No abstract

Digital-to-time converters (DTC) generate a clock with a time delay (or phase shift) based on a digital input code. They can be used in clock-and-data-recovery (CDR) circuits [1,2], in the feedback or reference path of a phase-locked loop (PLL) [3,4] or as direct phase modulators in outphasing transmitters (OT) [5]. While DTCs in PLLs often operate close to the reference oscillator frequency, CDR and OT DTCs are required to operate at frequencies in the GHz range. DTCs are often built using a multistage segmented architecture, employing separate coarse and fine delay tuning.Coarse tuning is usually implemented using multiphase generators based on delay-locked loops (DLLs) [5] or dividers [1], followed by a multiplexer (MUX). Compared to DLLs, dividers do not require a control loop and can achieve lower jitter, but their input clock frequency has to be a multiple of the output frequency. The fine tuning is commonly implemented either with delay cells [4,5], which tune the RC constant of a node to modify the signal zero crossing time, or with phase interpolators (PI) [1]. While delay cells offer high linearity, they cause unwanted supply modulation through a code-dependent current consumption, high jitter through the degradation of the (dis)charge slopes, and do not provide a well-defined delay range. Replica paths with inverted codes are necessary to equalize the current consumption over code [4] and calibration engines are used to cope with the undefined range [5]. PIs on the other hand can provide a full and exact 2π coverage, but have intrinsically high nonlinearity [1,2,6], caused mainly by (a) the ratio Δt/τ int of temporal spacing of the two input signals Δt and the time constant at the interpolation node τ int [2,6] and (b) contention between different interpolation cells (INTC), which leads to short-circuit (cross) currents between V DD and V SS during the interpolation [6]. Harmonic rejection filtering has been introduced to improve the PI linearity [1], but at expense of a lower slew rate of the internal signals and hence jitter.To achieve a wide range (500ps) with fine resolution (244fs), a full and exact 2π coverage, low jitter and high linearity, this paper presents a DTC incorporating: (a) a multi-modulus-divider (MMD) coarse-tuning stage for 2π coverage, (b) re-sampling of the coarse stage by rising and falling edges of the input clock for low-jitter, (c) a PI-based fine-tuning stage for correct-by-construction coverage, (d) embedding additional logic inside the INTCs to break the contention cross-currents for high linearity. The current consumption of the DTC is equal over all codes, resulting in lower sensitivity to the power supply impedance and avoiding replica paths. Figure 2.9.1 shows the DTC architecture. It is segmented into 3 stages, that produce successively finer time resolution: in an ultra-coarse tuning stage, an MMD with a 3b resolution produces 2 output signals at 2GHz spaced by Δt uc = T out /8 = 62.5ps, a coarse-tuning stage reduces this spacing with a 1b resolution to Δt c = 31.25p...

Understanding the factors that determine the realized and potential distribution of a species requires knowledge of abiotic, physiological, limitations as well as ecological interactions. Fungi of the order Laboulbeniales specialize on arthropods and are typically thought to be highly specialized on a single species or closely related group of species. Because infections are almost exclusively transmitted through direct contact between the hosts, the host ecology, to a large extent, determines the distribution and occurrence of the fungus. We examined ~20,000 fruit flies (Diptera: Dacinae) collected in Malaysia, Sulawesi, Australia, and the Solomon Islands between 2017 and 2019 for fungal infections and found 197 infected flies across eight different Bactrocera species. Morphology and 1,363 bps of small subunit (18S) DNA sequences both support that the infections are from a single polyphagous fungal species Stigmatomyces dacinus—a known ectoparasite of these fruit flies. This leads to the question: why is S. dacinus rare, when its hosts are widespread and abundant? In addition, the hosts are all Bactrocera, a genus with ~480 species, but 37 Bactrocera species found sympatric with the hosts were never infected. Host‐selection does not appear to be phylogenetically correlated. These results suggest a hidden complexity in how different, but closely related, host species vary in their susceptibility, which somehow limits the abundance and dispersal capability of the fungus.