Multi-frequency and multi-polarization Signals of Opportunity Reflectometry (SoOp-R) has the potential to monitor global root-zone soil moisture (RZSM) at a high spatiotemporal resolution using a constellation of small satellites. This paper introduces a mission design tool to assist the high-level design of multi-frequency SoOp-R missions and presents an example of tradespace analysis. Hundreds of constellation design alternatives were enumerated and compared using standard scoring functions that aggregate cost and coverage metrics for the multi-frequency SoOp-R mission. This comprehensive tool will be valuable in the development of future science missions using SoOp-R.

Signals of opportunity reflectometry (SoOp-R), the re-utilization of non-cooperative satellite transmissions for communication and navigation, is a promising approach to remote sensing of root-zone soil moisture (RZSM). Satellite transmissions in the frequency ranges of 137–138, 240–270, and 360–380 MHz are of interest due to the increased penetration depth. These can be combined with Global Navigation Satellite System Reflectometry (GNSS-R) in L-band (1575.42 MHz) to estimate the subsurface SM profile. The objective is to define requirements (e.g. frequency and polarization combinations, observation error, and temporal coincidence of multi-source observations) for satellite-based remote sensing of RZSM. Our approach is to use synthetic observations generated from multi-year time series ofin-situSM measurements from seven United States Climate Reference Network (USCRN) sites and dynamic vegetation structure based on a simple scaling method. A multi-frequency/polarimetric retrieval algorithm is developed and applied to these synthetic observations and used to predict retrieval errors for a range of changes in system parameters. We found that the use of both high and low frequencies improves retrieval accuracy by limiting uncertainties from vegetation and surface SM and providing sensitivity to deeper layers. Moreover, the retrieval errors were found to increase linearly with the reflectivity error and inter-frequency time delays. A bivariate model derived from this linear relationship will be useful for developing requirements on reflectivity precision based upon science requirements for SM/VWC retrievals. Although orbits of specific transmitter constellations were used to generate realistic distributions of incidence angle combinations, the method and results could be applied more generally.

In this paper, we propose a transformer-based image matting model called MatteFormer, which takes full advantage of trimap information in the transformer block. Our method first introduces a prior-token which is a global representation of each trimap region (e.g. foreground, background and unknown). These prior-tokens are used as global priors and participate in the self-attention mechanism of each block. Each stage of the encoder is composed of PAST (Prior-Attentive Swin Transformer) block, which is based on the Swin Transformer block, but differs in a couple of aspects: 1) It has PA-WSA (Prior-Attentive Window Self-Attention) layer, performing self-attention not only with spatial-tokens but also with prior-tokens. 2) It has prior-memory which saves prior-tokens accumulatively from the previous blocks and transfers them to the next block. We evaluate our MatteFormer on the commonly used image matting datasets: Composition-Ik and Distinctions-646. Experiment results show that our proposed method achieves state-of-the-art performance with a large margin. Our codes are available at https://github.com/webtoon/matteformer.

This poster presents system architecture, software stack, and performance analysis for SK hynix’s very first GDDR6-based processing-in-memory (PIM) product sample, called Accelerator-in-Memory (AiM).AiM is designed for the in-memory acceleration of matrix-vector product operations, which are commonly found in machine learning applications. The strength of AiM primarily comes from the two design factors, which are 1) all-bank operation support and 2) extended DRAM command set. All-bank operations allow AiM to fully utilize the abundant internal DRAM bandwidth, which makes it an attractive solution for memory-bound applications. The extended command set allows the host to address these new operations efficiently and provides a clean separation of concerns between the AiM architecture and its software stack design.We present a dedicated FPGA-based reference platform with a software stack, which is used to validate AiM design and evaluate its system-level performance. We also demonstrate FMC-based AiM extension cards that are compatible with the off-the-shelf FPGA boards and serve as an open research platform allowing potential collaborators and academic institutes to access our hardware and software systems.

SigNals Of Opportunity: P-band Investigation (SNOOPI) will be the first in-space validation of P-band (240–380 MHz) SoOp techniques and a prototype science instrument. These techniques have the potential to enable remote sensing of root-zone soil moisture (RZSM) and snow water equivalent (SWE). SNOOPI technology validation goals will be met by targeting observations within 9 km of the SMAP calibration/validation sites in the continental United States. A second priority is collection of continuous phase data over snow-covered regions. These goals are evaluated under constraints of a limited data budget and mission lifetime, with a launch readiness in August 2022. This presentation will review the instrument science plans aimed at achieving the validation objectives defined for the mission. Mission planning and data processing approaches are described.

Root-zone soil moisture (RZSM) is one of the least measured hydrological variables despite its critical role in understanding the global carbon cycle and forecasting agricultural drought and food production. Signals of opportunity (SoOp) has great potential to overcome the limitation of conventional microwave methods to utilize lower frequencies in space-borne remote sensing. Multi-frequency SoOp reflectometry (SoOp-R) offers a promising solution to measure RZSM by solving the inverse problem for obtaining multi-layer soil moisture profiles. This paper summarizes ongoing work to develop and validate multi-frequency SoOp- R to retrieve RZSM, including forward and inverse methods, sensitivity analysis for the optimal frequency combination, and tower-based field experiments.

