One-bit compressive sampling with time-varying thresholds for sparse parameter estimation

Gianelli, Christopher; Xu, Luzhou; Li, Jian; Stoica, Petre

doi:10.1109/sam.2016.7569634

Cited by 51 publications

(35 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…where p * andp * denote the minimums of optimization problems (24) and (25), respectively. Further, let r * ∈ K and r * T u * T ∈K stand for the minimizers of optimization problems (24) and (25), respectively. We first show that p * ≤p * .…”

Section: Estimation Via Lasserre's Sdp Relaxationmentioning

confidence: 99%

“…Originally proposed [16,20,21] in the context of compressive sensing [22], the 1-bit sampling has a rich heritage of research in statistical signal processing [23][24][25][26] and signal reconstruction [27,28]. In the past few years, the 1-bit processing has received significant attention in numerous modern applications such as classical communications [29][30][31], massive multiple-input multiple-output (MIMO) [32,33], deep learning [34], dictionary learning [35,36], and radar [37][38][39].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Localization Performance of 1-Bit Passive Radars in NB-IOT Applications

Sedighi

Mishra

Shankar

et al. 2019

2019 IEEE 8th International Workshop on Computational Advances in Multi-Sensor Adaptive Processing (CAMSAP)

View full text Add to dashboard Cite

Location-based services form an important use-case in emerging narrowband Internet-of-Things (NB-IoT) networks. Critical to this offering is an accurate estimation of the location without overlaying the network with additional active sensors. The massive number of devices, low power requirement, and low bandwidths restrict the sampling rates of NB-IoT receivers. In this paper, we propose a novel low-complexity approach for NB-IoT target delay estimation in cases where one-bit analog-to-digital-converters (ADCs) are employed to sample the received radar signal instead of high-resolution ADCs. This problem has potential applications in the design of inexpensive NB-IoT radar and sensing devices. We formulate the target estimation as a multivariate fractional optimization problem and solve it via Lasserre's semi-definite program relaxation. Numerical experiments suggest feasibility of the proposed approach yielding high localization accuracy with a very low number of 1-bit samples.

show abstract

Section: Estimation Via Lasserre's Sdp Relaxationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Localization Performance of 1-Bit Passive Radars in NB-IOT Applications

Sedighi

Mishra

Shankar

et al. 2019

2019 IEEE 8th International Workshop on Computational Advances in Multi-Sensor Adaptive Processing (CAMSAP)

View full text Add to dashboard Cite

show abstract

“…Recently, one-bit quantization has once again attracted considerable research interest in radar [10], [24]- [26]. In [10], the conventional radar measurement scheme is replaced by the one-bit sampling with Gaussian and uniform dithering.…”

Section: Introductionmentioning

confidence: 99%

“…The one-bit samples can be efficiently processed by using conventional methods proposed for high-precision samples. Sampling with dithering is also known as sampling with time-varying thresholds in [24]- [26]. The problem of sparse parameter estimation with one-bit quantization was considered in [24].…”

Section: Introductionmentioning

confidence: 99%

Signal Processing of Coast-Ship Bistatic High-Frequency Surface-Wave Radar Using One-Bit Quantized Measurement

2020

View full text Add to dashboard Cite

The feasibility of one-bit quantization in coast-ship bistatic high-frequency surface-wave radar (HFSWR) is mainly discussed in order to enjoy the advantages of simplicity, low cost and low power consumption. Our theoretical derivation shows that the received signal after one-bit quantization can be expressed as a linear combination of infinite components. These components include the original component and the high-order components that can be regarded as the interference terms. Then, the variation of the magnitude of each component with the signal-to-noise ratio (SNR) is given. To reduce the negative impact of these high-order components on the radar system, the detailed signal forms are analyzed. Furthermore, some comparative simulations of one-bit quantization and traditional high-precision quantization are provided to verify the correctness of our theoretical analysis. Both theories and simulation results indicate that the high-order components can be ignored in coast-ship bistatic HFSWR and that one-bit quantization is effective. Finally, real data analysis also validates the efficiency of one-bit quantization in practice. INDEX TERMS One-bit radar, bistatic high-frequency surface-wave radar, MIMO radar.

show abstract

“…In fact, the aforementioned techniques assume that the received signal is available in full precision. The resulting error of quantization can then be modeled as additive noise that usually has little to no impact on algorithms that assume the infinite precision case, provided that the sampling resolution is high enough [31]. The signals of interest in many modern applications, albeit, are extremely wide band and may pass through several RF chains that require multitudinous uses of ADCs.…”

mentioning

confidence: 99%

One-Bit Radar Processing With Time-Varying Sampling Thresholds

Ameri

Bose

et al. 2019

IEEE Trans. Signal Process.

Self Cite

View full text Add to dashboard Cite

Target parameter estimation in active sensing, and particularly radar signal processing, is a long-standing problem that has been studied extensively. In this paper, we propose a novel approach for target parameter estimation in cases where one-bit analog-to-digital-converters (ADCs), also known as signal comparators with time-varying thresholds, are employed to sample the received radar signal instead of high-resolution ADCs. The considered problem has potential applications in the design of inexpensive radar and sensing devices in civilian applications, and can likely pave the way for future radar systems employing low-resolution ADCs for faster sampling and high-resolution target determination.We formulate the target estimation as a multivariate weighted-least-squares optimization problem that can be solved in a cyclic manner. Numerical results are provided to exhibit the effectiveness of the proposed algorithms.

show abstract

One-bit compressive sampling with time-varying thresholds for sparse parameter estimation

Cited by 51 publications

References 22 publications

Localization Performance of 1-Bit Passive Radars in NB-IOT Applications

Localization Performance of 1-Bit Passive Radars in NB-IOT Applications

Signal Processing of Coast-Ship Bistatic High-Frequency Surface-Wave Radar Using One-Bit Quantized Measurement

One-Bit Radar Processing With Time-Varying Sampling Thresholds

Contact Info

Product

Resources

About