Towards Network Reliability: Harnessing Interference and Dynamics
In the past few decades, networked systems have revolutionized, among other things, the way we communicate, and the way we control physical processes. However, addressing the unreliability associated with such networked systems continues to be a fundamental challenge. The unreliability in such networked systems can be attributed to a variety of factors including the lack of co-ordination among network components, noisy measurements, and security vulnerabilities leading to malicious elements in the network. In view of the factors mentioned above, this dissertation takes steps towards enhancing network reliability by addressing some fundamental challenges in two contemporary network setups: wireless communication networks and cyber-physical systems (CPS).
In the context of wireless communication networks, we focus on the unreliability stemming from interference, i.e., a scenario where multiple transmitter-receiver pairs sharing the same frequency band interfere with each other. This is a fundamental bottleneck in reliably scaling data rates; an essential requirement for supporting the huge growth in mobile Internet traffic. The temporal nature of interference in such networks depends on the underlying traffic and the resource allocation decisions of neighboring base stations. In practice, due to the bursty nature of data traffic and uncoordinated resource allocations across base stations, the resulting interference at the physical layer tends to be bursty. Though recent advances in network information theory have led to a fundamental understanding of interference in wireless networks, the channel models usually assume that interference is always present and hence do not capture the impact of bursty interference. Thus, the following question emerges: how can we harness bursty interference and potentially improve the (information theoretic) system capacity? In this dissertation, we investigate the above question in the context of parallel (multicarrier) interference channels, and show positive results, i.e., we demonstrate that harnessing burstiness leads to a non-trivial increase in the system capacity. We develop interference management schemes which leverage feedback from receivers to recover interfered symbols in previous transmissions. For the case when feedback from receivers is not available, we develop opportunistic schemes based on a degraded message set approach to exploit burstiness. In addition, we develop tight outer bounds for a variety of regimes, and hence prove the information theoretic optimality of our interference management schemes in those regimes.
In the context of CPS, we focus on the unreliability arising out of security vulnerabilities. Due to the close interaction between the cyber and physical components, CPS pose unique security challenges. Furthermore, conventional cyber security methods which are oblivious to the underlying physical dynamics leave open the possibility of leveraging such dynamics for addressing security vulnerabilities in CPS. With this motivation, and considering the fact that state estimation is a crucial component in CPS, we study the problem of securing state estimation in linear dynamical systems despite active adversaries. In particular, we focus on two attack scenarios: (i) attacks on software/hardware where state estimation is performed (observer attacks), and (ii) attacks on sensors used for state estimation. To protect against observer attacks we propose an architecture where state estimation is performed across multiple computing nodes (observers), and derive tight bounds on the number of attacked observers which can be tolerated for accurate state estimation. To protect against sensor attacks, we propose a secure state estimation algorithm, and derive (optimal) bounds on the achievable state estimation error given an upper bound on the number of attacked sensors. The proposed state estimator involves Kalman filters operating over subsets of sensors to search for a sensor subset which is reliable for state estimation. As a result of independent
interest, we give a coding theoretic view of attack detection and state estimation against
sensor attacks in a noiseless dynamical system.
Hence, in a nutshell, this dissertation takes fundamental steps towards enhancing the reliability of wireless communication networks (by harnessing bursty interference), and cyber-physical systems (by leveraging physical dynamics for security).