Wireless Transimission Rate Scheduling For Energy Harvesting Systems

Since wireless devices are becoming more and more popular, their energy shortage problem becomes more and more serious, and thus attracts a lot attentions. Energy harvesting technique and transmission rate scheduling are thus proposed to solve this problem from’earning energy’ and ’saving energy’ aspects respectively. Energy harvesting technique collects energy from sur-rounding natural environments which provides extra energy, and is a promising solution to the energy shortage problem. Wireless transmission rate scheduling changes the rate during trans-mission, either to reduce energy consumption or to reduce transmission latency. For energy harvesting wireless devices, the general goals of rate scheduling are short transmission delay, low energy consumption and high throughput. However, short transmission delay means high transmission rate, which results in high energy consumption, vice versa. Therefore, the gen-eral goals can not be achieved simultaneously. How to balance the three goals on an energy harvesting wireless device becomes an urgent problem.Existing research works on rate scheduling of energy harvesting wireless devices mainly fo-cus on maximizing throughput, which have several drawbacks. First, a wireless communication channel is usually shared by multiple applications, hence packets have individual transmission delay constraints. However, the goal of maximizing throughput can not guarantee individual delay constraints to be satisfied. Second, although individual delay constraints are already con-sidered by rate scheduling problem on battery powered wireless devices. No existing algorithm can handle the case when a latter arrived packet carries a more urgent delay constraint. Finally, in mobile transmission, the transmission distance changes over time, hence it is rather challenging to design a rate schedule that maximize throughput. An existing work gives an approximation algorithm, whether or not an optimal algorithm exists remains unknown.To deal with the above drawbacks, this thesis proposes and solves the following three prob-lems. First, in the rate scheduling for energy harvesting wireless devices, the transmission de-lay constraints are introduced. The essence behind is to minimize energy consumption while given transmission amount and allowed delay. This thesis propose the Truncation method that truncates high rate and computes the optimal rate schedule by iteration. Second, for the rate scheduling problem of battery powered wireless devices, whose essence is to minimize energy consumption while given transmission data amount and delay constraints, but without energy harvesting, this thesis presents the Densest Interval First (DIF) policy which repeatedly locates the densest data interval and determines its optimal transmission rate. Third, in throughput max-imization problem for mobile energy harvesting wireless devices, the essence is to maximize throughput while given available energy. This thesis designs the WaterTanks method together with dynamic programming to determine slot assignments and power controls, and obtain the optimal solution. Finally, this thesis develops a wireless transmission rate scheduling tool pro-totype. It implements all proposed algorithms, and provides users unified interface as well as standard dynamic link library.This thesis solves the energy shortage problem by the ’earning’ and ’saving’ methods. The proposed rate scheduling algorithms for energy harvesting wireless devices are balanced amongst the general goals of short transmission delay, low energy consumption and high throughput. Because of the popularity of wireless devices, our proposed algorithms can play an important role in people’s lives.