Next generation mobile communication systems will have to support high data rate traffic. This traffic is predicted to be asymmetric since there will be higher demand for the bandwidth on the downlink (DL), than on the uplink (UL). Furthermore, this traffic asymmetry will be dynamic, which means that the traffic asymmetry ratio (DL/UL) will not be static over time.
Time division duplex (TDD) has the advantage to be able to adjust the number of allocated time slots in the DL and in the UL to the actual traffic requirement. Frequency division duplexing (FDD) on the other hand has a fixed DL/UL resource allocation and therefore cannot support varying traffic asymmetry ratio.
However, TDD systems are subject to additional intercellular interference if the DL/UL operation is not synchronized in all the cells in the system. This causes a challenge when different traffic asymmetry ratios are to be supported in different cells. This additional interference is the same-entity (base station to base station and mobile station to mobile station) interference. On the other hand, if DL/UL operation is synchronized in all the system, we experience different-entity (base station to mobile and mobile to base station) interference.
We first propose a time slot allocation (TSA) scheme which reduces the overall intercellular interference for symmetric traffic. The proposed TSA scheme is compared to the traditional all-synchronous TSA. Both mathematical analysis and simulation are used in this comparison.
Next, the proposed TSA is adapted to asymmetric traffic with different asymmetry ratios in different cells. Two adaptation strategies are proposed and simulations are performed to investigate signal-to-interference ratio (SIR) outage probability dropping rate,and throughput performances of the strategies.
Finally, TSA in hybrid division duplex (HDD) cellular systems is investigated. In HDD systems, time slot allocation in time division duplex (TDD) mode plays a key role for efficient use of radio resource. Various types of time slot allocation strategies for this special TDD op-eration are investigated. SIR and outage probability performances of these strategies are compared through extensive simulations and analysis.
