Wireless Communications and Signal Processing for the Internet of Things

点击翻译按钮获取中文摘要

Reliable high-rate wireless and mobile personal communications of multimedia has been largely achieved. Its success is due to physical signal layer improvements, Internet packet (IP) switching combined with channel sensing, and multiple access protocols with re-transmission of packet errors. The future is faced with new challenges motivated by automation in manufacturing, smart city infrastructure, transportation systems, e-health, environmental monitoring, and energy efficiency demands. Technical challenges standing in the way include communication among ever-larger numbers of devices rather than among people, low latency requirements rather than best effort service, and dense networks rather than minimum coverage achievement. These issues are starting to be addressed worldwide by fifth-generation (5G) standardization efforts targeted for year 2020. By necessity, the aggressive 5G time frame is narrowly focused. However, the above-mentioned future challenges are longer term, much broader and require comprehensive solutions. The proposed research program addresses several of these challenges by exploring the following themes: (i) increasing simultaneous communication capacity with low power devices well beyond the number of base station antennas, (ii) low delay event detection and change detection monitoring from a distributed sensor network with restricted communication, energy cost, error rate, failure rate, and resilience to malicious users, and (iii) multiple access protocols for dense networks to associate groups of devices to multi-antenna access points that have performance optimization in terms of throughput and fairness. To delineate our proposed research, we note first that communication with low-rate power-limited devices is not well modeled by conventional signalling assumptions used in classical information theory by Shannon nor is well served by classical maximum signal-to-noise ratio beamforming criteria. The large numbers of devices require signaling dimensions that far exceed the number of available antennas. Second, unlike classical Shannon principals, communication latency (or delay) is constrained severely so as to enable new applications, e.g., real-time remote control of complex networked systems, driver-less vehicular networks, and operation of networked robots. In addition, sensor networks will increasingly require monitoring tasks that transmit decisions with reliability, speed and security, and solutions are needed to address protocol overhead. Third, networks face formidable challenges in scaling of size and density. Resource allocation in dense networks faces computational explosion. Current protocols that enable dynamic spectrum resource sharing, while effective for current systems, are limited in the long term due to the many possible groupings of devices. The proposed research program will train 4 PhD, 4 MASc, 5 M.Eng. and 3 undergraduate students and enable future defense and industry related collaborations.