OFDM Ensemble Autoencoder Using CNN and SPSA for End-to-End Learning Communication Systems

2020 ◽  
Author(s):  
K Mohammed Asif ◽  
Aditya Trivedi
Keyword(s):  
Communication Systems ◽  
End To End

scholarly journals Distributed Channel Ranking Scheduling Function for Dense Industrial 6TiSCH Networks

Sensors ◽  
2021 ◽  
Vol 21 (5) ◽  
pp. 1593
Author(s):  
Ismael Amezcua Valdovinos ◽  
Patricia Elizabeth Figueroa Millán ◽  
Jesús Arturo Pérez-Díaz ◽  
Cesar Vargas-Rosales
Keyword(s):  
Kalman Filtering ◽  
Bandwidth Allocation ◽  
Communication Systems ◽  
High Reliability ◽  
Allocation Mechanism ◽  
Elapsed Time ◽  
Packet Arrival ◽  
End To End ◽  
Ieee 802.15.4E ◽  
Number Of Cells

The Industrial Internet of Things (IIoT) is considered a key enabler for Industry 4.0. Modern wireless industrial protocols such as the IEEE 802.15.4e Time-Slotted Channel Hopping (TSCH) deliver high reliability to fulfill the requirements in IIoT by following strict schedules computed in a Scheduling Function (SF) to avoid collisions and to provide determinism. The standard does not define how such schedules are built. The SF plays an essential role in 6TiSCH networks since it dictates when and where the nodes are communicating according to the application requirements, thus directly influencing the reliability of the network. Moreover, typical industrial environments consist of heavy machinery and complementary wireless communication systems that can create interference. Hence, we propose a distributed SF, namely the Channel Ranking Scheduling Function (CRSF), for IIoT networks supporting IPv6 over the IEEE 802.15.4e TSCH mode. CRSF computes the number of cells required for each node using a buffer-based bandwidth allocation mechanism with a Kalman filtering technique to avoid sudden allocation/deallocation of cells. CRSF also ranks channel quality using Exponential Weighted Moving Averages (EWMAs) based on the Received Signal Strength Indicator (RSSI), Background Noise (BN) level measurements, and the Packet Delivery Rate (PDR) metrics to select the best available channel to communicate. We compare the performance of CRSF with Orchestra and the Minimal Scheduling Function (MSF), in scenarios resembling industrial environmental characteristics. Performance is evaluated in terms of PDR, end-to-end latency, Radio Duty Cycle (RDC), and the elapsed time of first packet arrival. Results show that CRSF achieves high PDR and low RDC across all scenarios with periodic and burst traffic patterns at the cost of increased end-to-end latency. Moreover, CRSF delivers the first packet earlier than Orchestra and MSF in all scenarios. We conclude that CRSF is a viable option for IIoT networks with a large number of nodes and interference. The main contributions of our paper are threefold: (i) a bandwidth allocation mechanism that uses Kalman filtering techniques to effectively calculate the number of cells required for a given time, (ii) a channel ranking mechanism that combines metrics such as the PDR, RSSI, and BN to select channels with the best performance, and (iii) a new Key Performance Indicator (KPI) that measures the elapsed time from network formation until the first packet reception at the root.

scholarly journals Model-Free Training of End-to-End Communication Systems

2019 ◽  
Vol 37 (11) ◽  
pp. 2503-2516 ◽  
Author(s):  
Faycal Ait Aoudia ◽  
Jakob Hoydis
Keyword(s):  
Communication Systems ◽  
Model Free ◽  
End To End

scholarly journals Avoiding normalization uncertainties in deep learning architectures for end-to-end communication

2021 ◽  
Author(s):  
Simon Bos ◽  
Evgenii Vinogradov ◽  
Sofie Pollin
Keyword(s):  
Deep Learning ◽  
Network Architecture ◽  
Communication Systems ◽  
Digital Communication ◽  
Average Power ◽  
Communication Scheme ◽  
Batch Sizes ◽  
End To End ◽  
Improved Performance ◽  
Digital Communication Systems

Recently, deep learning is considered to optimize the end-to-end performance of digital communication systems. The promise of learning a digital communication scheme from data is attractive, since this makes the scheme adaptable and precisely tunable to many scenarios and channel models. In this paper, we analyse a widely used neural network architecture and show that the training of the end-to-end architecture suffers from normalization errors introduced by an average power constraint. To solve this issue, we propose a modified architecture: shifting the batch slicing after the normalization layer. This approach meets the normalization constraints better, especially in the case of small batch sizes. Finally, we experimentally demonstrate that our modified architecture leads to significantly improved performance of trained models, even for large batch sizes where normalization constraints are more easily met.<br>

scholarly journals A Risk-Sensitive Task Offloading Strategy for Edge Computing in Industrial Internet of Things

2021 ◽  
Author(s):  
Xiaoyu Hao ◽  
Ruohai Zhao ◽  
Tao Yang ◽  
Yulin Hu ◽  
Bo Hu ◽  
...  
Keyword(s):  
At Risk ◽  
Internet Of Things ◽  
Value At Risk ◽  
Communication Systems ◽  
Edge Computing ◽  
Conditional Value At Risk ◽  
Industrial Internet Of Things ◽  
Industrial Internet ◽  
End To End ◽  
Task Offloading

Abstract Edge computing has become one of the key enablers for ultra-reliable and low-latency communications in the industrial Internet of Things in the fifth generation communication systems, and is also a promising technology in the future sixth generation communication systems. In this work, we consider the application of edge computing to smart factories for mission-critical task offloading through wireless links. In such scenarios, although high end-to-end delays from the generation to completion of tasks happen with low probability, they may incur severe casualties and property loss, and should be seriously treated. Inspired by the risk management theory widely used in finance, we adopt the Conditional Value at Risk to capture the tail of the delay distribution. An upper bound of the Conditional Value at Risk is derived through analysis of the queues both at the devices and the edge computing servers. We aim to find out the optimal offloading policy taking into consideration both the average and the worst case delay performance of the system. Given that the formulated optimization problem is a non-convex mixed integer non-linear programming problem, a decomposition into sub-problems is performed and a two-stage heuristic algorithm is proposed. Simulation results validate our analysis and indicate that the proposed algorithm can reduce the risk in both the queueing and end-to-end delay.

On end to end network slicing for 5G communication systems

2016 ◽  
Vol 28 (4) ◽  
pp. e3058 ◽  
Author(s):  
X. An ◽  
C. Zhou ◽  
R. Trivisonno ◽  
R. Guerzoni ◽  
A. Kaloxylos ◽  
...  
Keyword(s):  
Communication Systems ◽  
Network Slicing ◽  
End To End ◽  
5G Communication
Sign in / Sign up

Export Citation Format

Share Document