Goodput maximization in opportunistic spectrum access networks under constraints on the inter-packet transmission waiting time

详细信息    查看全文

• 作者：Maurizio A. Bonuccelli^a ; ^{bonucce@di.unipi.it" class="auth_mail" title="E-mail the corresponding author}Author Vitae ; Donatella Ermini^b ; ^{erminido@yahoo.it" class="auth_mail" title="E-mail the corresponding author}Author Vitae ; Loreto Pescosolido ; ¹ ; ^b ; ^{loreto.pescosolido@uniroma1.it" class="auth_mail" title="E-mail the corresponding author}Author Vitae ; Chiara Petrioli^b ; ^{petrioli@di.uniroma1.it" class="auth_mail" title="E-mail the corresponding author}Author Vitae
• 关键词：Cognitive radio ; Opportunistic Spectrum Access (OSA) ; Goodput ; Theory of recurrent events ; Spectrum sensing
• 刊名：Ad Hoc Networks
• 出版年：2016
• 出版时间：1 March 2016
• 年：2016
• 卷：38
• 期：Complete
• 页码：19-37
• 全文大小：2547 K

文摘

We consider the problem of selecting the packet size, the Forward Error Correction (FEC) coding rate, and the duration of the spectrum sensing interval, which maximize the communication goodput (payload bits correctly received per unit time) of the Secondary Users (SUs) of a cognitive radio network. We start by deriving a closed form expression of the SUs achievable goodput under the assumption of Markovian primary channel use and perfect spectrum sensing. Then, we show that the optimal packet size and FEC coding rate selection change in case we impose a constraint on the maximum time SUs can wait before successfully transmitting a packet, i.e., on what we call the inter-packet transmission waiting time. Such waiting time depends on the alternating PU presence on the channel. Finally, we derive upper and lower bounds on the achievable goodput, which allow us to optimally select the packet size, the FEC coding rate, and the duration of the spectrum sensing time interval, in case of imperfect spectrum sensing. Extensive simulation results are provided that validate our findings. Specifically, we show that by adaptively and jointly selecting the system parameters as a function of received SU power and PU traffic statistics, it is possible to achieve a goodput increase in the order of 20%.