Neural networks have been proposed recently for positioning and channel charting of user equipments (UEs) in wireless systems. Both of these approaches process channel state information (CSI) that is acquired at a multi-antenna basestation in order to learn a function that maps CSI to location information. CSI-based positioning using deep neural networks requires a dataset that contains both CSI and associated location information. Channel charting (CC) only requires CSI information to extract relative position information. Since CC builds on dimensionality reduction, it can be implemented using autoencoders. In this paper, we propose a unified architecture based on Siamese networks that can be used for supervised UE positioning and unsupervised channel charting. In addition, our framework enables semisupervised positioning, where only a small set of location information is available during training. We use simulations to demonstrate that Siamese networks achieve similar or better performance than existing positioning and CC approaches with a single, unified neural network architecture.

Compared to DNA sequence analysis, matrix-assisted laser desorption ionization-time of flight mass spectrometry (MALDI-TOF MS) correctly identified 100% of Cryptococcus species, distinguishing the notable pathogens Cryptococcus neoformans and C. gattii. Identification was greatly enhanced by supplementing a commercial spectral library with additional entries to account for subspecies variability.Cryptococcosis is typically a severe infection of the central nervous system characterized by meningoencephalitis and other neurological complications, primarily in patients with AIDS or other forms of immune compromise (26). Cryptococcus isolates must be distinguished and characterized at the species or subspecies level due to differences in epidemiology, virulence, and antifungal drug susceptibility. Cryptococcus gattii has a greater propensity to infect immunocompetent people (26), with some strains being more virulent (3,16,27) or less susceptible to fluconazole and other triazoles (6,20,42). Cryptococcus neoformans consists of two varieties, var. grubii and var. neoformans. While C. neoformans var. grubii causes the majority of clinical infections worldwide, C. neoformans var. neoformans infections are more prevalent in India (18) and certain areas of Europe (11,24) and are strongly correlated with infections in the elderly, the presence of skin lesions, and corticosteroid use (12). Infections due to other Cryptococcus species are rare, but the incidence is increasing, requiring greater vigilance by clinical laboratories (21).Currently, identification of Cryptococcus to the genus or subgenus level from clinical specimens relies upon the microscopic examination of yeast cells in conjunction with biochemical tests, differential media, and/or DNA sequence analysis (9,22,23,31,33,(44)(45)(46). These tests may require multiday incubation or labor-intensive and costly protocols that may delay diagnosis.Recently, matrix-assisted laser desorption ionization-time of flight mass spectrometry (MALDI-TOF MS) has been used to successfully identify various species of bacteria and fungi (17, 28, 34) By identifying species based on characteristic protein spectra extracted from whole cells, MALDI-TOF MS produces a highly discriminatory identification of a pure culture in 5 to 20 min at minimal cost, depending on the sample preparation procedure utilized. In this study, we tested the ability of wholecell MALDI-TOF MS to identify clinical isolates of Cryptococcus spp. and to further differentiate them at the species or subspecies level.We analyzed a total of 160 yeast isolates, including 137 Cryptococcus strains and 23 non-Cryptococcus yeast strains. Twenty-five were type and reference strains (Table 1) As a reference "gold standard" for this evaluation, all isolates were identified by DNA sequence analysis of the rRNA internal transcribed spacer (ITS) region (7), which differentiated all yeast species except for C. neoformans and C. gattii. Clinical isolates demonstrated Ͼ99% similarity to type strains, with a between-species diverge...

Channel state information (CSI)-based fingerprinting via neural networks (NNs) is a promising approach to enable accurate indoor and outdoor positioning of user equipments (UEs), even under challenging propagation conditions. In this paper, we propose a positioning pipeline for wireless LAN MIMO-OFDM systems which uses uplink CSI measurements obtained from one or more unsynchronized access points (APs). For each AP receiver, novel features are first extracted from the CSI that are robust to system impairments arising in real-world transceivers. These features are the inputs to a NN that extracts a probability map indicating the likelihood of a UE being at a given grid point. The NN output is then fused across multiple APs to provide a final position estimate. We provide experimental results with realworld indoor measurements under line-of-sight (LoS) and non-LoS propagation conditions for an 80 MHz bandwidth IEEE 802.11ac system using a two-antenna transmit UE and two AP receivers each with four antennas. Our approach is shown to achieve centimeter-level median distance error, an order of magnitude improvement over a conventional baseline.

No abstract