The dissertation is dedicated to solving the scientific and technical problem of increasing the throughput of IEEE 802.11be multi-channel wireless networks through coordinated radio resource management, taking into account the complex impact of channel conditions, cross-link interference, and the constraints of multi-link operation (MLO) modes in IEEE 802.11be networks.
In the process of achieving the set goal and solving the scientific problem, the following main scientific results were obtained:
An improved channel estimation method for a multi-channel IEEE 802.11be device, which, unlike existing methods, is based on a comprehensive analysis of the radio environment state by introducing an integrated quality indicator and weighting coefficients that formalize current spectrum utilization, interference dynamics, and the impact of previous generation Wi-Fi devices, allowing for the assessment and justified selection of optimal frequency configurations and channels.
A newly developed mathematical model of traffic distribution for a multi-channel device, which is based on representing the device as a single queuing system, where the environment state and MLO constraints are formalized by a state vector and a decision matrix, and the justification for selection is achieved by establishing an analytical relationship between these control parameters and the expected quality of service indicators, which enables decision-making regarding the feasibility of utilizing traffic distribution policies for the current scheduling cycle during multi-channel operation.
A newly developed resource allocation method between channels of a multi-channel device, which integrates the results of the radio environment analysis and the traffic distribution model by combining data flow profiling by access categories with the prediction of their radio resource occupancy time and the optimization of the load distribution matrix, which ensures coordinated channel management to increase total throughput and minimize packet delivery latency during multi-channel operation.
In the first chapter, an analysis of resource management problems in IEEE 802.11be multi-link wireless networks under conditions of increasing load, spectrum fragmentation, cross-link interaction, and hardware constraints of MLO modes is carried out. Based on the analysis of multi-link device architecture, multi-channel operation modes, as well as medium access mechanisms and resource utilization, it is established that the performance of such networks is determined not by individual physical layer parameters, but by the consistency of time and frequency resource management at the MAC layer. The chapter systematizes factors reducing multi-channel transmission efficiency, demonstrates the limitations of local approaches to channel selection, and formulates the scientific task of the research.
In the second chapter, the problem of evaluating the spectral configuration of a multi-channel device under conditions where channel selection fails to provide efficient resource utilization due to mutual interference and hardware MLO constraints is solved. For this purpose, a mathematical model of spectral configuration was developed, and a methodology for multi-criteria channel state evaluation was improved based on an integrated quality metric that considers radio link quality, interference level, bandwidth, resource unit allocation, and cross-link effects.
In the third chapter, the problem of constructing a mathematical model for traffic distribution in a multi-link device, necessary for the formal description of the system state and the rules for using available links in a stochastic environment, is solved. Based on queuing theory, a mathematical model of a multi-link device with time discretization by decision cycles was built, a system state vector based on normalized link metrics and queue parameters was introduced, traffic distribution policies were formalized as a mapping of the state to a control decision, the space of feasible decisions considering MLO constraints was determined, and an analytical relationship between control parameters and quality of service indicators was established.
In the fourth chapter, the problem of organizing resource allocation in multi-link networks under conditions of dynamic changes in channel states, traffic characteristics, time overheads, and hardware synchronization constraints is solved. For this, a resource allocation methodology is proposed, implemented as a functional block at the U-MAC level for a closed control loop, within which link state monitoring and evaluation, flow profiling by access categories, channel occupancy time prediction, and load distribution matrix optimization based on the criterion of minimizing the maximum transmission completion time are performed.