SigNals Of Opportunity: P-band Investigation (SNOOPI) will be the first on-orbit demonstration of remote sensing using Signals of Opportunity (SoOp) in P-band (240–380 MHz). P-band is needed to penetrate through dense vegetation and into the root zone. The longer wavelength of P-band also increases the unwrapping interval for phase observations. These observations hold the potential for spaceborne remote sensing of root-zone soil moisture (RZSM) and snow water equivalent (SWE), two variables identified as priorities in the 2017–2027 Decadal Survey for Earth Science and Applications from Space. SNOOPI will provide in-space validation of both the P-band SoOp technique and a science instrument prototype. SNOOPI technology validation goals will be met by targeting observations within 9 km of the SMAP calibration/validation sites in the continental United States. A secondary priority is collection of continuous phase data over snow-covered regions. These goals are evaluated under constraints of a limited data budget and mission lifetime, with a launch readiness in early 2022. Updates on the development of measurement models and mission planning to support SNOOPI are provided. A ground-based station will be deployed to monitor the noncooperative sources, in order to reduce risk due to uncertainty in knowledge of the broadcast power, spectrum shape, and orbital position.

Root zone soil moisture (RZSM) is a key environmental variable needed for drought and flood forecasts as well as fundamental understanding of the water cycle. Signals of opportunity (SoOp) reflectometry offers a new possibility to directly measure RZSM using multiple microwave frequencies allocated for communications and navigation. This paper develops a multi-frequency SoOp RZSM retrieval algorithm and presents simulated results. We adopted the Principle of Maximum Entropy (POME) model to construct soil moisture profiles from in-situ observations at discrete depths. Synthetic observations, representing SoOp reflectivities at four (4) frequencies spanning I, P and L-bands were generated for each profile. A simulated annealing method was then used to generate soil moisture profile retrievals from these simulated reflectivities. An error assessment was conducted for these simulated retrievals.

SigNals of Opportunity: P-band Investigation (SNOOPI) will be an in-space technology demonstration of reflectometry using 240-380 MHz communications transmissions. SNOOPI will both demonstrate essential techniques for root-zone soil moisture (RZSM) and snow water equivalent (SWE) remote sensing as well as provide in-space validation of prototype instrument technology. This paper presents results from studies conducted to define key parameters of the SNOOPI mission, including orbital coverage, signal processing, and the estimated power from the non-cooperative sources.

Signals of opportunity (SoOp) is a promising technique to measure key geophysical variables at high spatiotemporal scales using multiple microwave frequencies outside bands allocated for science. To better understand spaceborne SoOp observations, an end-to-end mission simulator has been developed. It determines observation geometries of multiple transmitters and receivers and generates SoOp observables for land remote sensing by integrating geophysical data and instrument properties. Using the simulator, the coverage and the instrument data of a hypothetical constellation of spaceborne SoOp receivers were evaluated over different orbital planes. This integrative tool facilitates the spaceborne SoOp mission design by exploring and optimizing the tradespace of a variety of design parameters.

Advances in machine learning (ML) have ignited hardware innovations for efficient execution of the ML models many of which are memory-bound (e.g., long short-term memories, multi-level perceptrons, and recurrent neural networks). Specifically, inference using these ML models with small batches, as would be the case at the Cloud edge, has little reuse of the large filters and is deeply memory-bound. Simultaneously, processing-in or -near memory (PIM or PNM) is promising unprecedented high-bandwidth connection between compute and memory. Fortunately, the memory-bound ML models are a good fit for PIM. We focus on digital PIM which provides higher bandwidth than PNM and does not incur the reliability issues of analog PIM. Previous PIM and PNM approaches advocate full processor cores which do not conform to PIM's severe area and power constraints. We describe Newton, a major DRAM maker's upcoming accelerator-in-memory (AiM) product for machine learning, which makes the following contributions: (1) To satisfy PIM's area constraints, Newton (a) places a minimal compute of only multiply-accumulate units and buffers in the DRAM which avoids the full-core area and power overheads of previous work and thus makes PIM feasible for the first time, and (b) employs a DRAM-like interface for the host to issue commands to the PIM compute. The PIM compute is rate-matched to the internal DRAM bandwidth and employs a non-intuitive, global input vector buffer shared by the entire channel to capture input reuse while amortizing buffer area cost. To the host, Newton's interface is indistinguishable from regular DRAM without any offloading overheads and PIM/non-PIM mode switching, and with the same deterministic latencies even for floating-point commands. (2) To prevent the PIM-host interface from becoming a bottleneck, we include three optimizations: commands which gang multiple compute operations both within a bank and across banks; complex, multi-step compute commands - both of which save critical command bandwidth; and targeted reduction of tFAWoverhead. (3) To capture output vector reuse with reasonable buffering, Newton employs an unusually-wide interleaved layout for the matrix. Our simulations running state-of-the-art neural networks show that building on a realistic HBM2E-like DRAM, Newton achieves 10x and 54x average speedup over a non-PIM system with infinite compute that perfectly uses the external DRAM bandwidth and a realistic GPU, respectively.

Inductive Power Transfer (TPT) systems have been proposed in recent years as a safe, convenient and robust method of battery charging in EVs and robots. In an IPT system, the magnetic design of the primary and secondary pads is an important factor that determines the power transfer capability. This paper proposes the Tripolar Pad (TPP), a novel three coil magnetic pad utilising mutually decoupled coils. The TPP maintains tolerance to rotational displacements similar to nonpolarised pads, but uses polarised fields to improve power transfer. Simulation results are first shown in the paper and then validated by a scaled prototype.