A Reconfigurable Multiclass Support Vector Machine Architecture for Real-Time Embedded Systems Classification

Kane, Jason; Hernandez, Robert; Yang, Qing

doi:10.1109/fccm.2015.24

Cited by 6 publications

(13 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Two approaches were presented, the multi-class support vector machine polynomial kernel and multi-class support vector machine Linear Kernel using "winner take all". Even though, the results of using the one-versus-all or "winner take all" produced remarkable results; 98.57% for the polynomial kernel and 97.04% for the linear kernel, Kane et al [25] points out usefulness for a one-versus-one multi-class SVM for real-time FPGA implementation, as it requires minimal on-board memory during training.…”

Section: Related Workmentioning

confidence: 99%

“…Kane et al [25] proposed the first ever fully pipelined, floating point based, multi-use reconfigurable hardware ar-chitecture designed to act in conjunction with embedded processing as an accelerator for multi-class SVM classification. Their implementation show the benefits of one-versus-one against one-versus-all for real-time applications.…”

Section: Related Workmentioning

confidence: 99%

“…The approach used in this work is motivated by the one-versus-one multi-class linear SVM with majority voting presented in [22] and it's efficient implementation for realtime applications as presented in [25]. Most of the reviewed implementations focus more on improving the accuracy of the classifier with little or no consideration of real-time implementation, which is mostly needed for assistive care.…”

Section: Related Workmentioning

confidence: 99%

See 2 more Smart Citations

Ensemble One-vs-One SVM Classifier for Smartphone Accelerometer Activity Recognition

Bouchut

Appiah

Lotfi

et al. 2018

2018 IEEE 20th International Conference on High Performance Computing and Communications; IEEE 16th International Conference On

View full text Add to dashboard Cite

A recognition framework to identify six full body motion from smartphone sensory data is proposed. The proposed system relies on accelerometer, gyroscope and magnetometer data to classify user activities into six groups (sitting, standing, lying down, walking, walking up stairs and walking downstairs). The proposed solution is an improvement of a oneverse-one SVM classifier with an ensemble of different learning methods each trained to discriminate a single activity against another. The improvement presented here doesn't only focus on accuracy but also potential embedded implementation capable of performing real-time classification with mobile data from the cloud. The presented one-versus-one approach, based on a linear kernel achieved 97.50 percent accuracy on a public dataset; second best to 98.57 percent reported in literature which uses a polynomial kernel.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Ensemble One-vs-One SVM Classifier for Smartphone Accelerometer Activity Recognition

Bouchut

Appiah

Lotfi

et al. 2018

2018 IEEE 20th International Conference on High Performance Computing and Communications; IEEE 16th International Conference On

View full text Add to dashboard Cite

show abstract

“…Works that targeted on accelerating SVM with a coprocessor unit [1], [10]- [12] tend to focus on the kernel due to its compute-intensive task and also its innate nature to be parallelize. Prior work by Kane et al [13] implemented a generic SVM classification architecture that was tested with a wide variety of datasets. However, the analysis is only limited to comparison with software implementations only.…”

Section: Introductionmentioning

confidence: 99%

Reconfigurable Logic Embedded Architecture of Support Vector Machine Linear Kernel

Sirkunan¹,

Shaikh-Husin²,

Andromeda³

et al. 2017

EECSI

View full text Add to dashboard Cite

Support Vector Machine (SVM) is a linear binary classifier that requires a kernel function to handle non-linear problems. Most previous SVM implementations for embedded systems in literature were built targeting a certain application; where analyses were done through comparison with software implementations only. The impact of different application datasets towards SVM hardware performance were not analyzed. In this work, we propose a parameterizable linear kernel architecture that is fully pipelined. It is prototyped and analyzed on Altera Cyclone IV platform and results are verified with equivalent software model. Further analysis is done on determining the effect of the number of features and support vectors on the performance of the hardware architecture. From our proposed linear kernel implementation, the number of features determine the maximum operating frequency and amount of logic resource utilization, whereas the number of support vectors determines the amount of on-chip memory usage and also the throughput of the system.

show abstract

“…al 9 state that the computational process of Support Vector Machine classification suffers greatly from a large number of iterative mathematical operations and an overall complex algorithmic structure. Their work proposed a fully pipelined, floating point based, multi-use reconfigurable hardware architecture designed to act in conjunction with embedded processing as an accelerator for multiclass SVM classification.…”

Section: Related Workmentioning

confidence: 99%

Classifier Optimized for Resource-constrained Pervasive Systems and Energy-efficiency

Karvonen¹,

Jiménez²,

Simón³

et al. 2017

IJCIS

View full text Add to dashboard Cite

Computational intelligence is often used in smart environment applications in order to determine a user's context. Many computational intelligence algorithms are complex and resource-consuming which can be problematic for implementation devices such as FPGA:s, ASIC:s and low-level microcontrollers. These types of devices are, however, highly useful in pervasive and mobile computing due to their small size, energy-efficiency and ability to provide fast real-time responses. In this paper, we propose a classifier, CORPSE, specifically targeted for implementation in FPGA:s, ASIC:s or low-level microcontrollers. CORPSE has a small memory footprint, is computationally inexpensive, and is suitable for parallel processing. The classifier was evaluated on eight different datasets of various types. Our results show that CORPSE, despite its simplistic design, has comparable performance to some common machine learning algorithms. This makes the classifier a viable choice for use in pervasive systems that have limited resources, requires energy-efficiency, or have the need for fast real-time responses.

show abstract

A Reconfigurable Multiclass Support Vector Machine Architecture for Real-Time Embedded Systems Classification

Cited by 6 publications

References 18 publications

Ensemble One-vs-One SVM Classifier for Smartphone Accelerometer Activity Recognition

Ensemble One-vs-One SVM Classifier for Smartphone Accelerometer Activity Recognition

Reconfigurable Logic Embedded Architecture of Support Vector Machine Linear Kernel

Classifier Optimized for Resource-constrained Pervasive Systems and Energy-efficiency

Contact Info

Product

Resources

About