Radio Resource Management in LTE-Advanced Systems with Carrier Aggregation
In order to meet the ever-increasing demand for wireless broadband services from fast growing mobile users, the Long Term Evolution -Advanced (LTE-A) standard has been proposed to effectively improve the system capacity and the spectral efficiency for the fourth-generation (4G) wireless mobile communications. Many advanced techniques are incorporated in LTE-A systems to jointly ameliorate system performance, among which Carrier Aggregation (CA) is considered as one of the most promising improvements that has profound significance even in the upcoming 5G era. Component carriers (CCs) from various portions of the spectrum are logically concatenated to form a much larger virtual band, resulting in remarkable boosted system capacity and user data throughput. However, the unique features of CA have posed many emerging challenges as well as span-new opportunities on the Radio Resource Management (RRM) in the LTE-A systems. First, although multi-CC transmission can bring higher throughput, it may incur more intensive interference for each CC and more power consumption for users. Thus the performance gain of CA under different conditions needs fully evaluating. Besides, as CA offers flexible CC selection and cross-CC load balancing and scheduling, enhanced RRM strategies should be designed to further optimize the overall resource utilization. In addition, CA enables the frequency reuse on a CC resolution, adding another dimension to inter-cell interference management in heterogeneous networks (HetNets). New interference management mechanisms should be designed to take the advantage of CA. Last but not least, CA empowers the LTE-A systems to aggregate the licensed spectrum with the unlicensed spectrum, thus offering a capacity surge. Yet how to balance the traffic between licensed and unlicensed spectrum and how to achieve a harmony coexistence with other unlicensed systems are still open issues. To this end, the dissertation emphasizes on the new functionalities introduced by CA to optimize the RRM performance in LTE-A systems. The main objectives are four-fold: 1) to fully evaluate the benefits of CA from different perspectives under different conditions via both theoretical analysis and simulations; 2) to design cross-layer CC selection, packet scheduling and power control strategies to optimize the target performance; 3) to analytically model the interference of HetNets with CA and propose dynamic interference mitigation strategies in a CA scenario; and 4) to investigate the impact of LTE transmissions on other unlicensed systems and develop enhanced RRM mechanisms for harmony coexistence. To achieve these objectives, we first analyze the benefits of CA via investigating the user accommodation capabilities of the system in the downlink admission control process. The LTE-A users with CA capabilities and the legacy LTE users are considered. Analytical models are developed to derive the maximum number of users that can be admitted into the system given the user QoS requirements and traffic features. The results show that with only a slightly higher spectrum utilization, the system can admit as much as twice LTE-A users than LTE users when the user traffic is bursty. Second, we study the RRM in the single-tier LTE-A system and propose a cross-layer dynamic CC selection and power control strategy for uplink CA. Specifically, the uplink power offset effects caused by multi-CC transmission are considered. An estimation method for user bandwidth allocation is developed and a combinatorial optimization problem is formulated to improve the user throughput via maximizing the user power utilization. Third, we explore the interference management problem in multi-tier HetNets considering the CC-resolution frequency reuse. An analytical model is devised to capture the randomness behaviors of the femtocells exploiting the stochastic geometry theory. The interaction between the base stations of different tiers are formulated into a two-level Stackelberg game, and a backward induction method is exploited to obtain the Nash equilibrium. Last, we focus on the mechanism design for licensed and unlicensed spectrum aggregation. An LTE MAC protocol on unlicensed spectrum is developed considering the coexistence with the Wi-Fi systems. The protocol captures the asynchronous nature of Wi-Fi transmissions in time-slotted LTE frame structure and strike a tunable tradeoff between LTE and Wi-Fi performance. Analytical analysis is also presented to reveal the essential relation among different parameters of the two systems. In summary, the dissertation aims at fully evaluating the benefits of CA in different scenarios and making full use of the benefits to develop efficient and effective RRM strategies for better LTE-Advanced system performance.