MICPA: a client-assisted channel assignment scheme for throughput enhancement in WLANs

The emergence of high-density wireless local area network (WLAN) deployments in recent years is a testament to the insatiable demands for wireless broadband services. The increased density of WLAN deployments brings with it the potential of increased capacity, extended coverage, and exciting new applications. However, the corresponding increase in contention and interference can significantly degrade throughputs, unless new challenges in channel assignment are effectively addressed. In this paper, a client-assisted channel assignment scheme that can provide enhanced throughput is proposed. A study on the impact of interference on throughput with multiple access points (APs)is first undertaken using a novel approach that determines the possibility of parallel transmissions. A metric with a good correlation to the throughput, i.e., the number of conflict pairs, is used in the client-assisted minimum conflict pairs (MICPA) scheme. In this scheme, measurements from clients are used to assist the AP in determining the channel with the minimum number of conflict pairs to maximize its expected throughput. Simulation results show that the client-assisted MICPA scheme can provide meaningful throughput improvements over other schemes that only utilize the AP’s measurements.

Numerical simulations of intense tropical storms

A high resolution general circulation model has been used to study intense tropical storms. A five-year-long global integration with a spatial resolution of 125 km has been analysed. The geographical and seasonal distribution of tropical storms agrees remarkably well with observations. The structure of individual storms also agrees with observations, but the storms are generally more extensive in coverage and less extreme than the observed ones. A few additional calculations have also been done by a very high resolution limited-area version of the same model, where the boundary conditions successively have been interpolated from the global model. These results are very realistic in many details of the structure of the storms including simulated rain-bands and an eye structure. The global model has also been used in another five-year integration to study the influence of greenhouse warming. The sea surface temperatures have been taken from a transient climate change experiment carried out with a low resolution coupled ocean-atmosphere model. The result is a significant reduction in the number of hurricanes, particularly in the Southern Hemisphere. Main reasons for this can be found in changes in the largescale circulation, i.e. a weakening of the Hadley circulation, and a more intense warming of the upper tropical troposphere. A similar effect can be seen during warm ENSO events, where fewer North Atlantic hurricanes have been reported.

Numerical prediction of atmospheric blocking - A case study

One of the fundamental questions in dynamical meteorology, and one of the basic objectives of GARP, is to determine the predictability of the atmosphere. In the early planning stage and preparation for GARP a number of theoretical and numerical studies were undertaken, indicating that there existed an inherent unpredictability in the atmosphere which even with the most ideal observing system would limit useful weather forecasting to 2-3 weeks.