Cognitive Radio Context Problem & Contribution Results & Performance Evaluation École d’été RESCOM 2014 On Estimating the End-to-End Bandwidth in Multi-Transceiver Multi-Hop Cognitive Radio Networks Gentian Jakllari Guillaume Artero Gallardo Lucile Canourgues André-Luc Beylot May 13, 2014 G UILLAUME A RTERO ( RCF / T É SA ) Proprietary Information May 13, 2014 1 / 9
Cognitive Radio Context Problem & Contribution Results & Performance Evaluation A brief overview of Cognitive Radio networks (1/2) Emergence of Cognitive Radio networks 1 Software Defined Radio technologies 2 Increasing demand for wireless capacity G UILLAUME A RTERO ( RCF / T É SA ) Proprietary Information May 13, 2014 2 / 9
Cognitive Radio Context Problem & Contribution Results & Performance Evaluation A brief overview of Cognitive Radio networks (2/2) Cognitive Radio principle • Fixed Spectrum Allocation ⇒ Dynamic Spectrum Access • Opportunistic Spectrum Access based on Spectrum Sensing result G UILLAUME A RTERO ( RCF / T É SA ) Proprietary Information May 13, 2014 3 / 9
Cognitive Radio Context Problem & Contribution Results & Performance Evaluation Problem & Contribution (1/2) PU 2 CR 7 CR 8 CR 9 PU 4 CR 2 CR 1 CR 3 CR 4 CR 5 PU 1 CR 10 CR 11 PU 3 CR 6 CR 12 ⇒ Available end-to-end bandwidth estimation in TDMA-based multi-transceiver multi-hop cognitive radio networks G UILLAUME A RTERO ( RCF / T É SA ) Proprietary Information May 13, 2014 4 / 9
Cognitive Radio Context Problem & Contribution Results & Performance Evaluation Problem & Contribution (2/2) Cognitive Radio network characteristics • Intra-path interferences • TDMA access with Randomized slot scheduling throughput 𝑒 5 demand 𝑒 (1) (1) (1) (1) (1) ( 𝑇 1 , 𝑔 1 , 𝜚 1 ) ( 𝑇 2 , 𝑔 2 , 𝜚 2 ) ( 𝑇 3 , 𝑔 3 , 𝜚 3 ) ( 𝑇 4 , 𝑔 4 , 𝜚 4 ) ( 𝑇 5 , 𝑔 5 , 𝜚 5 ) forward 𝑒 1 forward 𝑒 2 forward 𝑒 3 forward 𝑒 4 forward 𝑒 5 Main Contribution ⇒ Centralized Exponential Time to Distributed Linear Time algorithm G UILLAUME A RTERO ( RCF / T É SA ) Proprietary Information May 13, 2014 5 / 9
Cognitive Radio Context Problem & Contribution Results & Performance Evaluation Results & Performance Evaluation Achievable End-to-End Throughput 500 p a = 33%, N H = 4 Exact end-to-end throughput (kbps) 450 p a = 50%, N H = 4 400 350 300 250 200 150 100 50 0 0 50 100 150 200 250 300 350 400 450 500 Computed end-to-end throughput (kbps) F IGURE 1: Numerical verification for u = 10 % of the correctness of the algorithm for computing the average throughput. G UILLAUME A RTERO ( RCF / T É SA ) Proprietary Information May 13, 2014 6 / 9
Cognitive Radio Context Problem & Contribution Results & Performance Evaluation Results & Performance Evaluation Admission Control Performance 600 N H = , u = 10% 4 Admitted rate in practice (kbps) N H = 10, u = 10% 500 400 300 200 100 0 0 100 200 300 400 500 600 Available end-to-end bandwidth (kbps) F IGURE 2: Admission control performance for the case of u = 10 % after only 10 numerical experiments. G UILLAUME A RTERO ( RCF / T É SA ) Proprietary Information May 13, 2014 7 / 9
Cognitive Radio Context Problem & Contribution Results & Performance Evaluation Results & Performance Evaluation Cognitive Radio Effect 1000 Admitted rate in practice (kbps) BRAND, N = 4 H 900 Zhu, N H = 4 800 700 600 500 400 300 200 100 0 0 100 200 300 400 500 600 700 800 900 1000 Available end-to-end bandwidth (kbps) F IGURE 3: Evaluation of both PU activity and multi-rates effects for the cases of u = 20 % . G UILLAUME A RTERO ( RCF / T É SA ) Proprietary Information May 13, 2014 8 / 9
Cognitive Radio Context Problem & Contribution Results & Performance Evaluation Questions ? G UILLAUME A RTERO ( RCF / T É SA ) Proprietary Information May 13, 2014 9 / 9
Recommend
More recommend
