Search

The ultra-dense network (UDN) has been identified as an appealing solution to address the huge service demands in future 5G and beyond. In this paper, we propose an optimization design of UDN balancing user mobility and network densification, where the massive users are divided into different groups according to their moving speeds, and different groups are served by different subnets. This way, the UDN can adjust subnet parameters to meet the service demands using minimum resource. Specifically, we first present the objective of the UDN optimization, then look at the area spectral efficiency of each individual subnet if selected to serve users within a speed range, and finally carry out the joint optimization across multiple subnets. Numerical examples demonstrate that the proposed approach is effective and practical, meeting the service demands with the least bandwidth resource consumption.

This study uses diffusion weighted NMR to determine yeast cell size at different times after those cells were suspended in water. Both the water diffusion and the metabolite diffusion where used in those measurements. Diffusion weighted NMR is a well-studied noninvasive tool that can be used to study various microstructures. Yeast is a cheap and versatile model system for more complex biological systems. To measure the metabolites they first had to be identified in the NMR spectrum of the yeast cells suspensions. The identification was done by extracting the metabolites from the yeast cells using ethanol. This allowed for identification of several specific metabolite signals, although there was still a lot of overlapping signals in the spectrum that could not be identified. Dried yeast, of the species Saccharomyces Cerevisiae, was washed and suspended in a 10 % D2O phosphate buffer. These suspensions were measured using several PFGSE NMR experiments, with varying diffusion times, and with and without water suppression. The data from these measurements were used to determine time dependent diffusion coefficients. Because of the restrictions on the diffusion of molecules due to the cell wall, there is a decline in time dependent diffusion coefficient with increasing diffusion time, and this decline can be used to determine the cell size. The metabolite measurements of cell size in the yeast samples show apparent growth from around 1,9 μm at 3 days suspended to around 3,5 μm at 12 days suspended. The intracellular water signal did not show this growth, staying around 1,6 μm. These values all fit within a reasonable range of known values for the cell radius of Saccharomyces Cerevisiae yeast, which is between 1,9 and 3,3 μm. [31] The growth in cell size can be attributed to the cells actually swelling, or to the cells dying and the cell walls becoming more permeable. The lack of growth in the water signal means that the water signal is likely less sensitive towards changes in microstructure. This might be because of influence from the extracellular water signal, or because there is less relative change in cell membrane permeability, as water already had comparatively rapid cell membrane exchange, as compared to the metabolites Cell size was also measured in a layered sample using slice selective measurements. One layer had yeast that had been suspended 12 days and the other had yeast that had been suspended 3 days. The slice selective diffusion measurements gave cell sizes that were larger than the cell sizes from the non-selective measurements, around 4 μm. It is difficult to determine the reason for the larger cell, but a gradient may not have been calibrated correctly at the location that was measured. The slice selective measurements also had larger uncertainties than the non-slice selective measurements. The larger uncertainties was likely from equipment of lower quality or the measurements having different geometry. Masteroppgave i kjemi KJEM399 MAMN-KJEM

It is shown that Enhanced UMTS will be an affordable solution for providing the required network quality and to reduce infrastructure investments in offices scenarios. System capacity results are obtained by using a system level simulator which considers traffic characterisation parameters and services usage in detail, among other. Results for the most profitable cell radius are obtained via an optimisation procedure based in economic aspects. A higher number of pico cells (with a smaller radius, around 30-32 m) can be installed in the future, when costs of deploying and maintaining the network decreases, allowing for supporting higher system capacity, and reducing prices. Our approach is based in a detailed services analysis, which represents a worst case situation relatively to the total services approach, because the later does not discriminate results for the different traffic classes. The impact of call blocking, handover failure, end-to-end delay, and delay variation are taken into account. © 2006 IEEE. Conference code: 72218 Cited By :2

Orthogonal multiple access/multiplexing (OFDMA/OFDM) techniques represent the radio interface technologies for the next generation broadband access networks. WiMAX system adopts them for point-to-multipoint transmission. In a typical cell radius deployment of one to tree kilometers, WiMAX systems can be expected to deliver capacity of up to 40 Mbps. However the co-channel interference due to cell frequency reuse can lead to decreasing performance. In particular, throughput and BER can get worse drastically. In this paper we deal the evaluation of interference in a WiMax cellular system. Analysis is carried out using a semi-analytical approach based on a general expression of the SINR accounting for multipath, frequency and time offsets between the reference terminal and the interfering devices. Performance have been obtained under different operating modes of the OFDMA/OFDM cellular network that is synchronous and asynchronous BSs.