Connecting sensors to LTE

The number of sensors that will be connected to the internet is expected to grow exponentially in the near future. As sensors are starting to get used for more and more applications, a large number of them will be placed in locations where wireless networks, such as Long Term Evolution (LTE), are the only available method of connectivity. Examples of such locations include remote areas (e.g. for Smartgrid and agricultural applications), and inhospitable environments (e.g. for industrial applications). Unfortunately, these wireless networks have not been designed for low power constrained devices, like battery operated sensors, and the procedures for attaching and staying connected to such networks would consume significant amounts of energy. Due to this increased power consumption the battery life on these devices would be too low to be useful. In addition to this due to the always-connected nature of the devices, the wireless network will quickly run out of resources when large numbers of sensors get connected. The resources include radio spectrum (that is extremely limited and prohibitively expensive) and signaling capacity on the network nodes. This thesis describes a new connection paradigm for connecting sensors to LTE networks along with the necessary wireless signaling changes that will allow for a battery life of the sensor to be at least 10 fold than that using the current mechanisms. This will enable the sensors to be placed in more applications where battery replacement cycles are very long and the battery replacement costs are very expensive. The new signaling mechanism will also conserve scarce resources in the wireless network so that the wireless network can scale to handle 10 fold more connections than today’s wireless networks.