arXiv subjects

Upamanyu Madhow

Publications and source records attributed to Upamanyu Madhow.

At least 19 recordsLinked to original sources

arXiv2026

Tiled Beamspace MVDR for 1024-element Wideband Radar

We present a tiled architecture for computationally efficient digital beamforming for wideband massive MIMO radar, using beamspace dimension reduction for each tile, and coordinated training of reduced-dimension MVDR beamformers across tiles. We illustrate the efficacy of our approach for a setting in which a 1024-element airborne radar platform beamforms towards airborne targets while suppressing strong interference from ground transmitters. The array is organized into eight 128-element tiles, each a 2D array with 4 (vertical) x 32 (horizontal) elements. Each tile applies a 2D spatial DFT to achieve energy concentration in beamspace, and a 1D temporal FFT to channelize the wideband signal into subbands for which narrowband array models apply. A small tile-level beamspace window is selected for each target (depending on its angle of arrival) in each subband, and coordinated training across tiles is used to compute reduced-dimension MVDR beamformers per-target, per-subband. While full-dimensional MVDR processing is infeasible for the system under consideration, we show that our proposed approach significantly outperforms beamspace MVDR beamforming for a single 128-element tile, where we set the dimensions of the spatial filter (and hence the complexity of MVDR training) to be equal in both systems.

eess.SP↗

arXiv2025

Beamspace Dimensionality Reduction for Massive MU-MIMO: Geometric Insights and Information-Theoretic Limits

Beamspace dimensionality reduction, a classical tool in array processing, has been shown in recent work to significantly reduce computational complexity and training overhead for adaptive reception in massive multiuser (MU) MIMO. For sparse multipath propagation and uniformly spaced antenna arrays, beamspace transformation, or application of a spatial FFT, concentrates the energy of each user into a small number of spatial frequency bins. Empirical evaluations demonstrate the efficacy of linear Minimum Mean Squared Error (LMMSE) detection performed in parallel using a beamspace window of small, fixed size for each user, even as the number of antennas and users scale up, while being robust to moderate variations in the relative powers of the users. In this paper, we develop a fundamental geometric understanding of this ``unreasonable effectiveness'' in a regime in which zero-forcing solutions do not exist. For simplified channel models, we show that, when we enforce a suitable separation in spatial frequency between users, the interference power falling into a desired user's beamspace window of size $W$ concentrates into a number of dominant eigenmodes smaller than $W$, with the desired user having relatively small projection onto these modes. Thus, linear suppression of dominant interference modes can be accomplished with small noise enhancement. We show that similar observations apply for MIMO-OFDM over wideband multipath channels synthesized from measured 28 GHz data. We propose, and evaluate via information-theoretic benchmarks, per-subcarrier reduced dimension beamspace LMMSE in this setting.

eess.SP↗

arXiv2025

Scaling Wideband Hybrid Beamforming for sub-THz Communication

We investigate the capacity attainable for a multiuser MIMO uplink as we scale both array size and bandwidth for regimes in which all-digital arrays incur excessive hardware complexity and power consumption. We consider a tiled hybrid beamforming architecture in which each tile, or subarray, is a phased array performing analog (or RF) beamforming, followed by DSP on the tile outputs. For parameters compatible with sub-THz fixed access links, we discuss hardware and power consumption considerations for choosing tile size and the number of tiles. Noting that the problem of optimal multiuser MIMO in our wideband regime is open even for the simplest possible channel models, we compare the spectral efficiencies attainable by a number of reasonable strategies for tile-level RF beamforming, assuming flexibility in the digital signal processing (DSP) of the tile outputs. We consider a number of beam broadening approaches for addressing the ``beam squint'' incurred by RF beamforming in our wideband regime, along with strategies for sharing tiles among users. Information-theoretic benchmarks are computed for an idealized MIMO-OFDM system, with linear per-subcarrier multiuser detection compared against an unconstrained complexity receiver.

