Peraso Technologies (Canada)

Benthic mapping relies on a range of methods to ground truth remotely sensed data to map physical and biotic features. Traditional methods are labor-intensive and time-consuming with limited scalability. Autonomous underwater vehicles (AUVs) and remote operated vehicles (ROVs) can collect high volumes of quality data but require automated processing routines and pose other logistic challenges. This paper explores the automated use of multiple machine learning (ML) models to detect and delineate benthic habitat components from structure-from-motion (SfM) orthomosaics. Overlapping AUV photos were processed into high-resolution, seamless orthomosaics for a portion of the Lake Michigan lakebed. ML models were trained to delineate rocks and live mussels and subsequently applied to the SfM orthomosaics. Results demonstrate that a robust pipeline with SfM and ML models is an effective approach to efficiently, accurately, and comprehensively map benthic habitats.

In this work, different solutions are discussed to integrate 60GHz phased array antennas with active devices on the same board. Capabilities of both ceramic- and organic-based manufacturing processes are explored to realize highperformance phased array antennas. Various phased array antennas, suitable for backhaul and consumer electronics, are designed and implemented based on different solutions.

A 1×4 phased array antenna in package is presented at 60 GHz frequency band. End-fire elements are double side dipoles which provide a near omni-directional beam in H-plane. In E-plane, the phased array facilitates ±45° beam steering which improves the spatial coverage. The achieved end-fire gain is 9.1 dBi and both impedance and gain bandwidths cover all WiGig channels. The wide bandwidth, high coverage, and high gain of the phased array make it a desirable candidate for multi-gigabit wireless communication.

We present a new method to phase-modulate RF pulses through injection locking. The modulator can produce BPSK or QPSK modulated pulses over a frequency span of 3 to 10 GHz. The RF pulses are produced using a fast on/off switching technique to power up and power down a tunable ring oscillator. Experiments show that the modulator's energy consumption ranges from 13 pJ/pulse to 18 pJ/pulse as the carrier frequency varies from 3 GHz to 10 GHz. The energy consumption figures are for 2 ns long pulses with amplitudes of 300 mV at 3 GHz and 200 mV at 10 GHz. The pulse repetition rate is 250 MHz. The chip was fabricated in 130 nm CMOS and the core circuit area is only 0.05 mm 2 .

This paper presents a compact antenna-in-package designed for 60 GHz short-range communication. The antenna is an array of 4 microstrip patches, with a size of 5 mm × 12 mm. The maximum measured gain of the antenna is 12 dBi. The measured bandwidth of the antenna is 6.5 GHz which covers 3 channels of 60 GHz spectrum. The 60 GHz chip is flip-chipped to the top of the package. The 13 mm × 13 mm package consists of 6 layers. The total thickness of the package is 650 µm without die, and 820–870 µm with die. Measured and simulated results for the antenna are presented in this paper.

Abstract Chemical flooding methods hold particular attraction for recovering the "residual oil" left in the reservoir after waterflooding. This paper describes and compares the results for two promising methods, viz. micellar flooding and alkaline-surfactant-polymer (ASP) flooding processes. Both of these methods have been tested successfully in the field, notably micellar flooding. Laboratory results are described for micellar floods in consolidated sandstone cores as well as in unconsolidated sand packs, including a three-dimensional model, equipped with horizontal or vertical wells. Floods were also carried out in unconsolidated cores using combinations of an alkali, surfactant and a polymer. Individual slugs were injected sequentially in some of the experiments, while the three components were mixed and injected as a single slug in other experiments. Oil recoveries in the two cases were similar. Results for the two processes are compared and contrasted, showing that, on the basis of oil volume recovered per unit mass of the chemical used, the two processes are similar, with micellar flooding having an edge. However, on the basis of total oil recovery, micellar flooding is the superior process, with oil recoveries ranging from 50 to 80﹪ of the oil left in the porous medium after a waterflood. Practical implications of the results are discussed. Introduction Among chemical flooding methods, micellar flooding and alkaline- surfactant-polymer (ASP) flooding processes are particularly effective for recovering a large fraction of the conventional oil (25 °CDATA[API, or higher) left in the reservoir after a waterflood, which could be as much as 60﹪ of the original oil in place. Many field tests of the micellar flooding process and several of ASP have established the effectiveness of these methods for mobilizing waterflood residual oil. The present laboratory study compares and contrasts the two processes, based on tertiary floods in sand packs and Berea sandstone cores. A number of investigators have noted the use of an alkali for reducing the divalent ion content and increasing the negative charge of the rock with a view to reducing chemical loss(1,2). Surkalo(3) reported the alkaline-surfactant-polymer (ASP) process as an alternative to micellar flooding. Several field tests have also been reported. Another function of alkali, if the Acid No. of the crude oil is large enough (> 0.5 mg KOH/g crude oil) is that the alkali may react with the acid components to form a surfactant in situ. Other factors, such as gravity segregation of alkali solutions, rate of diffusional and mechanical mixing, and subsequent mixing of surfactant formed would limit the effectiveness of this mechanism. The Process Chemical flooding methods are based on improving the mobility ratio, i.e., making the mobility of the displacing flood less than or equal to the mobility of the displaced fluid, and increasing the capillary number, mainly by making the interfacial tension (IFT) between the displacing and the displaced phases small, usually by about 1,000 fold. Other effects are also present, such as formation of macro- and microemulsions, formation of precipitates, wettability changes, relative permeability shifts, etc. Macroemulsions may improve the mobility ratio through drop entrainment and entrapment. At the same time, surfactant adsorption occurs on the rock surface.