1. Protocols for Low-Power Wide-Area Netwok in White Spaces

Source of support: NSF (2019--2024)

As a key technology driving the Internet-of-Things (IoT), Low-Power Wide-Area Networks (LPWANs) are evolving to overcome the range limits and scalability challenges in traditional wireless sensor networks. With the support from NSF and through collaboration with Microsoft Research, we have developed a highly scalable LPWAN architecture – called SNOW (Sensor Network Over White Spaces)- by designing sensor networks to operate over the TV white spaces. White spaces refer to the allocated but locally unused TV channels, and can be used by unlicensed devices, according to the FCC in the US. Compared to the existing LPWAN technologies, SNOW offers much higher scalability and energy efficiency and takes the advantages of freely available TV white spaces.

 SNOW is the first design of a highly scalable low power and long range wireless sensor network over the TV white spaces. At the heart of its design is a Distributed implementation of Orthogonal Frequency Division Multiplexing (OFDM), called D-OFDM. The base station splits the wide white space spectrum into narrowband orthogonal subcarriers allowing D-OFDM to carry parallel data streams to/from the distributed nodes from/to the base station. Each sensor uses only one narrow-band radio. The base station uses two wide-band radios, one for transmission and the other for reception, allowing transmission and reception in parallel. Each radio of the base station and a sensor is half-duplex and equipped with a single antenna. SNOW supports reliable, concurrent, and asynchronous receptions with one single-antenna radio and multiple concurrent data transmissions with the other single-antenna radio. This is achieved through a new physical layer design by adopting D-OFDM for multiple access in both directions and through a lightweight MAC protocol. While OFDM has been embraced for multiple access in various wireless broadband and cellular technologies recently, its adoption in low power, low data rate, narrowband, and sensor network design is novel. Taking the advantage of low data rate and short payloads, we adopt OFDM in SNOW through a much simpler and energy-efficient design.

In SNOW, a subcarrier bandwidth can be chosen as low as 100kHz, 200kHz, 400kHz depending on the packet size and expected bit rate. Using a subcarrier, a sensor node can have several kilometers of transmission range at 0dBm. Using a single base station, SNOW has the capability of supporting millions of devices. As an added advantage, it can use fragmented spectrum of white spaces. We implemented SNOW on two hardware platforms -- USRP using GNU radio and TI CC1310. CC1310 is a tiny, cheap (<$30), and commercially off-the-shelf (COTS) device with a programmable PHY. We implemented SNOW with CC1310 (using CC-ANTENNA-DK2 antenna)  as a node  and the BS is a USRP210 with a laptop. Experiments through deployments in multiple geographical areas as well as large-scale simulations demonstrated that SNOW drastically enhances the scalability of sensor network and outperforms existing techniques in terms of scalability, energy, and latency. Both analytical and experimental study hint that SNOW will be one of the most scalable and the key technologies to drive the IoT. The design of SNOW is available here.

This YouTube video shows a short demonstration of SNOW: https://youtu.be/y7Q932A24Zc

This project will design and implement an LPWAN architecture and complete protocol stack based on SNOW to support scalable integration, coexistence, mobility, and time-sensitive communication as follows. It will implement the proposed protocols on TI CC1310 (as SNOW nodes) and also on universal software radio peripheral devices. The protocols will be evaluated through experiments in on our LPWAN testbeds. 

SNOW Testbed:

Currently we have a temporary SNOW testbed of 43 nodes for experimental purposes that co-locates with our permanent wireless sensor network (WSN) testbed at Wayne State University. Specifically, we have 43 SNOW nodes of which 34 are based on CC1310 and 9 are based on USRP which are co-located with our WSN testbed as shown in the following figure. Currently all SNOW nodes are powered through USB from Laptop and the laptops are connected with those of WSN testbed through department Ethernet. The entire testbed is located in Maccabees Building at Wayne state University. 

Link to the open-source implementation of SNOW PHY: https://github.com/snowlab12/gr-snow

2. Low-Power Wide-Area Networks for Industrial Automation

    Source of support: NSF (2020--2024) 

This project proposes to adopt the Low-Power Wide-Area Network (LPWAN) technologies for industrial automation. Due to long-range, LPWANs can be adopted without complex configuration and at a fraction of costs for wide-area industrial Internet of Things (IoT) applications, compared to multi-hop solutions. This project will develop theoretical foundations and systems for enabling industrial automation using LPWANs. This project will particularly consider LoRa, a leading LPWAN technology. Adopting LoRa for industrial automation poses some evolutionarily challenges. The fundamental building blocks of any industrial automation system are feedback control loops that largely rely on real-time communication. Due to severe energy-constraint, LoRa uses a simple media access control protocol that is unsuited for real-time communication. It needs to adopt low duty-cycling in several regions (e.g., Europe). Besides, to optimize performance, industrial automation needs a codesign of real-time scheduling and control. Such a codesign becomes specially challenging in LoRa due to its large-scale and energy-limitations. 

3. Handling Wireless Coexistence

     Source of support: ONR (2021--2023) 

In this project, we handle the coexistence problem of LPWANs. The rapid growth of LPWANs in the limited spectrum brings forth the challenge of coexistence of many networks and devices in the same band. The immediate effect of such coexistence is degraded network performance in terms of throughput, latency, and energy. Some networks or devices may even suffer from spectrum starvation. Repeated attempts to access the spectrum will drain their batteries. In Naval applications, wireless monitoring operations can be severely disrupted if coexistence is not handled properly.  Today, LPWANs are not equipped to handle the impending challenge of coexistence. Their nodes have very low computation power, memory, and energy typically supplied from small batteries, making it difficult to employ sophisticated protocols. Coexistence handling for WiFi or traditional short-range wireless sensor network will not work for LPWANs. In massive crowds of coexisting networks, the interference pattern can be hard to detect for LPWAN nodes.  Due to long range, they are subject to an unprecedented number of hidden nodes, requiring new techniques for combating such interference. 

4. Handling Jamming of LPWAN

     Source of support: ONR (2022--2024) 

Internet of Things (IoT) applications rely on low-power wide-area networks (LPWANs) for gathering data from widely dispersed devices over long distances. As IoT applications are growing in various domains, LPWANs are getting prone to jamming attacks. Jamming causes excessive packet loss, throughput reduction, long delays, and lower energy efficiency in these networks. Existing anti-jamming work in IoT mainly considers jamming in the communication channel between the gateway/base station and the server. In this project, we propose to mitigate jamming in LPWANs using both link layer and physical layer approaches. 

5. Scaling Autonomous Vehicle Systems at the Edge: from On-Board Processing to Cloud Infrastructure

     Source of support: NSF (2021--2023) 

Connected and Autonomous Vehicles (CAVs) are expected to be the dominant vehicles of future transportation systems. This is due to their benefits including reduced driver stress, increased productivity, increased safety, and reduced energy consumption and pollution. The current CAV systems and the smart city infrastructure systems supporting vehicles operation suffer from several scalability issues that we aim to address in this project.

6. Towards the Design of a Large-Scale Wireless Sensor Network

     Source of support: NSF (2016--2018)