Designing Flexible, Energy Efficient and Secure Wireless Solutions for the Internet of Things

Abstract

The Internet of Things (IoT) is an emerging concept where ubiquitous physical objects (things) consisting of sensor, transceiver, processing hardware and software are interconnected via the Internet. The information collected by individual IoT nodes is shared among other often heterogeneous devices and over the Internet.

This dissertation presents flexible, energy efficient and secure wireless solutions in the IoT application domain. System design and architecture designs are discussed envisioning a near-future world where wireless communication among heterogeneous IoT devices are seamlessly enabled.

Firstly, an energy-autonomous wireless communication system for ultra-small, ultra-low power IoT platforms is presented. To achieve orders of magnitude energy efficiency improvement, a comprehensive system-level framework that jointly optimizes various system parameters is developed. A new synchronization protocol and modulation schemes are specified for energy-scarce ultra-small IoT nodes. The dynamic link adaptation is proposed to guarantee the ultra-small node to always operate in the most energy efficiency mode, given an operating scenario. The outcome is a truly energy-optimized wireless communication system to enable various new applications such as implanted smart-dust devices.

Secondly, a configurable Software Defined Radio (SDR) baseband processor is designed and shown to be an efficient platform on which to execute several IoT wireless standards. It is a custom SIMD execution model coupled with a scalar unit and several architectural optimizations: streaming registers, variable bitwidth, dedicated ALUs, and an optimized reduction network. Voltage scaling and clock gating are employed to further reduce the power, with a more than a 100% time margin reserved for reliable operation in the near-threshold region. Two upper bound systems are evaluated. A comprehensive power/area estimation indicates that the overhead of realizing SDR flexibility is insignificant. The benefit of baseband SDR is quantified and evaluated. To further augment the benefits of a flexible baseband solution and to address the security issue of IoT connectivity, a light-weight Galois Field (GF) processor is proposed. This processor enables both energy-efficient block coding and symmetric/asymmetric cryptography kernel processing for a wide range of GF sizes (2^m, m = 2, 3, ..., 233) and arbitrary irreducible polynomials. Program directed connections among primitive GF arithmetic units enable dynamically configured parallelism to efficiently perform either four-way SIMD GF operations, including multiplicative inverse, or a long bit-width GF product in a single cycle. This demonstrates the feasibility of a unified architecture to enable error correction coding flexibility and secure wireless communication in the low power IoT domain.