Research On Radio Resource Allocation In CoMP-based Dense Networks

The contradiction between rapidly growing demands of mobile communications and limited radio resources for transmissions is becoming increasingly outstanding. One of the major solutions to address this contradication is to improve the utilization of available radio resources. Coordinated Multiple Points(Co MP) based dense network is one of the key candidate schemes for mobile communications systems to improve utility efficiency of radio resources, due to its flexibility in d eployment, capability to combat with intercell interference and less pathloss duri ng transmissions. However, compared to conventional mobile communication s systems, radio resource management in Co MP-based dense network is much more difficult, due to introducing of numerous low-power Access Points(APs) and complex cooperation mechanisms between APs. Rational and effective radio resource management and optimization is the precondition for maximizing system performances. Therefore, it is necessary to make a deep study on radio resource optimization a ccording to the characteristics of Co MP-based dense networks.In this dissertation, we focused on the contradiction between increasing demands and constrained resources in mobile communications systems. For the purpose of improving utility efficiency of radio resource, we investigated on jointly resources allocation in Co MP-based dense network. Specially, we analyzed the characteristics of Co MP transmissions and the principle of its ability to combat with intercell interference, formulated the resource optimization problem under multiple practical constraints in mathematics, and proposed spectral-efficient and energyefficient algorithms of joint resource allocation, which were proved to be effective to improve the performance of dense networks. The main line of this dissertation is: at the first place, we established the transmission model with Co MP schemes in a multicell scenario, and analyzed the problem of joint resource allocation under the model; then we expanded multicell scenario into a complex dense network, and analyzed the problem of joint resource allocation in aspects of spectrum efficiency and energy efficiency, respectively. The major content of this dissertation includes:(1) First, we formulated Co MP transmissions in mathematics, and analyzed their capacity in a multicell scenario. Transmission models including Dynamic Cell Selection(DCS) and Joint Transmission(JT) Co MP techniques were built for est imating link capacity. The probability density function of Signal-to-Interference Ratios(SIRs) in non Co MP, DCS Co MP and JT Co MP systems were derived, and capacities in the three systems were estimated based on SIRs. Simulation results showed that the link capacity was tightly related with the numbers of transmit antennas, Base Stations(BSs) participating in Co MP transmission, and the distance between BSs and users.(2) Second, we studied on resource allocation problems in a multicell scenario based on Co MP. Based on the constructed Co MP transmission model s, we mathematically formulated a joint optimization problem involving both spectr um and power resources, which aims at improving spectrum efficiency while maintaining data rate fairness among users. To deal with the formulated problem, we first pr oposed a Resource Block(RB) scheduling algorithm through balancing the tradeoff between the diversity gain and the multiplexing gain, by which the spatial resource can be allocatated flexibly to further improve system performance. Then, on the basis of that, we proposed a power allocation algorithm based on Newton methods, and realized joint allocation of spectrum and power resources.(3) Third, we studied on spectrum-efficient resource allocation problem in a dense network based on Co MP technique. Based on the research results mentioned above, we expanded the multicell scenario into a dense network, and discussed the impact on the performance of Co MP transmissions caused by capacity of backhaul links in practice. We also formulated the optimization problem of joint resource a llocation in the dense network in mathematics, and proposed scheduling algorithms for DCS Co MP and dynamic JT Co MP systems, respectively, as well as a low-complexity power allocation algorithm.(4) Forth, we studied on energy-efficient resource allocation in a dense network based on Co MP technique. The problem with multiple optimization objectives involving energy efficiency and fairness was formulated in mathematics. Energy efficiency was defined as the obtained data rate when consuming 1 Watt power. In addition to the constraints of transmission power and backhaul capacity, the formulation also considered a given Qo S requirement in order to avoid the extreme situ ation where unqualified transmissions occur for the purpose of saving energy. A Cross Entropy(CE)-based scheduling algorithm was proposed. We also proposed two modes for practicing the proposed scheduling algorithm, which were the ce ntralized and the decentralized. In order to further improve the system performance, the respecitve power allocation algorithm was also proposed for both practicing modes.