eess.SP↗

arXiv2025

Scaling Wideband Massive MIMO Radar via Beamspace Dimension Reduction

We present an architecture for scaling digital beamforming for wideband massive MIMO radar. Conventional spatial processing becomes computationally prohibitive as array size grows; for example, the computational complexity of MVDR beamforming scales as O(N^3) for an N-element array. In this paper, we show that energy concentration in beamspace provides the basis for drastic complexity reduction, with array scaling governed by the O(NlogN) complexity of the spatial FFT used for beamspace transformation. Specifically, we propose an architecture for windowed beamspace MVDR beamforming, parallelized across targets and subbands, and evaluate its efficacy for beamforming and interference suppression for government-supplied wideband radar data from the DARPA SOAP (Scalable On-Array Processing) program. We demonstrate that our approach achieves detection performance comparable to full-dimensional benchmarks while significantly reducing computational and training overhead, and provide insight into tradeoffs between beamspace window size and FFT resolution in balancing complexity, detection accuracy, and interference suppression.

eess.SP↗

arXiv2025

mmSnap: Bayesian One-Shot Fusion in a Self-Calibrated mmWave Radar Network

We present mmSnap, a collaborative RF sensing framework using multiple radar nodes, and demonstrate its feasibility and efficacy using commercially available mmWave MIMO radars. Collaborative fusion requires network calibration, or estimates of the relative poses (positions and orientations) of the sensors. We experimentally validate a self-calibration algorithm developed in our prior work, which estimates relative poses in closed form by least squares matching of target tracks within the common field of view (FoV). We then develop and demonstrate a Bayesian framework for one-shot fusion of measurements from multiple calibrated nodes, which yields instantaneous estimates of position and velocity vectors that match smoothed estimates from multi-frame tracking. Our experiments, conducted outdoors with two radar nodes tracking a moving human target, validate the core assumptions required to develop a broader set of capabilities for networked sensing with opportunistically deployed nodes.

eess.SP↗

arXiv2025

Crowd Size Estimation for Non-Uniform Spatial Distributions with mmWave Radar

In this paper, we present a novel methodology for crowd size estimation using monostatic mmWave radar. Our aim is to accurately count large crowds that follow a non-uniform spatial distribution. Our estimation approach relies on the rigorous derivation of occlusion probabilities, which are then used to mathematically characterize the probability distributions that describe the number of agents visible to the radar as a function of the crowd size. We then estimate the true crowd size by comparing these derived mathematical models to the empirical distribution of the number of visible agents detected by the radar. This method requires minimal sensing capabilities (e.g., angle-of-arrival information is not needed), thus being well suited for either a dedicated mmWave radar or an integrated sensing and communication (ISAC) system. Extensive numerical simulations validate our methodology, demonstrating strong performance across diverse spatial distributions and for crowd sizes of up to (and including) 30 agents. We achieve a mean absolute error (MAE) of 0.48 agents, significantly outperforming a baseline which assumes that the agents are uniformly distributed in the area. Overall, our approach holds significant promise for a variety of applications including network resource allocation, crowd management, and urban planning.

eess.SP↗

arXiv2024

Multi-sensor Spatial Association using Joint Range-Doppler Features

We investigate the problem of localizing multiple targets using a single set of measurements from a network of radar sensors. Such "single snapshot imaging" provides timely situational awareness, but can utilize neither platform motion, as in synthetic aperture radar, nor track targets across time, as in Kalman filtering and its variants. Associating measurements with targets becomes a fundamental bottleneck in this setting. In this paper, we present a computationally efficient method to extract 2D position and velocity of multiple targets using a linear array of FMCW radar sensors by identifying and exploiting inherent geometric features to drastically reduce the complexity of spatial association. The proposed framework is robust to detection anomalies, and achieves order of magnitude lower complexity compared to conventional methods. While our approach is compatible with conventional FFT-based range-Doppler processing, we show that more sophisticated techniques for range-Doppler estimation lead to reduced data association complexity as well as higher accuracy estimates of target positions and velocities.

eess.SP↗

arXiv2024

Long-Term Fairness in Sequential Multi-Agent Selection with Positive Reinforcement

