Federated Learning and Control at the Wireless Network Edge

2021 ◽  
Vol 24 (3) ◽  
pp. 9-13
Author(s):  
Mehdi Bennis

We are at the cusp of two transformational technologies, namely the fifth generation of wireless communication systems, known as 5G, and machine learning (ML). On the one hand, while the evolutionary part of 5G, enhanced mobile broadband (eMBB), focusing mainly on millimeter-wave transmissions has made significant progress, fundamentals of ultra-reliable and low-latency communication (URLLC), one of the major tenets of the 5G revolution, are yet to be fully understood. In essence, URLLC warrants a departure from average-based system design toward a clean slate design centered on tail, risk, and scale [1]. While risk is encountered when dealing with decision making under uncertainty, scale is driven by the sheer amount of devices, antennas, sensors, and actuators, all of which pose unprecedented challenges in network design, optimization, and scalability.

2020 ◽  
pp. 1-17
Author(s):  
Francisco Javier Balea-Fernandez ◽  
Beatriz Martinez-Vega ◽  
Samuel Ortega ◽  
Himar Fabelo ◽  
Raquel Leon ◽  
...  

Background: Sociodemographic data indicate the progressive increase in life expectancy and the prevalence of Alzheimer’s disease (AD). AD is raised as one of the greatest public health problems. Its etiology is twofold: on the one hand, non-modifiable factors and on the other, modifiable. Objective: This study aims to develop a processing framework based on machine learning (ML) and optimization algorithms to study sociodemographic, clinical, and analytical variables, selecting the best combination among them for an accurate discrimination between controls and subjects with major neurocognitive disorder (MNCD). Methods: This research is based on an observational-analytical design. Two research groups were established: MNCD group (n = 46) and control group (n = 38). ML and optimization algorithms were employed to automatically diagnose MNCD. Results: Twelve out of 37 variables were identified in the validation set as the most relevant for MNCD diagnosis. Sensitivity of 100%and specificity of 71%were achieved using a Random Forest classifier. Conclusion: ML is a potential tool for automatic prediction of MNCD which can be applied to relatively small preclinical and clinical data sets. These results can be interpreted to support the influence of the environment on the development of AD.


2021 ◽  
Vol 9 (17) ◽  
pp. 26-39
Author(s):  
Hugo Wladimir Iza Benítez ◽  
Diego Javier Reinoso Chisaguano

UFMC (Universal Filtered Multi-Carrier) is a novel multi-carrier transmission technique that aims to replace the OFDM (Orthogonal Frequency Division Multiplexing) modulation technique for fifth generation (5G) wireless communication systems. UFMC, being a generalization of OFDM and FBMC (Filter Bank Multicarrier), combines the advantages of these systems and at the same time avoids their main disadvantages. Using a Matlab simulation, this article presents an analysis of the robustness of UFMC against fading effects of multipath channels without using a CP (cyclic prefix). The behavior of the UFMC system is analyzed in terms of the PSD (Power Spectral Density), BER (Bit Error Rate) and MSE (Mean Square Error). The results show that UFMC reduces the out-band side lobes produced in the PSD of the processed signal. Also, it is shown that the pilot-assisted channel estimation method applied in OFDM systems can also be applied in UFMC systems.


2021 ◽  
Vol 19 (2) ◽  
pp. 2056-2094
Author(s):  
Koji Oshima ◽  
◽  
Daisuke Yamamoto ◽  
Atsuhiro Yumoto ◽  
Song-Ju Kim ◽  
...  

<abstract><p>Data-driven and feedback cycle-based approaches are necessary to optimize the performance of modern complex wireless communication systems. Machine learning technologies can provide solutions for these requirements. This study shows a comprehensive framework of optimizing wireless communication systems and proposes two optimal decision schemes that have not been well-investigated in existing research. The first one is supervised learning modeling and optimal decision making by optimization, and the second is a simple and implementable reinforcement learning algorithm. The proposed schemes were verified through real-world experiments and computer simulations, which revealed the necessity and validity of this research.</p></abstract>


2018 ◽  
Vol 11 (1) ◽  
pp. 4 ◽  
Author(s):  
Dania Marabissi ◽  
Lorenzo Mucchi ◽  
Romano Fantacci ◽  
Maria Spada ◽  
Fabio Massimiani ◽  
...  

The fifth generation (5G) of wireless communication systems is considered the key technology to enable a wide range of application scenarios and the effective spreading of the smart city concept. Vertical business use cases, specifically designed for the future 5G city, will have a strong economical and social impact. For this reason, ongoing 5G field trials have to test newly deployed technologies as well as the capability of 5G to create a new digital economy. This paper describes the 5G field trial environment that was launched in Italy at the end of 2017. The aim is to evaluate the capability of the 5G network of supporting innovative services with reference to suitably designed key performance indicators and to evaluate the opportunities offered by these services. Indeed, vertical business use cases, specifically designed for the future 5G city, with a strong economic and social impact, are under implementation and will be evaluated. In particular, the paper provides a detailed description of the deployment of an actual complete integrated 5G network. It shows how 5G is effective enabling technology for a wide range of vertical business and use cases. Indeed, its flexibility allows to satisfy completely different performance requirements of real services. Some preliminary results, obtained during the first phase, are presented for a smart mobility scenario.


