A challenging aspect of mobile communications consists in exploring ways in
which the available run time of terminals can be maximized. In this paper,
we present a detailed electrochemical battery model and a simple stochasti
c model that captures the fundamental behavior of the battery, The stochast
ic model is then matched to the electrochemical model and used to investiga
te battery management techniques that may improve the energy efficiency of
radio communication devices. We consider an array of electrochemical cells.
Through simple scheduling algorithms, the discharge from each cell is prop
erly shaped to optimize the charge recovery mechanism, without introducing
any additional delay in supplying the required power. Then, a battery manag
ement scheme, which exploits knowledge of the cells' state of charge, is im
plemented to achieve a further improvement in the battery performance. In t
his case, the discharge demand may be delayed. Results indicate that the pr
oposed battery management techniques improve system performance no matter w
hich parameters values are chosen to characterize the cells' behavior.