Improving Channel Charting using a Split Triplet Loss and an Inertial Regularizer

Rappaport, Brian; Gönültaş, Emre; Hoydis, Jakob; Arnold, Maximilian; Srinath, Pavan Koteshwar; Studer, Christoph

doi:10.1109/iswcs49558.2021.9562215

Cited by 10 publications

(5 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The literature describes a variety of methods to learn neural network-based channel charting functions, such as autoencoders [2], Siamese neural networks [14], and neural networks trained with a timestamp-based triplet loss [19], [23], [24]. In what follows, we focus on the triplet-loss channel charting approach from [19].…”

Section: Channel Charting With Neural Networkmentioning

confidence: 99%

Channel Charting in Real-World Coordinates

Taner,

Palhares,

Studer

2023

GLOBECOM 2023 - 2023 IEEE Global Communications Conference

Self Cite

View full text Add to dashboard Cite

Channel charting is an emerging self-supervised method that maps channel state information (CSI) to a lowdimensional latent space, which represents pseudo-positions of user equipments (UEs). While this latent space preserves local geometry, i.e., nearby UEs are nearby in latent space, the pseudopositions are in arbitrary coordinates and global geometry is not preserved. In order to enable channel charting in real-world coordinates, we propose a novel bilateration loss for multipoint wireless systems in which only the access point (AP) locations are known-no geometrical models or ground-truth UE position information is required. The idea behind this bilateration loss is to compare the received power at pairs of APs in order to determine whether a UE should be placed closer to one AP or the other in latent space. We demonstrate the efficacy of our method using channel vectors from a commercial ray-tracer.

show abstract

Section: Channel Charting With Neural Networkmentioning

confidence: 99%

Channel Charting in Real-World Coordinates

Taner,

Palhares,

Studer

2023

GLOBECOM 2023 - 2023 IEEE Global Communications Conference

Self Cite

View full text Add to dashboard Cite

show abstract

“…The literature describes a variety of neural-network-based channel charting functions building on autoencoders [1], [16], networks trained with a triplet loss [2], [15], [17], [18], or Siamese neural networks [14], [19], [20]. In what follows, we focus on Siamese neural networks that match the distances between the positions in the channel chart to pairwise dissimilarities computed from CSI.…”

Section: B Channel Charting Pipelinementioning

confidence: 99%

Channel Charting for Streaming CSI Data

Taner,

Guillaud,

Tirkkonen

et al. 2023

2023 57th Asilomar Conference on Signals, Systems, and Computers

Self Cite

View full text Add to dashboard Cite

Channel charting (CC) applies dimensionality reduction to channel state information (CSI) data at the infrastructure basestation side with the goal of extracting pseudo-position information for each user. The self-supervised nature of CC enables predictive tasks that depend on user position without requiring any ground-truth position information. In this work, we focus on the practically relevant streaming CSI data scenario, in which CSI is constantly estimated. To deal with storage limitations, we develop a novel streaming CC architecture that maintains a small core CSI dataset from which the channel charts are learned. Curation of the core CSI dataset is achieved using a min-maxsimilarity criterion. Numerical validation with measured CSI data demonstrates that our method approaches the accuracy obtained from the complete CSI dataset while using only a fraction of CSI storage and avoiding catastrophic forgetting of old CSI data.

show abstract

“…The literature describes a variety of methods to learn neural-network-based channel charting functions, such as autoencoders [3], [29], Siamese neural networks [30], and neural networks trained with a timestamp-based triplet-loss [12], [42], [43]. In what follows, we focus on the timestamp-based triplet-loss approach from [12].…”

Section: Channel Charting With Neural Networkmentioning

confidence: 99%

Optimality of the Discrete Fourier Transform for Beamspace Massive MU-MIMO Communication

Taner

Studer

2021

2021 IEEE International Symposium on Information Theory (ISIT)

View full text Add to dashboard Cite

Beamspace processing is an emerging technique to reduce baseband complexity in massive multiuser (MU) multipleinput multiple-output (MIMO) communication systems operating at millimeter-wave (mmWave) and terahertz frequencies. The high directionality of wave propagation at such high frequencies ensures that only a small number of transmission paths exist between user equipments and basestation (BS). In order to resolve the sparse nature of wave propagation, beamspace processing traditionally computes a spatial discrete Fourier transform (DFT) across a uniform linear antenna array at the BS where each DFT output is associated with a specific beam. In this paper, we study optimality conditions of the DFT for sparsity-based beamspace processing with idealistic mmWave channel models and realistic channels. To this end, we propose two algorithms that learn unitary beamspace transforms using an 4 -norm-based sparsity measure, and we investigate their optimality theoretically and via simulations. p p i,k |A i,k | p the " p -norm," even though the pth root

show abstract

Improving Channel Charting using a Split Triplet Loss and an Inertial Regularizer

Cited by 10 publications

References 19 publications

Channel Charting in Real-World Coordinates

Channel Charting in Real-World Coordinates

Channel Charting for Streaming CSI Data

Optimality of the Discrete Fourier Transform for Beamspace Massive MU-MIMO Communication

Contact Info

Product

Resources

About