Uplink resource management for LTE-based IoT networks

Abstract

In this dissertation, the uplink resource management schemes for LTE-based internet of things (IoT) network were studied. Unlike the downlink transmission, where all the IoT devices can synchronize with the eNodeB through a synchronization channel (SCH) transmitted from the eNodeB, uplink transmissions from IoT devices are not initially synchronized because of the round trip delay, which varies depending on the location of each IoT device. Therefore, in order for IoT devices who are not connected to the eNodeB and try to establish a uplink with the eNodeB, non-synchronized random access is required where a resource request and an uplink synchronization acquisition are simultaneously performed at the eNodeB. Since it is not possible to know which subset of IoT devices are going to request resource at a certain moment in advance, the uplink random access should be operated in a contention-based manner. Therefore, it is worthwhile to investigate random access schemes that can improve the random access procedure, where the random access procedure is considered as one of the main bottlenecks for supporting IoT networks with numerous IoT devices. Therefore, in this dissertation, we propose two random access schemes which are compatible with the current LTE yet superior to the current LTE and other previously proposed LTE-based RA schemes. Both random access schemes utilized the fact that the IoT devices are likely to have no mobility. In detail, the proposed random access schemes utilize the round-trip delay time which is assumed to be known to each device. It is worth mentioning that both schemes are compatible with the current LTE systems with minor software modification or update. Among two newly proposed random access schemes, the first is named energy efficient random access (EERA). EERA not only increases the number of IoT devices that can be accommodated with the same amount of resources, but also makes IoT devices to consume less energy than existing random access schemes. The second is named time-shifted random access (TSRA). TSRA utilizes the characteristics of the Zadoff-Chu (ZC) sequences, where ZC sequences are used in LTE due to its constant amplitude zero auto-correlation characteristics. In detail, TSRA utilizes a time-shifted preamble, where a ZC sequence is additionally cyclically shifted (or time-shifted) at each IoT device. While the EERA is the scheme that focuses on reducing energy consumption, the TSRA can be applied with various objectives: enhancing fairness, reducing energy consumption, even supporting prioritized heterogeneous IoT networks. In this dissertation, the collision probability of the proposed two random access schemes are mathematically analyzed. Also, the fairness and energy consumption are also derived. The mathematically derived equations are verified by computer simulations. Furthermore, through computer simulations, the performance of the proposed random access schemes are evaluated in terms of collision probability, the average random access time, the average energy consumption, and the number of IoT devices in each random access opportunity.