Table 2.
Machine Learning Techniques in Resource-Constrained Environments.
| Reference | ML Method | Embedded/Mobile Platform | Application | Year |
|---|---|---|---|---|
| [2] | SVM | ARMv7, IBM PPC440 | Network Configuration | 2015 |
| [20] | DNN | FPGA Zedboard with 2 ARM Cortex Cores | Character Recognition | 2015 |
| [22] | DNN | Xilinx FPGA board | Image classification | 2016 |
| [23] | LSTM RNN | Zynq 7020 FPGA | Character Prediction | 2016 |
| [25] | CNN | VC707 Board with Xilinx FPGA chip | Image Classification | 2015 |
| [39] | GMM | Raspberry Pi | Integer processing | 2014 |
| [40] | k-NN, SVM | Mobile Device | Fingerprinting | 2014 |
| [41] | k-NN | Mobile Device | Fingerprinting | 2014 |
| [42] | k-NN, GMM | Mobile Device | Mobile Device Identification | 2015 |
| [43] | SVM | Xilinx Virtex 7 XC7VX980 FPGA | Histopathological image classification | 2015 |
| [44] | HMM | Nvidia Kepler | Speech Recognition | 2015 |
| [45] | Logistic Regression | Smart band | Stress Detection | 2015 |
| [46] | k-means | Smartphone | Indoor Localization | 2015 |
| [47] | Naïve Bayes | AVR ATmega-32 | Home Automation | 2015 |
| [48] | k-NN | Smartphone | Image Recognition | 2015 |
| [49] | Decision Tree | Mobile Device | Health Monitoring | 2015 |
| [50] | GMM | FRDM-K64F equipped with ARM Cortex-M4F core | IoT sensor data analysis | 2016 |
| [51] | CNN | FPGA Xilinx Zynq ZC706 Board | Image Classification | 2016 |
| [52] | CNN | Mobile Device | Mobile Sensing | 2016 |
| [53] | SVM | Mobile Device | Fingerprinting | 2016 |
| [54] | k-NN, SVM | Mobile Device | Fingerprinting | 2016 |
| [55] | k-NN | Xilinx Virtex-6 FPGA | Image Classification | 2016 |
| [56] | HMM | Arduino UNO | Disease detection | 2016 |
| [57] | Logistic Regression | Wearable Sensor | Stress Detection | 2016 |
| [58] | Naïve Bayes | Smartphone | Health Monitoring | 2016 |
| [59] | Naïve Bayes | Mobile Devices | Emotion Recognition | 2016 |
| [60] | k-NN | Smartphone | Data Mining | 2016 |
| [61] | HMM | Smartphone Sensors | Activity Recognition | 2017 |
| [62] | DNN | Smartphone | Face detection, activity recognition | 2017 |
| [63] | CNN | Mobile Device | Image classification | 2017 |
| [64] | SVM | Mobile Device | Mobile Device Identification | 2017 |
| [65] | SVM | Jetson-TK1 | Healthcare | 2017 |
| [66] | SVM, Logistic Regression | Arduino UNO | Stress Detection | 2017 |
| [67] | Naïve Bayes | Smartphone | Emotion Recognition | 2017 |
| [68] | k-means | Smartphones | Safe Driving | 2017 |
| [69] | HMM | Mobile Device | Health Monitoring | 2017 |
| [70] | k-NN | Arduino UNO | Image Classification | 2017 |
| [71] | SVM | Wearable Device (nRF51822 SoC+BLE) | Battery Life Management | 2018 |
| [72] | SVM | Zybo Board with Z-7010 FPSoC | Face Detection | 2018 |
| [73] | CNN | Raspberry Pi + Movidus Neural Compute Stick | Vehicular Edge Computing | 2018 |
| [74] | CNN | Jetson TX2 | Image Classification | 2018 |
| [75] | HMM | Smartphone | Healthcare | 2018 |
| [76] | k-NN | Smartphone | Health Monitoring | 2019 |
| [77] | Decision Trees | Arduino UNO | Wound Monitoring | 2019 |
| [78] | RNN | ATmega640 | Smart Sensors | 2019 |
| [79] | SVM, Logistic Regression, k-means, CNN | Raspberry Pi | Federated Learning | 2019 |
| [80] | DNN | Raspberry Pi | Transient Reduction | 2020 |
| [81] | MLP | Embedded SoC (ESP4ML) | Classification | 2020 |
| [82] | HMM | Smartphone | Indoor Localization | 2020 |
| [83] | k-NN | Smartphone | Energy Management | 2020 |
| [84] | ANN, Decision Trees | Raspberry Pi | Classification and Regression | 2021 |