Research On Cooperative-based Resource Allocation For Cognitive Radio Networks

With the rapid development of wireless mobile applications, the spectrum and energy resources have become severely scarce. Cognitive radio (CR) en-abling the dynamic and feasible access to authorized spectrum resources, wire-less energy transfer (WET) allowing the adaptive flow and re-allocation of en-ergy among devices, and wireless relay reducing the resource consumption for wireless data transmission significantly, have been widely acknowledged by both the academia and industry, and will play important roles in the 5G wire-less communication networks. As an effective integration of these three tech-nologies, WET-assisted cognitive cooperation aims at designing a more effi-cient reuse and allocation paradigm for both the spectrum and energy resources between primary networks and secondary networks, resulting in a notable im-provement on the overall resource utilization and a win-win solution for primary and secondary networks. In order to exploit the potential performance gain of spectrum/energy cooperation in cognitive radio networks (CRNs), the resource allocation strategies are critical, and the corresponding researches have attracted many researchers' attention.However, the research in this field is still in its infancy, and the estab-lished research models and optimization algorithms are relatively simple. In this paper, a further study on spectrum efficient and energy efficient resource allocation for cognitive cooperation is conducted. Firstly, based on the exist-ing One-plus-One cooperation model where one secondary users (SUs) coop-erates with one primary user (PU), this paper considers the possibility that one SU cooperates with multiple PUs to acquire more spectrum access opportuni-ties, and hence establishes a novel model, i.e., Multiple-plus-One cooperation model. Moreover, based on the traditional spectrum cooperation scheme in CRNs, this paper takes the energy cooperation between PUs and SUs into ac-count, and proposes a novel cooperation scheme, i.e. joint spectrum and energy cooperation (JSEC) scheme. Besides, different technologies, e.g., orthogonal frequency division multiplexing (OFDM), spatial multiplexing (SM), network coding (NC) and so on, are jointly employed to further improve the resource utilization in cognitive cooperation networks according to the specific appli-cation scenarios. Finally, by referring to the convex optimization and matrix optimization, the low-complexity and optimal resource allocation algorithms are devised for each cooperation model and cooperation scheme by means of various optimization methods, including the dual transformation, fixed point iteration, semi-definite relaxation, one-dimensional linear searching and so on.The studied models, proposed schemes and designed algorithms in this paper are meaningful and provide a good guidance for the suitable deployment of CRNs and the implementation of sophisticated resource allocation algorithms in the future 5G wireless communication systems.The main contents and contributions of this paper are summarized as fol-lows.Firstly, in order to increase the spectrum access opportunities for SU, this paper proposes a new cooperation model in which one SU, often equipped with multiple antennas, cooperates with two PUs. Meanwhile, to reduce the resource consumption for forwarding two PUs' data, NC is introduced to compress the data size to forward at SU. For this kind of NC-assisted Two-plus-One coop-eration model, we analyze the basic condition that one SU and two PUs are willing and capable to reach this three-party cooperation, and explain the duty as well as the strategies for each participant. Afterwards, the mathematic opti-mization problems are formulated for both flat-fading and frequency-selective-fading channel model with the aim of maximizing the SU's transmission rate while guaranteeing the successful data transmission for PUs. For flat-fading model, the Karush-Kuhn-Tucker (KKT) is adopted to design the optimal band and power allocation algorithm. For frequency-selective-fading model, the greedy-based subcarrier assignment algorithm and dual-transformation-based optimal power allocation algorithm are devised by decoupling these two opti-mization variables. The simulation results show that compared with the One-plus-One cooperation model, the proposed NC-assisted Two-plus-One model brings a significant improvement on the system resource utilization.Secondly, based on the traditional spectrum-only cooperation (SoC) scheme,this paper considers the wireless energy transfer from PUs to SUs and builds the novel joint spectrum and energy cooperation (JSEC) scheme, as well as proposes the detailed resource allocation algorithms. In particular, SU passes its received radio frequency (RF) signals from PU to a power splitter and har-vests parts of the signal energy to reduce its energy consumption, and in the meantime, employs spatial multiplexing to perform simultaneous signal relay-ing for PU and its own data delivery. The optimization model is established for the proposed JSEC scheme in the cases of decode-and-forward (DF) mode and amplify-and-forward (AF) mode, which aims at maximizing the achievable transmission rate of SU subject to the minimum transmission rate requirement for PU by jointly optimizing the PS factor, the signal forwarding vector for PU and the precoding vector for SU. Then the optimal resource allocation algo-rithms are also devised for DF and AF model, in which we first build the con-trapositive power minimization problem and employ the dual method to obtain the optimal forwarding vector and precoding vector, and then analyze the mono-tonic property or quasi-convex/quasi-concave property of the objective function and propose a bisection algorithm to acquire the optimal transmission rate for SU and the optimal PS factor. Simulation is performed for the proposed JSEC scheme and the corresponding resource allocation algorithms, through which their superiority over traditional SoC schemes is verified.Thirdly, by combining the aforementioned two research contents, this pa-per further studies Two-plus-One JSEC model: One one hand, the SU cooper-ates with two PUs to acquire more spectrum access opportunities, and on the other hand, SU employs the power splitting technology to harvest energy from RF signals transmitted by PUs. For this Two-plus-One JSEC model, we inves-tigate the detailed resource allocation for AF relay mode, DF relay mode and NC-assisted mode respectively, and establish the corresponding optimization models for these three relay modes. Our objective is to maximize the SU's rate by optimizing the PS factor, the forwarding vector and the precoding vector.For AF mode, the contrapositive power minimization problem is built to solve the optimal forwarding vector and precoding vector. Then, we prove that the objective function is quasi-concave about the PS factor and utilize the golden section searching method to find the optimal PS factor. For DF and NC-assisted modes, the optimal PS factor is acquired by the bisection searching method, fol-lowing which the optimal forwarding and precoding vectors are solved by either the dual transformation plus fixed point iteration method or the semi-definite relaxation plus rank reduction method. Through numerical simulations, it is shown that the proposed Two-plus-One JSEC scheme is an enhanced version of the One-plus-One JSEC scheme and Two-plus-One SoC scheme, resulting in a significant improvement on the data transmission reliability for PU and the achievable transmission rate for SU, and is of great application value in practice.