29 September 2009 in IEEE Journal on Selected Areas in Communications
In this paper, we propose a deflection routing scheme that improves effective throughput (defined as the successfully transmitted bits over the duration between two available sequential time slots) of millimeter-wave wireless personal area network (mmWave WPAN) systems. The upcoming mmWave WPAN is based on dynamic time division multiple access (TDMA) and designed to guarantee Gbps-order transmission capability for high definition TV (HDTV) transmission, high speed wireless docking and gaming, etc. The decode-and-forward (DF) type of relay offers a simple solution to the issues of mmWave WPAN systems, such as limited coverage range and unexpected blockage. However, due to the required extra time, DF relay on the other hand decreases the effective throughput, and may not be sufficient to satisfy the requirement of the above data-rate-greedy applications. Inspired by the fact that the significant path loss of a millimeter-wave environment can provide good space isolation, we propose a deflection routing scheme to improve the effective throughput by sharing time slots for direct path with relay path. Based on the sub-exhaustive search, a routing algorithm, named as best fit deflection routing (BFDR), has been developed to find the relay path with the least interference that maximizes the system throughput. To reduce the computational complexity of the BFDR, we have also developed a sub-optimal algorithm named as random fit deflection routing (RFDR). The RFDR algorithm finds the sub-optimized relay path, where the interference may not be the least but is sufficiently low to guarantee the concurrent transmissions. Computer simulations show that, in realistic 60 GHz environments, the effective system throughput can be improved up to 28% under grid topology and 35% under random topology. RFDR achieves almost the same order of throughput improvement with only 10% of the computational complexity of BFDR.
28 June 2012 in IEEE Transactions on Vehicular Technology
In this paper, we address the following two types of coexistence between wireless networks: 1) coexistence among secondary user (SU) networks and 2) coexistence between primary user (PU) and SU networks. The former type addresses the sharing of a common spectrum opportunity among SU networks to achieve an efficient spectrum usage, whereas the latter type addresses the interference from SUs to PUs. Following the trend of spectrum regulations and industrial standardization, we postulate a potential deployment scenario and describe the management procedure of SU networks under the interference constraints to the PU receivers. For the quantitative evaluation of coexistence, we first propose a new parameter called the quality of coexistence ( QoC). Then, we introduce a novel coexistence scheme called interference-constrained time slot splitting and allocation. In addition, we derive the distribution of the interference to the PU receiver caused by SU transmitters weighted by the time slot fraction assigned to each SU network. After optimizing individual SU transmission under the PU interference constraint, we use the analytical model as a basis to obtain the optimal time slot allocation ratios of multiple SU networks by maximizing the aforementioned QoC. This leads to an efficient spectrum usage by SUs without incurring harmful interference to the PUs. Finally, we compare the proposed scheme with other schemes that select a single SU network to operate based on simplistic criteria. It is shown that when the SU network employs adaptive transmit power control, the medium value of QoC is doubled using the optimal time slot allocation, resulting in significant improvement in coexistence.
1 January 2011 in IEICE Transactions on Communications