Sensitivity of energy-aware radio link control to imperfect average path loss knowledge

The need for higher data rates in wireless system has driven most of the recent wireless research arena. However, the increase of the system transmission rate implies higher system energy consumption. This creates a serious problem in the case of battery-powered devices such as wireless terminals. Therefore, a power management policy is necessary to dynamically trade off the system transmission rate and its energy consumption. In previous work, cross-layer energy-aware radio link control has been applied on OFDM-based WLAN transceivers. Such systems are designed for operating in indoor environment, where they can provide high throughput under low mobility conditions. Thus, the cross-layer energy-aware radio link control relies on a constant average path loss. However, in a wireless indoor environment, the average path loss can encounter significant random changes if, for instance, unpredictable object appears suddenly between the transmitter and the receiver, or simply if one or both terminal moves. The average path loss variation can reach up to 40 dB in some cases. Consequently, the power management stability can be dramatically affected. This paper aims at analyzing the sensitivity of the cross-layer energy radio link control due to such real-time average path loss variation. We also propose a more robust approach to ensure the stability of the considered radio link control strategy against random average path loss changes. From the simulation results, we have proven that the proposed radio link control approach can reduce the relative sub-optimal energy consumption per bit down to 5% compared with perfect calibration, which implies a factor 6 reduction in the sub-optimal energy consumed per bit regarding the existing radio link control.