File Download
There are no files associated with this item.
Links for fulltext
(May Require Subscription)
- Publisher Website: 10.1109/TWC.2022.3218579
- Scopus: eid_2-s2.0-85141563462
- WOS: WOS:000991554300036
- Find via
Supplementary
- Citations:
- Appears in Collections:
Article: Heterogeneous Transformer: A Scale Adaptable Neural Network Architecture for Device Activity Detection
Title | Heterogeneous Transformer: A Scale Adaptable Neural Network Architecture for Device Activity Detection |
---|---|
Authors | |
Keywords | Activity detection attention mechanism deep learning Internet-of-Things (IoT) machine-type communications (MTC) |
Issue Date | 1-May-2023 |
Publisher | Institute of Electrical and Electronics Engineers |
Citation | IEEE Transactions on Wireless Communications, 2023, v. 22, n. 5, p. 3432-3446 How to Cite? |
Abstract | To support modern machine-type communications, a crucial task during the random access phase is device activity detection, which is to identify the active devices from a large number of potential devices based on the received signal at the access point. By utilizing the statistical properties of the channel, state-of-the-art covariance based methods have been demonstrated to achieve better activity detection performance than compressed sensing based methods. However, covariance based methods require to solve a high dimensional nonconvex optimization problem by updating the estimate of the activity status of each device sequentially. Since the number of updates is proportional to the device number, the computational complexity and delay make the iterative updates difficult for real-time implementation especially when the device number scales up. Inspired by the success of deep learning for real-time inference, this paper proposes a learning based method with a customized heterogeneous transformer architecture for device activity detection. By adopting an attention mechanism in the architecture design, the proposed method is able to extract features reflecting relevance among device pilots and received signal, permutation equivariant with respect to devices, and its training parameter number is independent of the device number. Simulation results demonstrate that the proposed method achieves better activity detection performance with much shorter computation time than state-of-the-art covariance approach, and generalizes well to different numbers of devices and BS-antennas, different pilot lengths, transmit powers, and cell radii. |
Persistent Identifier | http://hdl.handle.net/10722/339302 |
ISSN | 2023 Impact Factor: 8.9 2023 SCImago Journal Rankings: 5.371 |
ISI Accession Number ID |
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Li, Yang | - |
dc.contributor.author | Chen, Zhilin | - |
dc.contributor.author | Wang, Yunqi | - |
dc.contributor.author | Yang, Chenyang | - |
dc.contributor.author | Ai, Bo | - |
dc.contributor.author | Wu, Yik-Chung | - |
dc.date.accessioned | 2024-03-11T10:35:32Z | - |
dc.date.available | 2024-03-11T10:35:32Z | - |
dc.date.issued | 2023-05-01 | - |
dc.identifier.citation | IEEE Transactions on Wireless Communications, 2023, v. 22, n. 5, p. 3432-3446 | - |
dc.identifier.issn | 1536-1276 | - |
dc.identifier.uri | http://hdl.handle.net/10722/339302 | - |
dc.description.abstract | <p>To support modern machine-type communications, a crucial task during the random access phase is device activity detection, which is to identify the active devices from a large number of potential devices based on the received signal at the access point. By utilizing the statistical properties of the channel, state-of-the-art covariance based methods have been demonstrated to achieve better activity detection performance than compressed sensing based methods. However, covariance based methods require to solve a high dimensional nonconvex optimization problem by updating the estimate of the activity status of each device sequentially. Since the number of updates is proportional to the device number, the computational complexity and delay make the iterative updates difficult for real-time implementation especially when the device number scales up. Inspired by the success of deep learning for real-time inference, this paper proposes a learning based method with a customized heterogeneous transformer architecture for device activity detection. By adopting an attention mechanism in the architecture design, the proposed method is able to extract features reflecting relevance among device pilots and received signal, permutation equivariant with respect to devices, and its training parameter number is independent of the device number. Simulation results demonstrate that the proposed method achieves better activity detection performance with much shorter computation time than state-of-the-art covariance approach, and generalizes well to different numbers of devices and BS-antennas, different pilot lengths, transmit powers, and cell radii.<br></p> | - |
dc.language | eng | - |
dc.publisher | Institute of Electrical and Electronics Engineers | - |
dc.relation.ispartof | IEEE Transactions on Wireless Communications | - |
dc.rights | This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License. | - |
dc.subject | Activity detection | - |
dc.subject | attention mechanism | - |
dc.subject | deep learning | - |
dc.subject | Internet-of-Things (IoT) | - |
dc.subject | machine-type communications (MTC) | - |
dc.title | Heterogeneous Transformer: A Scale Adaptable Neural Network Architecture for Device Activity Detection | - |
dc.type | Article | - |
dc.identifier.doi | 10.1109/TWC.2022.3218579 | - |
dc.identifier.scopus | eid_2-s2.0-85141563462 | - |
dc.identifier.volume | 22 | - |
dc.identifier.issue | 5 | - |
dc.identifier.spage | 3432 | - |
dc.identifier.epage | 3446 | - |
dc.identifier.eissn | 1558-2248 | - |
dc.identifier.isi | WOS:000991554300036 | - |
dc.identifier.issnl | 1536-1276 | - |