Distributed channel prediction for multi-agent systems

Chowdappa, Vinay-Prasad; Frohole, Markus; Wymeersch, Henk; Botella, Carmen

doi:10.1109/icc.2017.7997044

Cited by 4 publications

(4 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In 5G networks, channel gain prediction can be used in resource allocation at different layers of the protocol stack with the benefit of reducing latency and/or overheads (see [60] for specific examples). Further applications involve cooperative spectrum sensing [61], where uGP exploits location uncertainty of the recorded measurement and the test locations, building of radio-maps in a distributed way [62], and for optimal sensor placement [30] to collect channel measurements (c.f. Sec.…”

Section: A Channel Gain Prediction For Network Processingmentioning

confidence: 99%

Channel Prediction With Location Uncertainty for Ad Hoc Networks

Fröhle

Charalambous

Nevat³

et al. 2018

IEEE Trans. on Signal and Inf. Process. over Networks

Self Cite

View full text Add to dashboard Cite

Multi-agent systems (MAS) rely on positioning technologies to determine their physical location, and on wireless communication technologies to exchange information. Both positioning and communication are affected by uncertainties, which should be accounted for. This paper considers an agent placement problem to optimize end-to-end communication quality in a MAS, in the presence of uncertainties. Using Gaussian processes (GPs), operating on the input space of location distributions, we are able to model, learn, and predict the wireless channel. Predictions, in the form of distributions, are fed into the communication optimization problems. This approach inherently avoids regions of the workspace with high position uncertainty and leads to better average communication performance. We illustrate the benefits of our approach via extensive simulations, based on real wireless channel measurements. Finally, we demonstrate the improved channel learning and prediction performance, as well as the increased robustness in agent placement. Index Terms-Ad-hoc networks, bit error rate, channel prediction, Gaussian processes, multi-agent systems M. Fröhle and H. Wymeersch are with the Department of Electrical Engineering

show abstract

Section: A Channel Gain Prediction For Network Processingmentioning

confidence: 99%

Channel Prediction With Location Uncertainty for Ad Hoc Networks

Fröhle

Charalambous

Nevat³

et al. 2018

IEEE Trans. on Signal and Inf. Process. over Networks

Self Cite

View full text Add to dashboard Cite

show abstract

“…Traditionally, one of the key tasks in REM reconstruction is deciding an spatial interpolation method offering a good quality and complexity trade-off when the number of locations (or samples) to be considered increases. The most popular techniques in the literature are nearest neighbor [17], inverse distance weighting [17,18], natural neighbor [19], thin plate splines [19], Gaussian process regression [20,21] and Kriging [10,14,[17][18][19][22][23][24]. Kriging is a method that was originally used in geostatistics, but it has been applied since then in many fields, and it is actually one of the most frequently applied methods for REM reconstruction.…”

Section: Introductionmentioning

confidence: 99%

Application of Radio Environment Map Reconstruction Techniques to Platoon-based Cellular V2X Communications

Roger

Botella

Pérez-Solano

et al. 2020

Sensors

Self Cite

View full text Add to dashboard Cite

Vehicle platoons involve groups of vehicles travelling together at a constant inter-vehicle distance, with different common benefits such as increasing road efficiency and fuel saving. Vehicle platooning requires highly reliable wireless communications to keep the group structure and carry out coordinated maneuvers in a safe manner. Focusing on infrastructure-assisted cellular vehicle to anything (V2X) communications, the amount of control information to be exchanged between each platoon vehicle and the base station is a critical factor affecting the communication latency. This paper exploits the particular structure and characteristics of platooning to decrease the control information exchange necessary for the channel acquisition stage. More precisely, a scheme based on radio environment map (REM) reconstruction is proposed, where geo-localized received power values are available at only a subset of platoon vehicles, while large-scale channel parameters estimates for the rest of platoon members are provided through the application of spatial Ordinary Kriging (OK) interpolation. Distinctive features of the vehicle platooning use case are explored, such as the optimal patterns of vehicles within the platoon with available REM values for improving the quality of the reconstruction, the need for an accurate semivariogram modeling in OK, or the communication cost when establishing a centralized or a distributed architecture for achieving REM reconstruction. The evaluation results show that OK is able to reconstruct the REM in the platoon with acceptable mean squared estimation error, while reducing the control information for REM acquisition in up to 64% in the best-case scenario.

show abstract

“…Therefore, it is cost and resource effective to update the radio map depending on the mobile data traffic and time in a day. Radio maps also find their potential applications in 5G heterogeneous networks, where their availability could be crucial in spectrum sensing in cognitive radios [8]- [10], interference management [11], coverage analysis [12]- [14], device to device communications [15], formation control and connectivity maintenance in multi-agent system [16], and proactive resource allocation in anticipatory networks [6], [17], [18]. A radio map contains information such as radio signal strength, delay spread and interference levels over a finite geographical region.…”

Section: Introductionmentioning

confidence: 99%

“…One of the key challenging tasks in radio map reconstruction is choosing an appropriate interpolation method offering good quality and complexity trade-off. Methods that have been specifically proposed are: nearest neighbor [19], thin plate splines [20], [21], natural neighbor [20], inverse distance weighting [22]- [25], Kriging [26] and GPR [16], [27]. Kriging is one of the most frequently applied methods for radio map reconstruction [8], [9], [13], [14], [28]- [34].…”

Section: Introductionmentioning

confidence: 99%

Distributed Radio Map Reconstruction for 5G Automotive

Chowdappa

Botella

Zapater

et al. 2018

IEEE Intell. Transport. Syst. Mag.

Self Cite

View full text Add to dashboard Cite

Radio maps are expected to be an essential tool for the resource optimization and management of 5G automotive. In this paper, we consider the problem of radio map reconstruction using a wireless sensor network formed by sensor nodes in vehicles, access nodes from a smart city infrastructure, etc. Due to limited resource constraints in sensor networks, it is crucial to select a small number of sensor measurements for field reconstruction. In this context, we present a novel distributed incremental clustering algorithm based on the regression Kriging method for radio map reconstruction in terms of average received power at locations where no sensor measurements are available. The path-loss and shadowing components of the wireless channel are separately estimated. For shadowing estimation, clusters of sensors are adaptively formed and their size is optimized in terms of the least number of sensors by minimizing the ordinary Kriging variance. The complexity of the proposed algorithm is analyzed and simulation results are presented to showcase the algorithm efficacy to field reconstruction.

show abstract

Distributed channel prediction for multi-agent systems

Cited by 4 publications

References 23 publications

Channel Prediction With Location Uncertainty for Ad Hoc Networks

Channel Prediction With Location Uncertainty for Ad Hoc Networks

Application of Radio Environment Map Reconstruction Techniques to Platoon-based Cellular V2X Communications

Distributed Radio Map Reconstruction for 5G Automotive

Contact Info

Product

Resources

About