Resource Optimization Strategy In Cognitive Radio And Cooperative Networks

The surge in demand for high data rates on mobile devices is driving the current imperative for either improving spectral efficiency on the existing spectrum and/or additional spectrum allocations. Cognitive radios (CR) and cooperative transmission have been regarded as two promising techniques to satisfy this demand and extensively studied. CR performing dynamic spectrum access seems to be a natural pathway for realizing improved spectrum efficiencies by allowing secondary users (SUs) on hitherto licensed spectrum. While, cooperative transmission in relay-based networks realize spectrum efficiencies enhancement and enlarge the coverage of the networks by assisting sources forwarding the message to destinations. To achieve higher spectrum efficiencies, the dessertation has investigated the resource optimization strategy in both CR and cooperative transmission networks. The major work and contributions of this dessertation consist in:1. Improved Idle Channel Utilization in Distributed Multi-channel Cognitive Radio Systems. In the absence of a control channel, the idle channel utilization (ICU) is dictated by both the spectrum sensing strategy and the packet access protocol in a distributed CR network. In the disseration, we investigate a distributed multi-channel random access CR network under both perfect and imperfect spectrum sensing, where each SU randomly selects a fixed number of channels for sensing, termed as fixed time spectrum sensing (FTSS), and transmits on arbitrary detected idle channels. Based on this model, we first develop a slotted opportunistic access aloha protocol for multiuser random access. Then the probability of the idle channel selected for access are derived and a range for the number of channels to be sensed per SU access in perfect sensing is obtaind. Based on that, we derive the distribution of the successful packet transmissions and attain the expression of the average ICU. Finally, the analytical results are validated by substantial simulations. We show that their impacts on ICU by adjusting the number of the sensed channels and packet transmission probability, along with the distance from SUs to primary users.2. Throughput Analysis for a Multi-user, Multi-channel ALOHA Cognitive Radio System. First, we derive the analytical average system throughput based on FTSS for cases where the number of idle channels are fixed and a random variable, respectively. Based on that, a relationship between the average system throughput and the number of sensing channels is attained. Subsequently, we formulate a joint optimization problem in order to maximize average system throughput. The analytical results are validated by substantial simulations.Second, due to the absence of a control channel, system parameters such as the number of active SU and the number of idle channels are generally unknown to an accessing terminal in distributed CR networks. Hence, we propose a novel method for estimating these key parameters based on FTSS. The estimates can be used to adjust the packet transmission rates and the number of the sensed channels, consequently to improve the overall throughput of the distributed CR network. The performance of the estimation method is evaluated through substantial simulations.3. Achievable Rate Analysis for Multi-channel Cognitive Ad Hoc Networks. We analyze the average achievable rate of the multi-channel cognitive ad hoc network which consists of the finite number of SU coexisting with PU in a unit (finite) area. Without the control channel, the average achievable rate of the network is investigated and affected by the spectrum sensing and access strategy and the distribution of SUs in the network. Accordingly, we first obtain the average number of the idle channels selected by a certain amount of SUs and determine the range of the number of sensed channels. Secondly, the average achievable rate for one SU pair with co-channel SUs interference is derived. The average achievable rate of the network is then arrived and depends on the number of the sensed channels. Finally, the analytical results are validated by substantial simulations and show the optimal number of the sensed channels.4. Cooperative Strategy in OFDM-based Relay Networks. First, we consider a two-hop orthogonal frequency division multiplexing (OFDM) relay link, and investigate the system capacity in hybrid forward (HF) relay through the power allocation (PA). Two situations about the power constraint are concerned, and the selection criterion among different forwarding schemes in HF relay is proposed under each situation, by means of capacity comparison on the subcarrier pair. From the selection criterions, the source or relay adaptively choose the best scheme for every subcarrier pair according to the channel condition. To maximize the capacity, the optimal solutions of the relay PA and the joint PA are attained. In the simulation, the advantage of the HF relay is shown in terms of capacity improvement compared with the conventional forward relay.Secondly, we consider a four-node relay-based Orthogonal Frequency Division Multiplexing Access (OFDMA) system. The expression of the upper bound for the achievable rate in Successive Relaying (SUR) protocol is derived by using the cut-set theorem for half-duplex systems. Based on this expression, the near-optimal solution of the achievable rate is obtained in the joint power and subcarrier allocation according to the dual problem decomposition approach and the subgradient method. Moreover, we make a comparison on the achievable rate between SUR and Simultaneous Relaying (SIR) in the pathloss model accepted by the3rd Generation Partnership Project (3GPP) Long Term Evolution (LTE) Advanced system. The simulation results demonstrate the enhancement of the achievable rate in SUR protocol is expanded in high signal-to-noise ratio (SNR) compared with the achievable rate in SIR protocol under the symmetric scenario. And for SUR protocol, the achievable rate is higher when the relays are close to the destination or the source.5. Joint Resource Allocation in OFDMA Cooperative Networks with User Classification. We focus on the joint resource allocation in the OFDMA cooperative network under the downlink transmission. Different from the existed research, we first propose the principle for user classification based on the reliable SNR at the UE side. According to the classification result, the expression of the weighted achievable rate of the network is obtained and set to be the optimization object. The optimization is divided into two phases,1) the optimal PA in the given subcarrier allocation;2) joint decision on the subcarrier allocation and relay selection, to maximize the weighted achievable rate of the network. Finally, the simulation results are illustrated to validate the analytical optimization results. Through the simulation with different weighted factors, we obtain the tradeoff between the fairness for the classified UE and the achievable rate of the network.