Discrete-Time Markov Chain Analysis of Energy Efficiency in a CR Network Regarding Primary and Secondary Traffic With Primary User Returns

In cognitive radio (CR) networks coexistence between primary and secondary networks is allowed to exploit the spectrum bands more efficiently. To have a green CR network, energy efficiency (EE) as well as collision probability between such networks are vital key performance. While, it is desired to improve EE, the latter should be decreased as much as possible. Therefore, interference between the primary user (PU) and secondary user (SU) is of high importance. Two kinds of interference are unavoidable. They are sensing error and PU re-occupancies interferences. While most studies focus on the sensing error interference, with the dynamic PU, the latter is also quite intense. Moreover, a precise investigation on the effect of such interference necessitates considering the PU and SU traffic behavior. Therefore, in this paper a new analytical framework is proposed in which the EE and collision probability are formulated. The traffic behaviors of PU and a general SU traffic are also considered. Using renewal reward theory, EE, and collision probability are then extracted and investigated for different dynamic rates of the PU traffic. For this, sensing error interference, PU-re-occupancy interference, and total energy consumption are derived in closed form expressions. The queue stability of SU's buffer is also investigated. Finally, simulation results are provided to justify the theoretical results. Suitable values for sensing and transmission times are obtained to maximize the EE under collision and queue stability constraints.