CSI-based Positioning in Massive MIMO systems using Convolutional Neural Networks

Bast, Sibren De; Guevara, Andrea P.; Pollin, Sofie

doi:10.1109/vtc2020-spring48590.2020.9129126

Cited by 75 publications

(73 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The positioning accuracy of the proposed CNN on these three datasets can be seen in Table I. As can be seen, the model reaches an accuracy of 17.16 -17.30 mm (0.150 -0.151 λ) on the LoS datasets and 20.26 mm (0.176 λ) on the nLoS dataset. For the boardroom dataset, this new model more than doubles the accuracy in comparison to the previous reported mean estimation error of 55.35 mm (0.48 λ) [6]. This is due to the architecture of the proposed model.…”

Section: B Csi-based Positioning Performancementioning

confidence: 55%

“…The first dataset was recorded in the boardroom of our office building with all users in direct LoS of the base station. The full details of the creation of this dataset is presented in [6]. This room was chosen for its simple and clean layout, to minimise the scattering of the signals in the environment.…”

Section: B Csi-based Positioning Performancementioning

confidence: 99%

“…This is due to the architecture of the proposed model. This model has many more Convolutional layers in comparison to the model proposed in [6] and is therefore a deeper model. In general, a deeper model can learn more complex features than a more shallow model.…”

Section: B Csi-based Positioning Performancementioning

confidence: 99%

“…To further explore the limits to the accuracy of the proposed MaMIMO CSI-based localisation systems using CNNs, we generated three indoor spatially labelled dense open datasets [6]. Here, each dataset represented the same room but the MIMO array topology was changed.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

MaMIMO CSI-Based Positioning using CNNs: Peeking inside the Black Box

Bast

Pollin

2020

2020 IEEE International Conference on Communications Workshops (ICC Workshops)

Self Cite

View full text Add to dashboard Cite

Massive MIMO (MaMIMO) Channel State Information (CSI) based user positioning systems using Convolutional Neural Networks (CNNs) show great potential, reaching a very high accuracy without introducing any overhead in the MaMIMO communication system. In this study, we show that both these systems can position indoor users in both Line-of-Sight and in non-Line-of-Sight conditions with an accuracy of around 20 mm. However, to further develop these positioning systems, more insight in how the CNN infers the position is needed. The used CNNs are a black box and we can only guess how they position the users. Therefore, the second focus of this paper is on opening the black box using several experiments. We explore the current limitations and promises using the open dataset gathered on a real-life 64-antenna MaMIMO testbed. In this way, extra insight in the system is gathered, guiding research on MaMIMO CSIbased positioning systems using CNNs in the right direction.

show abstract

Section: B Csi-based Positioning Performancementioning

confidence: 55%

Section: B Csi-based Positioning Performancementioning

confidence: 99%

Section: B Csi-based Positioning Performancementioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

MaMIMO CSI-Based Positioning using CNNs: Peeking inside the Black Box

Bast

Pollin

2020

2020 IEEE International Conference on Communications Workshops (ICC Workshops)

Self Cite

View full text Add to dashboard Cite

show abstract

“…Multiple techniques to localise users based on the estimated channel have been studied broadly. For example, in [9], the location of the users is calculated with very high accuracy and is applied to similar scenarios.…”

Section: B Distance User Groupingmentioning

confidence: 99%

MaMIMO User Grouping Strategies: How much does it matter?

Guevara

Bast

Pollin

2019

2019 53rd Asilomar Conference on Signals, Systems, and Computers

Self Cite

View full text Add to dashboard Cite

In a Massive MIMO communication system, a large number of antennas increases the spectral efficiency; however, if this number is fixed, then the number and selection of users influence the performance of the system. In this work, an ultra-dense massive MIMO channel database is obtained for different antenna geometries, centralised and distributed scenarios, and propagation conditions. The greedy search selection of multiple users to form a group is analysed under capacity and simplified heuristic approximations. Experimental results show that the grouping strategy has a significant impact on the nLoS scenario only when a small number of users is served (multiplexing regime). However, in LoS cases, the distribution of the base station antennas has a higher impact on the system performance than the selection of the users.

show abstract

Potential, concepts, and key advances for a ubiquitous adaptive indigenous microengineering and nanoengineering in 6G network

John

Thakur

Singh

2022

Int J Communication

View full text Add to dashboard Cite

SummaryThe wireless communication is an integral part of the human life, which has been evolved from first generation (1G) to fifth generation (5G) where prime emphasis is on the enhanced mobile broadband (eMBB), ultra‐reliable low latency communication (urLLC), and massive machine type communication (mMTC). Further, we are working towards the sixth generation (6G) network where microtechnology and nanotechnology will play a key role to provide worldwide coverage, increased spectral/energy/cost efficiency, and improved intelligence and protection. This paper emphasizes on the prime enabling technologies that are enabled by the microengineering and nanoengineering in 6G framework to meet these needs. Started with the detailed study and analysis of 6G networks such as evolution from 1G to 6G and prime difference between 5G and 6G. Further, the potential 6G dimensions, architecture of microtechnology and nanotechnology designs, and key techniques required in 6G networks are explored in detail. Moreover, the role of cognitive technologies for development of micro‐ and nano‐perspective in 6G networks is presented. Furthermore, after discussing about the application areas, finally, the prime research challenges and potential future research directions are illustrated.

show abstract

CSI-based Positioning in Massive MIMO systems using Convolutional Neural Networks

Cited by 75 publications

References 8 publications

MaMIMO CSI-Based Positioning using CNNs: Peeking inside the Black Box

MaMIMO CSI-Based Positioning using CNNs: Peeking inside the Black Box

MaMIMO User Grouping Strategies: How much does it matter?

Potential, concepts, and key advances for a ubiquitous adaptive indigenous microengineering and nanoengineering in 6G network

Contact Info

Product

Resources

About