Downlink Throughput Driven Channel Access Framework for Cognitive LTE Femto-Cells

TL;DR

This paper introduces an optimized channel access framework for LTE cognitive femto-cells that maximizes downlink throughput by intelligently sensing macro-cell channels and determining the optimal sensing interval, verified through analytical and simulation methods.

Contribution

It presents a novel spectrum sensing and access strategy with an analytical solution for optimal sensing interval to enhance femto-cell throughput in LTE networks.

Findings

Maximum throughput varies with macro-cell occupancy statistics.

Optimal sensing interval maximizes femto-cell throughput.

Throughput increases by around 15% at peak macro-cell traffic.

Abstract

This paper proposes an optimized sensing based channel access framework for the LTE cognitive femto-cells, with an objective of maximizing the femto-cells downlink throughput. Cognitive femto-cells opportunistically transmit on the macro-cell channels when they are free of use. Those free channels are located by means of spectrum sensing using energy detection. Moreover, periodic sensing is adopted to detect any changes of the sensing outcomes. The maximum attainable femto-cell downlink throughput varies with the macro-cell channel occupancy statistics. Therefore, the LTE macro-cell occupancy is empirically modeled using exponential distributions mixture. The LTE cognitive femto-cell downlink throughput is maximized by compromising the transmission efficiency, the explored spectrum opportunities and the interference from the macro-cell. An analytical solution for the optimal periodic sensing interval that maximizes the throughput is found and verified by simulations. The obtained results show that there is indeed a single periodic sensing interval value that maximizes the LTE cognitive femto-cell downlink throughput. At the peak of the macro-cell traffic, our framework increases the femto-cell throughput by around 15% compared to the senseless case. The impact of the available number of channels for opportunistic access is studied and no significant impact is found for more than three channels.