While much of the rapidly growing literature on fair decision-making focuses on metrics for one-shot decisions, recent work has raised the intriguing possibility of designing sequential decision-making to positively impact long-term social fairness. In selection processes such as college admissions or hiring, biasing slightly towards applicants from under-represented groups is hypothesized to provide positive feedback that increases the pool of under-represented applicants in future selection rounds, thus enhancing fairness in the long term. In this paper, we examine this hypothesis and its consequences in a setting in which multiple agents are selecting from a common pool of applicants. We propose the Multi-agent Fair-Greedy policy, that balances greedy score maximization and fairness. Under this policy, we prove that the resource pool and the admissions converge to a long-term fairness target set by the agents when the score distributions across the groups in the population are identical. We provide empirical evidence of existence of equilibria under non-identical score distributions through synthetic and adapted real-world datasets. We then sound a cautionary note for more complex applicant pool evolution models, under which uncoordinated behavior by the agents can cause negative reinforcement, leading to a reduction in the fraction of under-represented applicants. Our results indicate that, while positive reinforcement is a promising mechanism for long-term fairness, policies must be designed carefully to be robust to variations in the evolution model, with a number of open issues that remain to be explored by algorithm designers, social scientists, and policymakers.

stat.ML↗

arXiv2024

Improving Robustness via Tilted Exponential Layer: A Communication-Theoretic Perspective

State-of-the-art techniques for enhancing robustness of deep networks mostly rely on empirical risk minimization with suitable data augmentation. In this paper, we propose a complementary approach motivated by communication theory, aimed at enhancing the signal-to-noise ratio at the output of a neural network layer via neural competition during learning and inference. In addition to standard empirical risk minimization, neurons compete to sparsely represent layer inputs by maximization of a tilted exponential (TEXP) objective function for the layer. TEXP learning can be interpreted as maximum likelihood estimation of matched filters under a Gaussian model for data noise. Inference in a TEXP layer is accomplished by replacing batch norm by a tilted softmax, which can be interpreted as computation of posterior probabilities for the competing signaling hypotheses represented by each neuron. After providing insights via simplified models, we show, by experimentation on standard image datasets, that TEXP learning and inference enhances robustness against noise and other common corruptions, without requiring data augmentation. Further cumulative gains in robustness against this array of distortions can be obtained by appropriately combining TEXP with data augmentation techniques. The code for all our experiments is available at https://github.com/bhagyapuranik/texp_for_robustness.

cs.LG↗

arXiv2022

Neuro-Inspired Deep Neural Networks with Sparse, Strong Activations

While end-to-end training of Deep Neural Networks (DNNs) yields state of the art performance in an increasing array of applications, it does not provide insight into, or control over, the features being extracted. We report here on a promising neuro-inspired approach to DNNs with sparser and stronger activations. We use standard stochastic gradient training, supplementing the end-to-end discriminative cost function with layer-wise costs promoting Hebbian ("fire together," "wire together") updates for highly active neurons, and anti-Hebbian updates for the remaining neurons. Instead of batch norm, we use divisive normalization of activations (suppressing weak outputs using strong outputs), along with implicit $\ell_2$ normalization of neuronal weights. Experiments with standard image classification tasks on CIFAR-10 demonstrate that, relative to baseline end-to-end trained architectures, our proposed architecture (a) leads to sparser activations (with only a slight compromise on accuracy), (b) exhibits more robustness to noise (without being trained on noisy data), (c) exhibits more robustness to adversarial perturbations (without adversarial training).

cs.NE↗

arXiv2022

Self-supervised Speaker Recognition Training Using Human-Machine Dialogues

Speaker recognition, recognizing speaker identities based on voice alone, enables important downstream applications, such as personalization and authentication. Learning speaker representations, in the context of supervised learning, heavily depends on both clean and sufficient labeled data, which is always difficult to acquire. Noisy unlabeled data, on the other hand, also provides valuable information that can be exploited using self-supervised training methods. In this work, we investigate how to pretrain speaker recognition models by leveraging dialogues between customers and smart-speaker devices. However, the supervisory information in such dialogues is inherently noisy, as multiple speakers may speak to a device in the course of the same dialogue. To address this issue, we propose an effective rejection mechanism that selectively learns from dialogues based on their acoustic homogeneity. Both reconstruction-based and contrastive-learning-based self-supervised methods are compared. Experiments demonstrate that the proposed method provides significant performance improvements, superior to earlier work. Dialogue pretraining when combined with the rejection mechanism yields 27.10% equal error rate (EER) reduction in speaker recognition, compared to a model without self-supervised pretraining.

