Joint versus separate spectrum sensing and resource allocation in OFDMA-based cognitive radio networks

In this study, the authors investigate the resource allocation issue for sensing-based orthogonal frequency-division multiple access (OFDMA) cognitive radio networks. They consider a network consisting multiple secondary users (SUs) and a secondary base station (BS) implementing a two-phase protocol. In the first phase, cooperative spectrum sensing is carried out to detect the vacant subchannels. In the second phase, SUs transmit data in the uplink to the BS by using OFDMA. They optimise the sensing parameters, transmit power and subchannel assignments jointly to minimise the total energy consumption with the constraints on SUs’ quality of service and detection probability of the primary user. This is a mixed binary integer programming problem which is NP (non-deterministic polynomial-time)-hard and generally intractable. They represent the problem as a bilevel problem and propose two efficient algorithms to solve the slave and master subproblems. They also study the separate optimisation, in which the sensing parameters of SUs are set regardless of the allocated resources. They investigate the energy savings of joint versus separate optimisation using numerical experiments. The results show that the joint optimisation method can introduce up to 16% of energy saving in zero sensing signal-to-noise ratio with the same total transmission bandwidth of 2.5 MHz.