NSF: CNS: Towards Sustainable, Efficient and Secure Low-Power IoT Systems

I. Project Overview:

To meet the ever-growing need of energy supply for internet of things (IoT), energy harvesting is emerging as a promising alternative to traditional power grids and batteries. The goal of this project is to provide energy-efficient and secure wireless communication solutions to the renewable energy powered IoT. This project conducts fundamental research on network synchronization, efficient data transmission and security and tackles the challenges from practical issues such as limited power supply and nonlinear circuits and unique features of renewable energy powered systems. The project is expected to enable large-scale data collection and processing through renewable energy powered IoT devices. In addition, the research activities in the project will provide a rich environment and platform that can facilitate educating and training students from high school level to graduate level.

This project is organized around three research topics. The first topic develops a wake-up radio based time synchronization method, which can achieve high synchronization precision and ultra-low energy cost. In the second topic, a new nonlinear recursive energy harvesting model that considers practical challenges in the wireless systems will be developed and several optimal data transmission strategies will be proposed to boost the throughput of IoT networks. The third topic studies new security threats caused by malicious energy attacks and develops reinforcement learning-based intelligent attack and defense mechanisms. This project will advance close interactions of multiple fields in networking, energy harvesting and security and motivate a wide range of research within sustainability, efficiency and security in IoT.

II. Energy-Preserving Time Synchronization

Motivations: Given the limited power supply from the renewable energy, an energy harvesting node (EHN) may stay asleep most of the time and be briefly activated when communication or sensing is needed. The communicating parties thus must be synchronized, otherwise asleep receivers will miss messages from the sender. The conventional time synchronization protocols rely on the timestamp exchange for clock skew and offset estimation. In the initial synchronization stage of these protocols, nodes need to keep idle listening until the initial synchronization is finished. Therefore, nodes generally have heavy energy consumption in the initial phase of timestamp-based time sync methods (Table 1) and making conventional time synchronization not work well in ultra-low-power energy harvesting wireless networks. 

In this project, we develop a new WUR-based time synchronization method for ultra-low-power wireless networks with energy harvesting capabilities. Compared with existing methods, WUR-TS does not rely on timestamp exchange but uses short energy tones to synchronize the nodes within the receiving range. WUR-TS is a centralized synchronization method based on wake-up radio circuit, which can provide a variety of synchronization accuracy to meet different application requirements while consuming much lower energy than timestamp-based synchronization methods. In WUR-TS, energy harvesting nodes (EHNs) do not exchange any timing information, but only receive the wake-up signal broadcast periodically by the central node to synchronize time throughout the network. To achieve this, the central node is scheduled to periodically broadcast a wake-up signal. Afterward, the EHN calculates the arrival time of the wake-up signal to estimate the clock skew.

II. A. Fully-Passive WUR-based Time Synchronization

Fully-passive WuRs contain only passive components such as Schottky diodes, complementary metal-oxide- semiconductor (CMOS), and capacitors. The fully-passive WuR circuit consumes zero energy in the idle status. According to experiment results, it can achieve an average of 5ms and 70μs time precision for rough synchronization and fine synchronization, respectively. The fully-passive solution has extremely low energy cost: 1.2 μW in idle state, 3.65 μJ on one round of synchronization.

II. B. Semi-Passive WUR-based Time Synchronization

Semi-Passive WUR-TS contains an active component, a ultra-low-power comparator in the system design. It is capable of achieving a higher synchronization accuracy than the fully-passive solution, at the cost of higher energy consumption. Based on the experimental data, we developed a model to accurately estimate the arrival time of each wake-up signal. By using this model, the EHN can calculate its clock drift in complex radio environments. We have implemented with a minimum number of commercial components. According to the experimental results, Semi-Passive WUR-TS can achieve a synchronization accuracy of 3 μs when the power supply voltage is 2.8 V and the received wake-up signal strength is higher than −33 dBm. Semi-Passive WUR-TS is an ultra-low-power synchronization method. If no wake-up signal is detected, the power consumption of each EHN is 3.2 μW. If the wake-up signal is detected, the EHN consumes only 3.6 μJ of energy to complete time synchronization.

III. Efficient Communication Design in RF Energy Harvesting Powered IoT

Existing transmission strategies based on unrealistic assumptions are infeasible in practice. Through experimental study, we reveal several practical issues, including the nonlinear battery charging, and dropping supply voltage, in the wireless systems that will challenge the optimal transmission policy design. 

• Nonlinear battery charging: Due to the nonlinear charging characteristics of supercapacitor, the amount of harvested energy is not a predetermined value as assumed in the conventional model, but a variable depending on the battery’s residual energy. Meanwhile, the residual energy is affected by the data scheduling, which in turn causes a feedback loop from data transmission to harvested energy. Without considering the dependency of energy harvest on data transmission resulting from the nonlinear charging feature of battery, the existing data transmission scheduling strategies that are based on inaccurate estimation of the harvested energy is thus infeasible in reality.

• Dropping supply voltage during successive transmissions: Due to the limited energy storage capacity of IoT nodes (e.g., 1 mF capacitor) resulted from low RF harvesting rates, the voltage of the energy storage, which also serves as the IoT’s supply voltage, may experience a significant drop during successive transmissions. Besides communications, other energy consuming operations, such as ADC sampling and computing, will also result in a significant falling of supply voltage. Conventional energy consumption model that assumes constant supply voltage will be inaccurate.

