Low Power Wide Area Networks (lpwan): Technology Review And Experimental Study on Mobility Effect

LESSONS LEARNED

In this section we discuss the various scenarios in which results of this research can play an important role during the design phase of the project. This section we talk about three
LPWAN applications and show how the results of this research can be used in designing LPWAN applications.

1. AGRICULTURE LIVESTOCK TRACKING

New technological innovations and ideas can help agriculture industry to better manage the available resources and also results an increase in yields and profit. LPWAN application in the field of agriculture can perform various tasks like:

• 
  Remotely controlling self-driving tractors can help to reduce cost [14].
  
• 
  Sensors attached to livestock can help the farmers keep tracks of their location and health. Sick animals can be separated from the heard thus stopping any further spread of infection to other animals [17].
  
• 
  Sensors deployed to track the health of soil can notify farmers of various irregularities like if the soil becomes to acidic or too dry for farming [17].
  

Based on our results on performance of LPWAN in outdoor environment, here are some considerations needed to be assessed before trying to design a LPWAN based
agricultural application.

• 
  Type of sensors and location : Different sensors are used to gather different data and they also vary in sizes. It is no easy for farmers to mount a big sensor collar onto the animals. If the animals being tracked are in an area with dense vegetation or if the population of the animals is high, it might degrade the performance on the
  

application.

• 
  How often is data gathered : It is not necessarily true that if more data is gathered by a sensor the better it is. More data often means a larger data packet size or frequent transmissions. It is seen from our experiments that a larger data packet size has considerable impact on mobility performance of the sensors.
  
• 
  Range Needed : It has been concluded from our experiments that more distance the data packet needs to travel the more is the impact of mobility on network performance.
  
• 
  Type of Livestock being tracked : Different animals move at different speed. For example sensors tracking a horse would often be moving at a higher speed than a sensor tracking a cow. The former network will suffer from the impact of mobility on network delay and packet loss rate as discussed in our experiments.
  
• 
  Gateway Location : It is also crucial to consider how the gateways that communicate with the end nodes are deployed in the field. It has been seen from our experiments that a gateway places inside a concrete building does suffer from interference due to walls.
  
• 
  Geographical Location : Application performance can also be affected if the area in which it is deployed has dense vegetation cover like big trees or small hills which can interfere with network performance.
  

2.