cs.LG↗

arXiv2021

Generalized Likelihood Ratio Test for Adversarially Robust Hypothesis Testing

Machine learning models are known to be susceptible to adversarial attacks which can cause misclassification by introducing small but well designed perturbations. In this paper, we consider a classical hypothesis testing problem in order to develop fundamental insight into defending against such adversarial perturbations. We interpret an adversarial perturbation as a nuisance parameter, and propose a defense based on applying the generalized likelihood ratio test (GLRT) to the resulting composite hypothesis testing problem, jointly estimating the class of interest and the adversarial perturbation. While the GLRT approach is applicable to general multi-class hypothesis testing, we first evaluate it for binary hypothesis testing in white Gaussian noise under $\ell_{\infty}$ norm-bounded adversarial perturbations, for which a known minimax defense optimizing for the worst-case attack provides a benchmark. We derive the worst-case attack for the GLRT defense, and show that its asymptotic performance (as the dimension of the data increases) approaches that of the minimax defense. For non-asymptotic regimes, we show via simulations that the GLRT defense is competitive with the minimax approach under the worst-case attack, while yielding a better robustness-accuracy tradeoff under weaker attacks. We also illustrate the GLRT approach for a multi-class hypothesis testing problem, for which a minimax strategy is not known, evaluating its performance under both noise-agnostic and noise-aware adversarial settings, by providing a method to find optimal noise-aware attacks, and heuristics to find noise-agnostic attacks that are close to optimal in the high SNR regime.

stat.ML↗

arXiv2021

All-Digital LoS MIMO with Low-Precision Analog-to-Digital Conversion

Line-of-sight (LoS) multi-input multi-output (MIMO) systems exhibit attractive scaling properties with increase in carrier frequency: for a fixed form factor and range, the spatial degrees of freedom increase quadratically for 2D arrays, in addition to the typically linear increase in available bandwidth. In this paper, we investigate whether modern all-digital baseband signal processing architectures can be devised for such regimes, given the difficulty of analog-to-digital conversion for large bandwidths. We propose low-precision quantizer designs and accompanying spatial demultiplexing algorithms, considering 2x2 LoS MIMO with QPSK for analytical insight, and 4x4 MIMO with QPSK and 16QAM for performance evaluation. Unlike prior work, channel state information is utilized only at the receiver (i.e., transmit precoding is not employed). We investigate quantizers with regular structure whose high-SNR mutual information approaches that of an unquantized system. We prove that amplitude-phase quantization is necessary to attain this benchmark; phase-only quantization falls short. We show that quantizers based on maximizing per-antenna output entropy perform better than standard Minimum Mean Squared Quantization Error (MMSQE) quantization. For spatial demultiplexing with severely quantized observations, we introduce the novel concept of virtual quantization which, combined with linear detection, provides reliable demodulation at significantly reduced complexity compared to maximum likelihood detection.

eess.SP↗

arXiv2021

Robust Adversarial Learning via Sparsifying Front Ends

It is by now well-known that small adversarial perturbations can induce classification errors in deep neural networks. In this paper, we take a bottom-up signal processing perspective to this problem and show that a systematic exploitation of sparsity in natural data is a promising tool for defense. For linear classifiers, we show that a sparsifying front end is provably effective against $\ell_{\infty}$-bounded attacks, reducing output distortion due to the attack by a factor of roughly $K/N$ where $N$ is the data dimension and $K$ is the sparsity level. We then extend this concept to deep networks, showing that a "locally linear" model can be used to develop a theoretical foundation for crafting attacks and defenses. We also devise attacks based on the locally linear model that outperform the well-known FGSM attack. We supplement our theoretical results with experiments on the MNIST and CIFAR-10 datasets, showing the efficacy of the proposed sparsity-based defense schemes.

