SpecEES: DoS Resilience, Secrecy, and Throughput in Massive MIMO

点击翻译按钮获取中文摘要

The project's objective is to develop the foundations for realizing denial of service (DoS)-resilience, secrecy, and high throughput in next-generation wireless networks in which base stations or access points (APs) scale their antenna and computing resources to massive multiple-input/multiple-output (massive MIMO). This project considers a practical yet strong adversary that launches both active attacks, attempting to collapse network throughput by polluting AP measurement of Channel State Information (CSI), as well as passive attacks, that attempt to intercept communication via potentially distributed and nomadic eavesdroppers. The first project thrust targets to detect and defend against pilot contamination attacks with zero startup cost, i.e., without requiring prior measurements of conditions without an attack. The second project thrust experimentally explores passive eavesdropping under realworld factors such as a large but finite number of antennas and a discrete and limited set of modulation and coding schemes (MCS). With a discovery of MCS saturation vulnerability, the project will study power control designed to overcome this vulnerability and thwart the adversary. On the uplink, the third project thrust develops antenna index coding as means to target reception on a subset of the massive MIMO array. A key outcome will be both derivation of the secrecy rate as well as experimental evaluation of this new mechanism. All three project thrusts will include an extensive experimental study using the Rice massive MIMO platform and over-the-air measurements. As computing and communication resources must be devoted to realizing the aforementioned counter mechanisms, this project will yield empirical understanding of the tradeoffs in realizing both high spectral efficiency and security.The proposed research agenda will enable new wireless security defenses and will provide an in-depth understanding of securing next generation communication systems such as massive MIMO. This project will yield new understanding of the design space that spans from traditional throughput optimization to jointly realizing security features and providing security by design. This project will yield new understanding of the security capabilities and limits of designs based on CSI as a shared secret. This project will impact standards bodies as it will expose fundamental limits of existing security mechanisms and will show how enhancements to standards can yield dramatically improved security capabilities. This project will impact industry through demonstration of results coupled with the researchers' extensive collaborative industry network. Finally, the project includes an inter-disciplinary methodology and the team includes multiple Ph.D. students from under-represented groups.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.