Future 6G wireless networks are expected to evolve toward AI-native systems that support intelligent, environment-aware services, including positioning and mapping. Current research has already integrated AI into several aspects of communication, particularly wireless channel estimation. However, most learning-based channel estimators focus on communication-oriented objectives rather than explicitly estimating multipath parameters such as delay, angle of arrival (AoA), and angle of departure (AoD). These parameters provide essential geometric information for downstream tasks including positioning and radio simultaneous localization and mapping (SLAM). In this work, we propose a Spectral-Convolutional Factorized Transformer (SCFT) for joint multipath component (MPC) estimation from noisy complex channel estimates and physical beam-pair descriptors. SCFT combines local spectral convolution, permutation-equivariant beam-set attention, and a query-based Transformer decoder to detect and classify line-of-sight (LoS) and non-line-of-sight (NLoS) paths and estimate their observed delays, AoAs, and AoDs. Experiments on realistic ray-tracing data show that SCFT reduces the generalized optimal sub-pattern assignment (GOSPA) error by 13.4% compared with ResNet, the strongest competing baseline. We further demonstrate this advantage in independent snapshot positioning and radio SLAM, where SCFT reduces position RMSE by 18.5% and 65.0%, respectively, relative to ResNet. These findings highlight the importance of explicitly recovering multipath geometry, as improvements in MPC estimation can translate into substantial gains in downstream tasks.
Figures & tables
Fig. 1: Overview of the proposed Spectral-Convolutional Factorized Transformer (SCFT) framework for joint multipath component estimation.
Item
Value
Carrier; subcarrier spacing
28 GHz; 120 kHz
PRS subcarriers; selected pairs
Nsc=768 ; S=32
BS/RX arrays; codebooks
8×8 / 4×4 ; Kronecker DFT
Train/val./cal./test samples
35000/5000/5000/5000
SNR; MPC queries
0 – 30 dB; M=5
SCFT encoder/decoder
Nb=4 , R=2 , d=128 , 8 heads
TABLE I: Experimental configuration.
Method
Params (M)
GFLOPs
GOSPA μ±σ
F1
Delay RMSE (ns)
AoA MAE ( ∘ )
AoD MAE ( ∘ )
SCFT
1.490
3.621
2.331±2.730
0.903
21.37
5.54
2.31
Flat Transformer
1.491
4.081
3.276±3.073
0.853
24.91
7.93
3.13
ResNet
1.520
3.614
2.692±2.838
0.887
20.41
6.29
3.02
Vanilla CNN
1.520
3.612
6.770±2.552
0.646
38.73
20.71
4.88
Spectral U-Net
1.563
3.583
6.343±2.632
0.688
37.04
19.57
4.41
SAGE
–
–
7.827±1.746
0.302
62.03
7.98
7.76
TABLE II: Test-set MPC estimation performance. Geometric errors are evaluated over matched paths; computational complexity is reported per sample.
Fig. 2: MPC estimation performance versus SNR: (a) delay RMSE, (b) AoA MAE, (c) AoD MAE, and (d) GOSPA. Solid and dashed curves denote LoS and NLoS paths, respectively, in (a)–(c).
Variant
GFLOPs
GOSPA
F1
SCFT (spectral convolution)
3.621
2.331
0.903
Full spectral attention
3.777
2.784
0.885
No cross-pair attention
3.681
2.790
0.884
Neutral beam descriptors
3.777
7.966
0.467
TABLE III: Controlled architectural ablations.
Metric
SCFT
ResNet
SAGE
Oracle
Snapshot RMSE
2.55±0.16
3.13±0.14
35.30±1.11
0.78±0.41
SLAM RMSE
1.78±0.37
5.09±7.93
25.79±85.03
0.83±0.32
RPE
0.47±0.05
0.94±0.87
21.86±119.15
0.18±0.03
Map GOSPA
4.94±1.12
3.997±1.08
11.90±2.91
1.26±1.33
Association F1
0.879±0.024
0.813±0.093
0.081±0.038
0.977±0.013
TABLE IV: Downstream results over 50 paired noisy channel realizations ( μ±σ ). RMSE, RPE, and map GOSPA are in metres.