In our research, we put forward a nonlinear model for energy harvesting, devised to precisely predict the amount of energy to be collected for effective transmission scheduling. Moreover, we incorporate the hardware characteristics of IoT devices into our model to accurately estimate their energy consumption.

IV. Emerging Energy Attack in Renewable Energy Powered IoT Networks

Malicious energy attack is a unique attack method in RF energy harvesting wireless networks, that threatens the information security by consciously charging energy to manipulate routing paths in the network layer. In this attack, a malicious energy source actively charges specific EHNs in a network. The infected nodes that receive extra energy from the energy attacker will become more active than ordinary nodes to work as data forwarders or information aggregators. If the energy attacker is able to select the infected nodes properly, it can “encourage” most of the data traffic into passing through a compromised node who was originally deviated from the source and the destination. In this way, the energy attacker can create opportunities for an eavesdropper to sniff confidential data from any target node.

As an emerging attack method, the malicious energy attack is immune to many security mechanisms since it is an indirect attack method that disrupts the network protocols through energy, a “harmless” or even “beneficial” resource in the environment. To be specific, the legality of information can be verified by inserting artificial imprints (e.g., packet identification, key, and preamble), however, it is difficulty for current energy harvesting systems to distinguish the malicious energy from the ambient energy. This makes malicious energy attack have high seclusion against a traditional security method.

V. Education and Outreach Activities

Course Development: 

• ECE 8990 Special Topics in IoT and IoT Security, Mississippi State University 

• CS 648 IoT and IoT Security, University of Alabama

Outreach Activities: 

• MathCount-Tuscaloosa 2023 STEM Demo -- RF-powered IoT (Location: University of Alabama campus)

• IoT Lab Tour in the Mississippi Summer Transportation Institute (MSTI) 2023 Summer Camp (Location: Mississippi State University campus)

V. Personnels

Faculty: 

• Dr. Yu Luo (Lead PI) from Mississippi State University 

• Dr. Lina Pu (PI) from the University of Alabama

Graduate Students: 

• Jing Yang (PhD Candidate, Expected to graduate in September 2023 from Mississippi State University)

• Nitish Patil (PhD Student from Mississippi State University )

• Johnson A. Uduka (Master Student from Mississippi State University)

• Long Li (PhD Candidate from University of Alabama)

• Xinyu Zhang (PhD student from University of Alabama)

Undergraduate Students:

• Michael Reis (May 2023 -- Present, REU Student from University of Alabama)

• Kevin Liao (Spring 2021, MSU ORED student from Mississippi State University)

• Marco Lopez Bustillo (Fall 2019 -- Spring 2020, MSU ORED student from Mississippi State University)

Publications:

1. ​Yu Luo, and Lina Pu, "UAV remotely-powered underground IoT for soil monitoring," IEEE Transactions on Industrial Informatics, 2023. DOI: 10.1109/TII.2023.3272016

2. Yu Luo, Lina Pu, and Chun-Hung Liu "CPU frequency scaling optimization in sustainable edge computing", IEEE Transactions on Sustainable Computing, vol. 8, no. 2, 194 - 207, 2022. DOI: 10.1109/TSUSC.2022.3217970

3. Yu Luo, Lina Pu, and Zheng Peng, "Energy stimulated time synchronization for energy harvesting wireless networks," IEEE Transactions on Network Science and Engineering, 2022. DOI: 10.1109/TNSE.2022.3154355

4. Long Li, Yu Luo, Jing Yang, and Lina Pu, "Reinforcement learning enabled intelligent energy attack in green IoT networks," IEEE Transactions on Information Forensics and Security, 2022. DOI: 10.1109/TIFS.2022.3149148

5. Yu Luo, and Lina Pu, "WUR-TS: semi-passive wake-up radio receiver based time synchronization method for energy harvesting wireless networks," IEEE Transactions on Mobile Computing, 2021. DOI: 10.1109/TMC.2021.3064374

6. Yu Luo, Lina Pu, and Lei Lei, “Impact of varying radio power density on wireless communications of RF energy harvesting systems," IEEE Transactions on Communications, vol. 69, no. 3, pp. 1960 - 1974, 2020. DOI: 10.1109/TCOMM.2020.3040397

7. Long Li, Yu Luo, and Lina Pu, "Q-learning enabled intelligent energy attack in sustainable wireless communication networks", In Proceedings of the IEEE International Conference on Communications (ICC), pp. 1-7, 2021.

8. Yu Luo, and Lina Pu, "Practical issues of RF energy harvest and data transmission in renewable radio energy powered IoT," IEEE Transactions on Sustainable Computing, vol. 6, no. 4, pp. 667-678, 2020. DOI: 10.1109/TSUSC.2020.3000085

9. Yu Luo, and Lina Pu, "ESTS: energy stimulated time synchronization for energy harvesting wireless networks", In Proceedings of the IEEE Global Communications Conference (GLOBECOM), pp. 1-6, IEEE, 2020.

10. Yu Luo, Lina Pu, Guodong Wang, and Yanxiao Zhao, “RF energy harvesting wireless communications: RF environment, device hardware and practical issues," Sensors, vol. 19, no. 13, pp. 3010-3037, 2019. DOI:10.3390/s19133010

Poster Presentations (students):

1. Kevin Liao, Yu Luo, and Jing Yang, “Poster: Semi-passive wake-up radio based time synchronization”, MSU Summer Undergraduate Research Symposium, 2021.