Implementing an adaptive bandwidth management scheme for efficient use of spectrum

Abstract

Licensing regimes grant entities, such as cellular companies, exclusive rights to portions of the wireless spectrum and leave a small segment of unlicensed spectrum for public use. This method of spectrum management has lead to inefficiency in spectrum usage since not all what is allocated can be used at all times. While the frequency spectrum is a limited resource, and yet more spectrum resource is needed for faster growth of the wireless technologies, it is important to make efficient use of both the licensed and unlicensed bands.

This research focused on improving spectrum utilization and efficiency in the unlicensed band. The study focused on spectrum optimization in WLANs to solve the artificial spectral shortage problem, in order to support ever-growing user demands for wireless applications. Appropriate number of channels were created to accommodate all potentially interfering APs, and to adjust the channel bandwidth (channel-width) depending on the instantaneous number of users/traffic associated to the neighboring APs.

In this dissertation the Dynamic Spectrum Allocation model and Greedy raising algorithm were adopted to optimize spectrum allocations. The simulations adaptively divided the available spectrum of 64 KHz into non-overlapping frequency segments of different bandwidth considering the number of users in a local environment to achieve the following research objectives: (i) To maximize spectrum efficiency by increasing parallel transmissions and reducing co-channel interferences; (ii) To improve fairness across a network by balancing spectrum assignments.

The simulation results showed that implementing dynamic channelization improves not only network capacity but per-client fairness by allocating channels of variable width for access points in a WLAN. From the simulation results of dynamic channelization we can confidently conclude that a solid foundation for efficient management of the precious spectrum resource especially in large WLANs has been set.