Optimizing energy efficiency of a multi-radio mobile device in heterogeneous beyond-4G networks

Abstract

In this paper, we address the operation of a multi-radio mobile device in heterogeneous wireless deployments. We assume that such a device may efficiently control its radio interfaces when using the available radio access technologies. In particular, we investigate the potential of flexible transmit power allocation and develop a provably optimal power control scheme that strictly maximizes the energy efficiency of the mobile device, while at the same time satisfies the minimum required level of the user data rate. When compared against simpler (heuristic) power control strategies, our solution always demonstrates the best energy efficiency of the multi-radio device by enabling collaborative operation between several radio technologies, which makes it a useful benchmark for the future integrated beyond-4G wireless networks.

Introduction

Wireless networks demonstrate worldwide proliferation, which has further advanced recently with the introduction of novel fourth generation (4G) communication technologies  [1], [2]. Adoption of these 4G technologies is becoming increasingly widespread, allowing for improved access to services and applications previously only supported through fixed broadband systems  [3]. However, existing wireless deployments are still unable to deliver their users the desired ubiquitous connectivity experience due to the shortage of available capacity and lack of service uniformity  [4].

Whereas there is currently no technical definition of what comes after the state-of-the-art 4G technologies, experts agree on that future beyond-4G wireless communications will probably be a converged set of co-existing radio access networks, rather than one single technology  [5]. As wireless spectrum continues to be scarce and expensive, the success of future beyond-4G systems requires effective solutions to overcome the divide between the demanded quality-of-service (QoS) and the limited network resources.

Over the years, wireless spectrum has become one of the most valuable natural resources, which accentuates the importance of its efficient use (i.e., spectral efficiency). However, energy efficiency is also becoming increasingly important primarily for small form-factor mobile devices, where wireless power consumption dominates the total device power budget. This is due to the increasing disproportion between the available and the required battery capacity, which is demanded by the ubiquitous multimedia applications  [6]. To compensate for this growing gap, aggressive improvements in all aspects of wireless system design are necessary  [7].

Whereas energy efficiency is accentuated by the need of extending client device operation time without recharging, the need for improved service continuity is dictated by the ubiquitous wireless multimedia applications. Currently, wireless cellular, local, and personal area networking technologies as well as supportive network architectures are evolving towards more advanced and complex converged networks. On the other hand, consumer electronics is spawning a huge explosion in the number and variety of multi-radio devices  [8], driven by the user demand for “anytime, anywhere” connectivity.

The problem of energy efficient interworking between the available wireless technologies in a user multi-radio device is therefore addressed in this work in order to develop provably efficient techniques that allow for significant energy performance improvement in heterogeneous wireless environments.

Conventional wireless devices are typically communicating their data by choosing one of the fixed set of modulation and coding schemes, which sacrifices flexible power adaptation for design simplicity  [9]. This often causes excessive energy consumption or pessimistic data rates selected for peak channel conditions  [10]. Hence, physical layer parameters should be flexibly adjusted to actually account for the client QoS requirements as well as for the state of the wireless channel to reach a compromise between energy and spectral efficiencies  [11]. In this regard, throughput optimization has long been an attractive research direction  [12], [13]. However, as wireless clients become increasingly mobile, the focus of recent efforts tends to shift towards energy consumption at all layers of communication systems, from architectures  [14] to algorithms  [15].

Energy efficiency is becoming increasingly important for wireless networks due to the limited battery lifetime of mobile clients. For maximizing energy efficiency, so-called “bits-per-Joule”  [16] or “throughput-per-Joule”  [17] metrics are often considered. Several approaches are known to focus on energy efficiency. These include water-filling power allocation techniques that optimize throughput with respect to the fixed total transmit power limitation  [18], [19], as well as adaptation of both the total transmit power and its allocation according to the channel state information  [10], [20].

However, the vast majority of existing information-theoretic approaches (see, e.g.,  [21], [22]) account only for the transmit power when investigating energy consumption. Typically, a client device will also consume extra circuit power, which is independent of its data rate  [23], [24] and can actually be on the same order (and even comparable) with the maximum allowed transmit power. As such, the circuit power consumption should be considered explicitly when optimizing energy efficiency  [25]. With the emphasis on circuit power, recent work suggests the use of optimization theory for establishing energy-optimal communication settings  [10], [26] to balance transmit and circuit power consumption. These findings indicate that the conventional water-filling approach to simply extend the transmission time of the device may not be attractive anymore since circuit energy consumption grows with transmission duration.

With the growing use of smaller cells to improve the capacity of 4G systems, the coverage ranges of cellular, local, and personal area networks are increasingly overlapping. In the extreme, contemporary urban wireless deployments often include areas where different communication networks are co-located  [27], [28]. As long as these technologies occupy non-overlapping frequency bands  [28], they may coexist simultaneously without any significant performance degradation. This creates an attractive opportunity to cooperatively utilize several radio access networks for improved wireless connectivity  [29].

Over the last few years, much literature has accumulated  [30] exploring the interworking solutions within the core network and above, including seamless mobility between 3GPP and WLAN technologies, trusted access to 3GPP services with WLAN devices, and support for Access Network Discovery and Selection functions  [31]. In particular, the network selection problem in heterogeneous wireless environment using IEEE 802.21 and IEEE 1900.4 frameworks has recently been studied  [32]. We emphasize that our focus in this paper is, however, on the joint use of multiple networks that requires cooperation on the Radio Access Network (RAN) layer, which enables more flexible control of the transmission parameters  [33]. We expect this work to be useful in the ongoing 3GPP discussions on WLAN/3GPP radio interworking  [34].

We expect that intelligent coupling between multiple radio access technologies (such as LTE-Advanced, HSPA, WiMAX, WiFi, Bluetooth, ZigBee, etc.) will enable efficient operation of a multi-radio device and thus realize the desired uniform user experience. To achieve this, both short- and long-range technologies (e.g., WiFi and LTE-Advanced) may need to work cooperatively to augment system capacity and improve service continuity  [35] in a beyond-4G network (see Fig. 1). Consequently, we seek to explore the potential of adaptive power control to improve the energy efficiency of a multi-radio device using heterogeneous connectivity at different scales.

The rest of this text is organized as follows. In Section  2, we introduce our system model with its main assumptions. Section  3 formulates a practical constrained optimization problem, where the energy efficiency of a multi-radio mobile device needs to be maximized subject to some realistic restrictions. We then solve this problem directly and obtain the exact solution. In Section  4, we provide several important numerical examples to conclude on the feasibility of our approach.