stat.ML↗

arXiv2021

Sparse Coding Frontend for Robust Neural Networks

Deep Neural Networks are known to be vulnerable to small, adversarially crafted, perturbations. The current most effective defense methods against these adversarial attacks are variants of adversarial training. In this paper, we introduce a radically different defense trained only on clean images: a sparse coding based frontend which significantly attenuates adversarial attacks before they reach the classifier. We evaluate our defense on CIFAR-10 dataset under a wide range of attack types (including Linf , L2, and L1 bounded attacks), demonstrating its promise as a general-purpose approach for defense.

cs.LG↗

arXiv2021

Wireless Fingerprinting via Deep Learning: The Impact of Confounding Factors

Can we distinguish between two wireless transmitters sending exactly the same message, using the same protocol? The opportunity for doing so arises due to subtle nonlinear variations across transmitters, even those made by the same manufacturer. Since these effects are difficult to model explicitly, we investigate learning device fingerprints using complex-valued deep neural networks (DNNs) that take as input the complex baseband signal at the receiver. We ask whether such fingerprints can be made robust to distribution shifts across time and locations due to clock drift and variations in the wireless channel. In this paper, we point out that, unless proactively discouraged from doing so, DNNs learn these strong confounding features rather than the nonlinear device-specific characteristics that we seek to learn. We propose and evaluate strategies, based on augmentation and estimation, to promote generalization across realizations of these confounding factors, using data from WiFi and ADS-B protocols. We conclude that, while DNN training has the advantage of not requiring explicit signal models, significant modeling insights are required to focus the learning on the effects we wish to capture.

eess.SP↗

arXiv2020

A Neuro-Inspired Autoencoding Defense Against Adversarial Perturbations

Deep Neural Networks (DNNs) are vulnerable to adversarial attacks: carefully constructed perturbations to an image can seriously impair classification accuracy, while being imperceptible to humans. While there has been a significant amount of research on defending against such attacks, most defenses based on systematic design principles have been defeated by appropriately modified attacks. For a fixed set of data, the most effective current defense is to train the network using adversarially perturbed examples. In this paper, we investigate a radically different, neuro-inspired defense mechanism, starting from the observation that human vision is virtually unaffected by adversarial examples designed for machines. We aim to reject L^inf bounded adversarial perturbations before they reach a classifier DNN, using an encoder with characteristics commonly observed in biological vision: sparse overcomplete representations, randomness due to synaptic noise, and drastic nonlinearities. Encoder training is unsupervised, using standard dictionary learning. A CNN-based decoder restores the size of the encoder output to that of the original image, enabling the use of a standard CNN for classification. Our nominal design is to train the decoder and classifier together in standard supervised fashion, but we also consider unsupervised decoder training based on a regression objective (as in a conventional autoencoder) with separate supervised training of the classifier. Unlike adversarial training, all training is based on clean images. Our experiments on the CIFAR-10 show performance competitive with state-of-the-art defenses based on adversarial training, and point to the promise of neuro-inspired techniques for the design of robust neural networks. In addition, we provide results for a subset of the Imagenet dataset to verify that our approach scales to larger images.

cs.LG↗

arXiv2020

Adversarially Robust Classification based on GLRT

Machine learning models are vulnerable to adversarial attacks that can often cause misclassification by introducing small but well designed perturbations. In this paper, we explore, in the setting of classical composite hypothesis testing, a defense strategy based on the generalized likelihood ratio test (GLRT), which jointly estimates the class of interest and the adversarial perturbation. We evaluate the GLRT approach for the special case of binary hypothesis testing in white Gaussian noise under $\ell_{\infty}$ norm-bounded adversarial perturbations, a setting for which a minimax strategy optimizing for the worst-case attack is known. We show that the GLRT approach yields performance competitive with that of the minimax approach under the worst-case attack, and observe that it yields a better robustness-accuracy trade-off under weaker attacks, depending on the values of signal components relative to the attack budget. We also observe that the GLRT defense generalizes naturally to more complex models for which optimal minimax classifiers are not known.

stat.ML↗