Fig. 3: Positioning and radio SLAM results: (a) snapshot positioning error, (b) radio SLAM position error, and (c) representative trajectories and virtual-anchor maps. Panels (a) and (b) show mean empirical CDFs with pointwise standard-deviation bands across 50 noisy realizations. Panel (c) shows a common realization selected based on proximity to the method-wise median SLAM RMSE and map GOSPA.
Channel-gain maps provide the channel gain between any two locations in a geographical region. They find numerous applications, from resource allocation and interference control to path planning for autonomous vehicles. Channel-gain map estimation (CGME) is considerably more challenging than conventional radio map estimation (RME) because channel-gain maps are functions over a 6-dimensional input space. This calls for specialized methods, which currently rely on the (inaccurate) radio tomographic model or require a prohibitively large number of measurements since they do not exploit any spatial structure. This paper overcomes this issue by leveraging spatial patterns that channel-gain maps exhibit across environments, as dictated by the laws of physics and typical environmental characteristics (e.g. building materials and layouts). Adopting a metalearning perspective, a transformer-based estimator is proposed to implicitly learn this common structure from measurements collected in multiple environments. This enables CGME in new environments from significantly fewer measurements (five times less in our experiments). To maximize learning efficiency, the transformer is composed with a feature map that enforces the invariances of CGME, such as those following from reciprocity. Numerical experiments corroborate the merits of the proposed estimator relative to existing methods.
Prasenjit Dhara, Daniel Romero
Dept. of Information and Communication Technology, University of Agder, Grimstad, Norway.
Recent advances in machine learning have enabled training of wireless foundation models, which aim to support tasks such as channel estimation, beam prediction, and localization based on wireless signals. Existing wireless foundation models typically pretrain on channel tensors using masked reconstruction over subcarriers, antennas, or time but ignore the physical characteristics of wireless propagation. In this work, we propose to instead use multipath propagation as the fundamental pretraining object. We present MultiPathFormer, an autoregressive foundation model that represents each transmitter-receiver link as an ordered sequence of continuous-valued path tokens and pretrains with next-path prediction. We introduce an Environmental RAG (retrieval-augmented generation) mechanism and a first-path codebook on top of the transformer backbone, leveraging environment knowledge to improve path statistics estimation like delay and power by up to 59%. MultiPathFormer pretrained on 27 environments transfers to unseen users and, after scenario-specific fine-tuning, outperforms training the corresponding models from scratch in new environments. Across downstream tasks, it outperforms SOTA channel-based foundation models, achieving 5.57 m mean localization error, 0.914 top-3 beam accuracy, 0.994 line-of-sight classification accuracy, and 0.561 channel estimation NMSE. These results show that path-level pretraining can learn reusable representations of wireless propagation.
Accurate multipath parameter estimation is critical for modern wireless communication systems, particularly in challenging low-SNR environments. Traditional Maximum Likelihood Estimation algorithms, such as CLEAN, provide high-resolution parameter extraction but suffer from prohibitive computational complexity due to exhaustive grid search. Conversely, purely data-driven deep learning approaches lack physical grounding and struggle to generalize across variable multipath densities and off-grid parameters. To address these limitations, this paper proposes Neural Network-Assisted CLEAN (NN-CLEAN), a hybrid framework that embeds a multi-head residual network directly into the iterative CLEAN extraction loop. By replacing the exhaustive grid search with rapid, parallelizable forward passes while delegating residual subtraction to exact mathematical models, NN-CLEAN isolates physical multipath parameters without accumulating non- physical errors. Extensive Monte Carlo simulations demonstrate that NN-CLEAN achieves estimation accuracy exceeding 96% at 5 dB SNR, matching the traditional Grid-Search CLEAN (GS- CLEAN) baseline, while providing a massive reduction in computational complexity and substantially outperforming subspace methods and standalone one-shot neural networks. Crucially, NN-CLEAN exhibits a near-flat scaling in execution runtime and memory consumption as batch sizes increase. This highly efficient parallelization establishes NN-CLEAN as a robust, real- time solution for channel estimation in MIMO systems.
Chaofan Deng, Linyu Sun, Jaeho Lee +1
School of Electrical and Computer Engineering, Georgia Institute of Technology, Atlanta, GA, USA