uplink processing... (More) - Research into distributed multiple-input multiple-output (D-MIMO) systems has gained increasing attention as a candidate technology for future 6G networks. However, evaluating such systems under realistic implementation constraints requires real-time testbeds that can capture effects such as synchronization, processing latency, hardware resource usage, and radio-frequency impairments.
This thesis further develops the Lund University Large Intelligent Surface (LuLIS) testbed. The testbed is a scalable, field programmable gate array (FPGA) based testbed with 16 distributed nodes. The work consists of two main contributions. First, a new distributed digital baseband architecture is developed to increase the supported uplink processing capability from four to 32 users. The architecture supports a distributed part of maximum-ratio combining (MRC) and zero-forcing (ZF) detection schemes. Second, a front-end design for operation in Frequency Range 3 (FR3).
The implemented digital baseband is evaluated in terms of functionality, latency, and FPGA resource utilization, while the analog front-end is characterized through measurements on an evaluation front-end. The results show that the baseband design is functional but that the implementation should be improved in a few ways, to reduce the use of BlockRAM resources on the FPGAs, and to reduce the latency in certain configurations. The evaluation front-end demonstrated performance similar to the theoretical estimates. The 16 channel front-end featured an improved design based on further research and learnings from the evaluation front-end but was not manufactured due to time constraints and potential for further improvement. Together, these contributions extend the LuLIS testbed toward larger-scale multi-user experiments and future upper-mid-band 6G research. (Less) - Popular Abstract
- Imagine a room full of people with a large speaker in the middle. Everyone in the room hears the same music, regardless of where they are standing. This is similar to how today’s cellular base stations work: a single signal is broadcast over a wide area for many users to share.
Now imagine replacing that one large speaker with an array of small directional speakers. By cleverly controlling the speakers, sound waves corresponding to the music of each person can be directed in space. This is similar to what in wireless communications is called beamforming, and the system described is a close analogy to how current mobile communications are meant to work.
To improve this, the array of speakers can be split up and spread throughout the... (More) - Imagine a room full of people with a large speaker in the middle. Everyone in the room hears the same music, regardless of where they are standing. This is similar to how today’s cellular base stations work: a single signal is broadcast over a wide area for many users to share.
Now imagine replacing that one large speaker with an array of small directional speakers. By cleverly controlling the speakers, sound waves corresponding to the music of each person can be directed in space. This is similar to what in wireless communications is called beamforming, and the system described is a close analogy to how current mobile communications are meant to work.
To improve this, the array of speakers can be split up and spread throughout the room. Each speaker array can more easily focus sound toward a specific person, allowing different people to listen to different music at the same time with less interference from one another.
With distributed multiple-input multiple-output (D-MIMO), the goal is to coordinate all of the speaker arrays in the room so they work together as a single system. The system tries to maximize the loudness and clarity of the music for each person while minimizing interference from everyone else’s music.
Each speaker array also acts as a microphone array, listening to how the sound reaches different microphones in the room. By sharing this information with the other arrays, the system can continuously adjust which speakers transmit which sounds, adapting in real time as people move around the room.
It’s easy to see that the more speaker arrays you have, the more precisely they can direct sound, the more people you can serve simultaneously. In the same way, D-MIMO uses many distributed antennas working together to increase both network capacity and user performance. This thesis focuses on accelerating the calculations necessary to focus the antennas and increasing the frequency range to allow for more users in the future. (Less)
Please use this url to cite or link to this publication:
https://lup.lub.lu.se/student-papers/record/9245254
- author
- Svensson, August LU and Niklasson Lundqvist, Lukas
- supervisor
-
- Liang Liu LU
- organization
- alternative title
- Distribuerad massiv MIMO testbädd för 6G i FR3-frekvensbandet
- course
- EITM01 20261
- year
- 2026
- type
- H2 - Master's Degree (Two Years)
- subject
- keywords
- 6G, Wireless Communication, D-MIMO, Massive MIMO, MIMO, Precoding, Detection, Panel-Based Large Intelligent Surface, LIS, Decentralized Processing, VLSI, FR3, Analog Front-End, FPGA
- report number
- LU/LTH-EIT 2026-1176
- language
- English
- id
- 9245254
- date added to LUP
- 2026-06-29 13:37:31
- date last changed
- 2026-06-29 13:37:31
@misc{9245254,
abstract = {{Research into distributed multiple-input multiple-output (D-MIMO) systems has gained increasing attention as a candidate technology for future 6G networks. However, evaluating such systems under realistic implementation constraints requires real-time testbeds that can capture effects such as synchronization, processing latency, hardware resource usage, and radio-frequency impairments.
This thesis further develops the Lund University Large Intelligent Surface (LuLIS) testbed. The testbed is a scalable, field programmable gate array (FPGA) based testbed with 16 distributed nodes. The work consists of two main contributions. First, a new distributed digital baseband architecture is developed to increase the supported uplink processing capability from four to 32 users. The architecture supports a distributed part of maximum-ratio combining (MRC) and zero-forcing (ZF) detection schemes. Second, a front-end design for operation in Frequency Range 3 (FR3).
The implemented digital baseband is evaluated in terms of functionality, latency, and FPGA resource utilization, while the analog front-end is characterized through measurements on an evaluation front-end. The results show that the baseband design is functional but that the implementation should be improved in a few ways, to reduce the use of BlockRAM resources on the FPGAs, and to reduce the latency in certain configurations. The evaluation front-end demonstrated performance similar to the theoretical estimates. The 16 channel front-end featured an improved design based on further research and learnings from the evaluation front-end but was not manufactured due to time constraints and potential for further improvement. Together, these contributions extend the LuLIS testbed toward larger-scale multi-user experiments and future upper-mid-band 6G research.}},
author = {{Svensson, August and Niklasson Lundqvist, Lukas}},
language = {{eng}},
note = {{Student Paper}},
title = {{6G Distributed MIMO Testbed for FR3}},
year = {{2026}},
}