2014 ◽  
Vol 6 (3-4) ◽  
pp. 331-341 ◽  
Author(s):  
Arnaut Dierck ◽  
Sam Agneessens ◽  
Frederick Declercq ◽  
Bart Spinnewyn ◽  
Gert-Jan Stockman ◽  
...  

New wireless wearable monitoring systems integrated in professional garments require a high degree of reliability and autonomy. Active textile antenna systems may serve as platforms for body-centric sensing, localisation, and wireless communication systems, in the meanwhile being comfortable and invisible to the wearer. We present a new dedicated comprehensive design paradigm and combine this with adapted signal-processing techniques that greatly enhance the robustness and the autonomy of these systems. On the one hand, the large amount of real estate available in professional rescue worker garments may be exploited to deploy multiple textile antennas. On the other hand, the size of each radiator may be designed large enough to ensure high radiation efficiency when deployed on the body. This antenna area is then reused by placing active electronics directly underneath and energy harvesters directly on top of the antenna patch. We illustrate this design paradigm by means of recent textile antenna prototypes integrated in professional garments, providing sensing, positioning, and communication capabilities. In particular, a novel wearable active Galileo E1-band antenna is presented and fully characterized, including noise figure, and linearity performance.


The unused frequencies of terrestrial TV are being explored in the recent years, to satisfy the bandwidth demands of ever-increasing wireless communication systems. Allocating these unused TV frequencies is often a challenging task. Allocations that fulfil the user requirements at various instants, while maximizing the utilization of available TV frequencies is the one that is desired. In this paper, interval-graph method is implemented to identify the optimum number of channels needed for the given demand of bandwidths. Simulations are carried out by using GLPK 4.65 solver. Results show the appropriate number of channels required or conversely the reduction of data rates to individual users based on the available bandwidths.


2021 ◽  
pp. 583-588
Author(s):  
Mohamed Ibrahim Shujaa ◽  
◽  
Nada Qasim Mohammed ◽  
Moustafa K. Ibrahim ◽  
Qasim Mohammed Hussein

In next-generation of wireless communication systems, Fifth-Generation (5G), small cells deployment is one of the most important issues that must be taking in the account. This paper discusses this issue in three aspects. First, it aims to derive the Critical Handover Location (CHL) point for neighbouring wireless stations which in turn is considered an entrance to the second aspect of this work that decides the small cell placement in one network. Finally, the work proposed a new approach to evaluating the Number of Small Cells (NRS) deployment mathematically. The proposed approach provides the balance in resources allocation in the network in terms of transmitted power of each small cell and their placement in order to provide maximum capacity and coverage area with a lower level of interference between nearest wireless stations thus decreasing the total cost of network insulation.


Sensors ◽  
2019 ◽  
Vol 19 (16) ◽  
pp. 3575 ◽  
Author(s):  
M. Carmen Lucas-Estañ ◽  
Javier Gozalvez ◽  
Miguel Sepulcre

5G and beyond networks are being designed to support the future digital society, where numerous sensors, machinery, vehicles and humans will be connected in the so-called Internet of Things (IoT). The support of time-critical verticals such as Industry 4.0 will be especially challenging, due to the demanding communication requirements of manufacturing applications such as motion control, control-to-control applications and factory automation, which will require the exchange of critical sensing and control information among the factory nodes. To this aim, important changes have been introduced in 5G for Ultra-Reliable and Low-Latency Communications (URLLC). One of these changes is the introduction of grant-free scheduling for uplink transmissions. The objective is to reduce latency by eliminating the need for User Equipments (UEs—sensors, devices or machinery) to request resources and wait until the network grants them. Grant-free scheduling can reserve radio resources for dedicated UEs or for groups of UEs. The latter option is particularly relevant to support applications with aperiodic or sporadic traffic and deterministic low latency requirements. In this case, when a UE has information to transmit, it must contend for the usage of radio resources. This can lead to potential packet collisions between UEs. 5G introduces the possibility of transmitting K replicas of the same packet to combat such collisions. Previous studies have shown that grant-free scheduling with K replicas and shared resources increases the packet delivery. However, relying upon the transmission of K replicas to achieve a target reliability level can result in additional delays, and it is yet unknown whether grant-free scheduling with K replicas and shared resources can guarantee very high reliability levels with very low latency. This is the objective of this study, that identifies the reliability and latency levels that can be achieved by 5G grant-free scheduling with K replicas and shared resources in the presence of aperiodic traffic, and as a function of the number of UEs, reserved radio resources and replicas K. The study demonstrates that current Fifth Generation New Radio (5G NR) grant-free scheduling has limitations to sustain stringent reliability and latency levels for aperiodic traffic.


Sign in / Sign up

Export Citation Format

Share Document