On spectrum sharing games

Efficient spectrum-sharing mechanisms are crucial to alleviate the bandwidth limitation in wireless networks. In this paper, we consider the following question: can free spectrum be shared efficiently? We study this problem in the context of 802.11 or WiFi networks. Each access point (AP) in a WiFi network must be assigned a channel for it to service users. There are only finitely many possible channels that can be assigned. Moreover, neighboring access points must use different channels so as to avoid interference. Currently these channels are assigned by administrators who carefully consider channel conflicts and network loads. Channel conflicts among APs operated by different entities are currently resolved in an ad hoc manner (i.e., not in a coordinated way) or not resolved at all. We view the channel assignment problem as a game, where the players are the service providers and APs are acquired sequentially. We consider the price of anarchy of this game, which is the ratio between the total coverage of the APs in the worst Nash equilibrium of the game and what the total coverage of the APs would be if the channel assignment were done optimally by a central authority. We provide bounds on the price of anarchy depending on assumptions on the underlying network and the type of bargaining allowed between service providers. The key tool in the analysis is the identification of the Nash equilibria with the solutions to a maximal coloring problem in an appropriate graph. We relate the price of anarchy of these games to the approximation factor of local optimization algorithms for the maximum k-colorable subgraph problem. We also study the speed of convergence in these games.     

Pure alexia and word-meaning deafness in a patient with multiple sclerosis

OBJECTIVE: To describe pure alexia and auditory comprehension problems in a young woman with multiple sclerosis (MS). PATIENT: A 33-year-old woman with MS who complained of difficulties in reading and comprehending spoken language was referred for a neuropsychological examination. Reading difficulties were confirmed and most of the reading errors were additions, omissions, and substitutions of single letters. While the patient reported that the letters seemed to disappear before her eyes, no general problems with visual attention, visual discrimination, or scanning were detected. No difficulties with spelling were reported. The auditory comprehension deficit is interpreted as a form of a semantic access disorder and is not due to generalized slowing in information processing or conceptual disintegration. CONCLUSIONS: Pure alexia is unusual in MS and to our knowledge only 1 other case has been reported (in Japanese). Memory impairments and slowed information processing are probably the most frequent cognitive sequelae of the disease and, consequently, the literature is biased toward the study of those cognitive domains. However, given the wide distribution of sclerotic plaques in MS, it could be argued that we should expect some variability of cognitive changes in MS. Striking deficits as seen in this patient should make us more sensitive to this possibility.     

Effect of various marine lipids on transdermal drug delivery--in vitro evaluation

The effects of several marine lipids on the penetration of hydrocortisone and nitroglycerin through excised hairless mouse skin have been studied. Fatty acid extracts obtained by hydrolysis of Portuguese dog-fish-liver-oil or by hydrolysis of cod-liver-oil were shown to be effective skin penetration enhancers. Phospholipid obtained from squid was also shown to be effective enhancer. However, the enhancing effect of the marine products could generally be associated with their content of free unsaturated fatty acids. The fatty acid extract obtained from cod-liver-oil caused insignificant skin irritation when incorporated into an ointment base and applied to human skin.     

Nearly optimal bounds for distributed wireless scheduling in the SINR model

We study the wireless scheduling problem in the SINR model. More specifically, given a set of \(n\) links, each a sender–receiver pair, we wish to partition (or schedule) the links into the minimum number of slots, each satisfying interference constraints allowing simultaneous transmission. In the basic problem, all senders transmit with the same uniform power. We analyze a randomized distributed scheduling algorithm proposed by Kesselheim and Vöcking, and show that it achieves \(O(\log n)\)-approximation, an improvement of a logarithmic factor. This matches the best ratio known for centralized algorithms and holds in arbitrary metric space and for every length-monotone and sublinear power assignment. We also show that every distributed algorithm uses \(\varOmega (\log n)\) slots to schedule certain instances that require only two slots, which implies that the best possible absolute performance guarantee is logarithmic.     

Structural studies on Li1−xTa3O8−xFx solid solutions by full-profile refinements of guinier-hägg X-ray film data

Three compositions of solid solutions Li_1?xTa₃O_8?xF_x (x = 0) 0.25 and 0.5) have been studied by full-profile analysis of microdensitometer-measured Guinier-Hägg powder diffraction data. From sequential changes in disorder and interatomic distances within this series, the preferred lithium position is proposed to be approximately (0.11, $\text{1}\text{2}$, 0.03). The results obtained for the high temperature modification of LiTa₃O₈ by A.G. Nord and J.O. Thomas from X-ray single crystal and neutron powder diffraction data [Nord, A.G. and Thomas, J.O., Acta Chem. Scand., in press] are discussed in view of the present investigation.     

Structure and ionic conductivity of Lix(MeO)4Me30O90 (Me = NbV, WVI)

The compound Li_x(MeO)₄Me₃₀O₉₀ (Me = Nb^V, W^VI) has been prepared and investigated by X-ray powder and electron diffraction techniques as well as by measurements of complex impedance. The structure is of the tripled tetragonal tungsten bronze type with one third of the pentagonal tunnels filled with equal numbers of Me and 0 atoms. The conductivity at 300°C is 0.20 and 0.25 ohm^?1m^?1 for sintered samples with x equal to 2 and 4, respectively. The location and probable routes of transport of the lithium ions are discussed.