As the daily usage time of smart phone increases, battery life time becomes increasingly important. Recently, a complaint is raised for frequent battery drain in smart phones due to applications running in the background. In most of the mobile applications, after end of the data transfer, longer idle time occurs in a Transmission Control Protocol (TCP) connection for connection closure. Usually such idle TCP connections are terminated by servers based on a static timeout value configured at the server side. This results in frequent transition from idle Radio Frequency (RF) state to active RF state thereby increasing mobile network signalling and device battery power consumption. To address this problem, we propose a novel TCP connection control mechanism in view of power saving, called Low Power TCP (LPTCP). Our solution is lightweight, application agnostic, a client deployed solution. LPTCP identifies inactive TCP connections that cause the delayed server close and pro-actively closes those TCP connections. We evaluated LPTCP with the latest Samsung smart phones such as Galaxy S8 and S9. LPTCP achieved up to 18% of power gain and significantly reduced 24% of mobile network signalling overhead.

Energy awareness of mobile devices with limited battery capacity can be achieved by embedding battery drain rate monitoring capability into the devices. With Li-ion battery, battery drain rate varies with temperature and battery aging, since they affect battery characteristics such as capacity and internal resistance. We develop BattTracker, an algorithm to estimate battery drain rate without knowing the exact capacity and internal resistance by incorporating the concept of effective resistance. BattTracker tracks the instantaneous battery drain rate with up to 0.5 second time granularity. Extensive evaluation with smartphones demonstrates that BattTracker accurately estimates the battery drain rate with less than 5% estimation error, thus enabling energy-aware operation of smartphones with finegrained time granularity. To the best of our knowledge, this is the first effort to estimate the instantaneous battery drain rate by considering both temperature and aging effects on the battery characteristics.

Voice over LTE (VoLTE), adopted as a standard technology by 3GPP is confronted by a number of challenges due to different Quality-of-Service (QoS) requirements as well as the Inter-Cell Interference (ICI) problem. In this paper, a novel Resource Allocation (RA) strategy for improving VoIP capacity in LTE uplink system is proposed. Our strategy improves the VoLTE performance simply by prioritizing packets according to Signal to Generating Interference plus Noise Ratio (SGINR) based metric, and at the same time, allocating the ordered packets to Resource Blocks (RBs) in a predefined way, which protects vulnerable users with low uplink received signal strength (RSS) and separates threatening users in each cell into different RBs in an orthogonal manner. Through system level simulation, we demonstrate that our resource allocation scheme outperforms commonly-used frequency reuse-based schemes by increasing the VoIP capacity by up to 20%.

We propose E$PA, an adaptive LTE/WiFi network interface activation algorithm with supporting system design and multi-attribute cost function for smartphones file transfer services (e.g., downloading a movie file). E$PA's cost function incorporates battery life (energy), data usage quota ($), and expected file transfer completion time (performance) simultaneously, and is motivated by the growing sensitivity of todays smartphone users to these attributes. Each time the individual attributes are calculated and updated, E$PA selects one of the three modes that minimizes the overall cost: (i) activation of both LTE and WiFi interfaces for parallel data transfer; (ii) LTE interface activation only; or (iii) WiFi interface activation only. The primary benefit of the E$PA is that it enables the smartphone to always operate in the “best” mode without the need for users manual control; the energy saving mode if the remaining battery energy is becoming nearly depleted; the cost-saving mode if the remaining data quota is almost running out; or, the maximum throughput mode if remaining data quota and battery life are both sufficient. Our multi-attribute cost model also takes into account the overheads (delays and energy consumption) associated with network interface turn-on/off and switching, as they impact the estimations of both performance (transfer time) and energy (battery life) attributes. Simulation results show that E$PA indeed chooses the “best” operational mode by maintaining dynamic balance among transfer time, energy consumption, and service charge.

Self-organizing network (SON), which is an essential technology of future radio networks, is proposed in the 3GPP Long Term Evolution (LTE) specification as a usage case. In this paper, we focus on user equipment (UE)'s power control considering interference mitigation and retransmission constraints especially for physical random access channel (PRACH) of LTE. When a UE initially accesses an evolved-NodeB (eNB), it sends a preamble with an initial access level, and raises the power by a fixed step whenever the preamble was not detected by the eNB. We present an analysis and simulation to study the tradeoff relationship between inter-cell interference and retransmission according to the PRACH power setting, i.e., the initial target received power and power ramping step, assuming Rayleigh fading channels. Furthermore, we propose an algorithm tracing the optimal power setting that minimizes the resource wastage of the cell in consideration and its adjacent cells satisfying the constraint of retransmission and failure probability bound.