Internet of Things (IoT) recently becomes a hot topic in both research communities and the public. It has already opened a level of connectivity above and beyond what we expected from computer networks. However, the technology is not matured yet, and there is a clear ongoing trend in the market. Customers start having more sophisticated requirements from the offered services. Return on investment and business profitability dominate the growth of IoT’s market share. Computing power becomes more capable in embedded systems. Serving multiple concurrent users in an IoT system is expected to be a common theme in the near future.
As a result, managing network resources effectively and efficiently is a vital key to the success for this technology. Applying solutions found in the current literature can suffer from serious issues and challenges. Some solutions have rigid assumptions that are hard to meet in dynamic IoT environments, while others are powerful but impossible to implement in systems with limited resources. To overcome these problems, this work develops a set of distributed algorithms to optimize resources effectively in IoT, and prepare the technology for upcoming advances and future directions, rather than being damaged by them.
In this work, we optimize resources in four different network layers: application, transport, routing and link layers. We consider various types of jobs and operations, including data routing and forwarding, link scheduling, sensing, data processing, energy budgeting, and other complex user-defined jobs. Our proposed solutions have been evaluated using testbed experiments and extensive network simulations. Theoretical analysis and guarantees are also provided wherever applicable.
This work is useful to researchers in the area of IoT, wireless sensor networks, embedded networked systems and general wireless networking.