Low-Complexity Scalable Iterative Algorithms for IEEE 802.11p Receivers
| Autor: | Tobias J. Oechtering, Frederic Gabry, Gwilherm Baudic, Olivier Goubet |
|---|---|
| Přispěvatelé: | Institut Supérieur de l'Aéronautique et de l'Espace - ISAE-SUPAERO (FRANCE), Royal Institute of Technology – KTH (SWEDEN) |
| Jazyk: | angličtina |
| Rok vydání: | 2015 |
| Předmět: |
Computational complexity theory
Computer Networks and Communications Computer science Orthogonal frequency-division multiplexing IEEE 802.11p standard Algorithme et structure de données Factor graphs vehicular technology Aerospace Engineering Channel estimation Signalbehandling Multiplexing IEEE 802.11p Electrical and Electronic Engineering 802.11p Adaptive algorithm Estimator time-varying channels wireless communication Vehicular communications Automotive Engineering Signal Processing Bit error rate Iterative receiver Algorithm Factor graph Communication channel |
| Popis: | In this paper, we investigate receivers for vehicle-to- vehicle (V2V) and vehicle-to-infrastructure (V2I) communications. Vehicular channels are characterized by multiple paths and time variations, which introduce challenges in the design of receivers. We propose an algorithm for IEEE 802.11p-compliant receivers, based on orthogonal frequency-division multiplexing (OFDM). We employ iterative structures in the receiver as a way to estimate the channel despite variations within a frame. The channel estimator is based on factor graphs (FGs), which allow the design of soft iterative receivers while keeping acceptable computational complexity. Throughout this paper, we focus on designing a receiver that offers a good complexity-performance tradeoff. Moreover, we propose a scalable algorithm to be able to tune the tradeoff, depending on the channel conditions. Our algorithm allows reliable communications while offering a considerable decrease in computational complexity. In particular, numerical results show the tradeoff between complexity and performance measured in computational time and bit error rate (BER), as well as frame error rate (FER) achieved by various interpolation lengths used by the estimator, which both outperform by decades the standard least squares (LS) solution. Furthermore, our adaptive algorithm shows a considerable improvement in terms of computational time and complexity against state-of-the-art and classical receptors while showing acceptable BER and FER performance. QC 20151021 |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|