Aalto University, P.O. Box 11000, FI-00076 Aalto www.aalto.fi Author M. M. Aftab Hossain Name of the doctoral dissertation Energy Efficiency of Wireless Access Impact of Power Amplifiers and Load Variations Publisher School of Electrical Engineering Unit Department of Communications and Networking Series Aalto University publication series DOCTORAL DISSERTATIONS 117/2016 Field of research Communications Engineering Manuscript submitted 18 February 2016 Date of the defence 22 June 2016 Permission to publish granted (date) 19 May 2016 Language English Monograph Article dissertation Essay dissertation Abstract In order to ensure seamless coverage and sustainable exponential growth of capacity, there is a keen interest in the development and deployment of highly energy efficient wireless systems and solutions. To design an energy efficient network, it is important to consider the facts that the variations in traffic demand in both temporal and spatial domains are significant, and the power consumption of cellular networks is mostly dominated by macro base stations where the power amplifiers (PAs) consume around 55-70 percent of total energy. One of the main challenges lies in coping with this load variations considering that the PAs attain high efficiency only at around the maximum output power level. In this thesis, we propose energy efficient system level solutions for wireless access network that consider the non-ideal efficiency characteristics of the PA and the load variations. We model and incorporate the PA efficiency in the energy-delay trade-off present in Shannon's channel capacity model in order to investigate the energy saving potential in a wireless access network at the cost of additional flow-level delay. We propose a best response iteration based distributed power control algorithm where the cells identify the power levels for different user locations to minimize energy consumption under delay constraints. We observe that energy saving potential strongly depends on the network load and PA efficiency characteristics. We also investigate the impact of additional delay in the downlink on the energy consumption of the mobile terminal. Heterogeneous network is the leading technology for the next-generation cellular networks. We investigate the energy-efficient densification and load sharing between the layers of a heterogeneous network while taking into consideration the PA efficiency and temporal load variations (TLV). We also study the impact of PA efficiency and energy-delay trade-off on the energy efficient network densification. Massive MIMO (MM) is another leading candidate technology to cater for very high capacity demand. We consider a multi-cell MM system and provide the guidelines to dimension the PA for the antennas. We also develop energy efficient antenna adaptation schemes that allow the cells to dynamically adapt the number of antennas to the TLV in order to maintain high energy efficiency (EE) throughout the day. Our results indicate that these proposed antenna adaptation schemes can improve the EE significantly.

Energy Efficiency of Wireless Access - Impact of Power Amplifiers and Load Variations

The phenomenal progress of computing devices has been largely made possible by the sustained efforts of semiconductor industry in innovating techniques for extremely large scale integration. Indeed, gigantically-integrated-circuits today contain multi-billion interconnects which enable the transistors to talk to each other ? all in a space of few mm2. Such aggressively downscaled components (transistors and interconnects) silently suffer from increasing electric fields and impurities/defects during manufacturing. Compounded by the Gigahertz switching, the challenges of reliability and design integrity remains very much alive for chip designers, with Electromigration (EM) being the foremost interconnect reliability challenge.
Traditionally, EM containment revolves around EM guidelines, generated at single-component level, whose non-compliance means that the component fails. Failure usually refers to deformation due to EM ? manifested in form of resistance increase, which is unacceptable from circuit performance point of view. Subsequent aspects deal with correct-by-construct design of the chip followed by the signoff-verification of EM reliability. Interestingly, chip designs today have reached a dilemma point of reduced margin between the actual and reliably-allowed current densities, versus, comparatively scarce system-failures. Consequently, this research is focused on improved algorithms and methodologies for interconnect reliability analysis enabling accurate and design-specific interpretation of EM events. 
In the first part, we present a new methodology for logic-IP (cell) internal EM verification: an inadequately attended area in the literature. Our SPICE-correlated model helps in evaluating the cell lifetime under any arbitrary reliability speciations, without generating additional data ? unlike the traditional approaches. The model is apt for today's fab less eco-system, where there is a) increasing reuse of standard cells optimized for one market condition to another (e.g., wireless to automotive), as well as b) increasing 3rd party content on the chip requiring a rigorous sign-off. We present results from a 28nm production setup, demonstrating significant violations relaxation and flexibility to allow runtime level reliability retargeting.
Subsequently, we focus on an important aspect of connecting the individual component-level failures to that of the system failure. We note that existing EM methodologies are based on serial reliability assumption, which deems the entire system to fail as soon as the first component in the system fails. With a highly redundant circuit topology ? that of a clock grid ? in perspective, we present algorithms for EM assessment, which allow us to incorporate and quantify the benefit from system redundancies. With the skew metric of clock-grid as a failure criterion, we demonstrate that unless such incorporations are done, chip lifetimes are underestimated by over 2x.
This component-to-system reliability bridge is further extended through an extreme order statistics based approach, wherein, we demonstrate that system failures can be approximated by an asymptotic kth-component failure model, otherwise requiring costly Monte Carlo simulations. Using such approach, we can efficiently predict a system-criterion based time to failure within existing EDA frameworks.
The last part of the research is related to incorporating the impact of global/local process variation on current densities as well as fundamental physical factors on EM. Through Hermite polynomial chaos based approach, we arrive at novel variations-aware current density models, which demonstrate significant margins (> 30%) in EM lifetime when compared with the traditional worst case approach.The above research problems have been motivated by the decade-long work experience of the author dealing with reliability issues in industrial SoCs, first at Texas Instruments and later at Qualcomm.

Algorithms and methodologies for lnterconnect reliability analysis of lntegrated circuits

Benabou (2002) strengthens the so-called Efficient Redistribution Hypothesis (ERH) by demonstrating how income redistribution can promote growth and welfare by mitigating economic waste from resource misallocation that is caused by credit market frictions to production, which is subject to diminishing returns. We ask how the presence of knowledge externality in private education technology and social frictions resisting knowledge diffusion changes the optimal progressivity. In the model, the distribution of human capital of the parent's generation determines the stock of nonrival knowledge through a process of diffusion which is subject to communication frictions. The output cost of inequality and the optimal progressivity increase with the degree of knowledge externality and hence with social returns to education, if and only if, frictions to diffusion outweigh frictions to production. If the inequality augmenting neighborhood effect of parental human capital is relatively large, compared to the inequality mitigating impact of the economy's stock of nonrival knowledge, then the ERH fails with low communication frictions. However, if frictions and social returns to education are large, irrespective of its local or global origin, then even with endogenous growth led by constant returns to capital, the ERH survives.

Optimal progressivity of redistribution with frictions to knowledge diffusion

Schwartz & Yeh's method and Wilkinson's method are widely used to compute the moments of the total co-channel interference in wireless communi- cation, usually modeled as the sum of lognormal ran- dom variables. The accuracy of these methods has been studied in previous works, under the assump- tion of having all summands signals (individual inter- ference signals) identically distributed. Such assump- tion rarely holds in practical cases of emerging wireless systems, where interference may stem from far-away macrocells and nearby transmitters, causing the inter- ference signals to have different moments. In this pa- per we present an analysis of Wilkinson's method and Schwartz & Yeh's method, for the general case when the summands have different mean values and standard deviations in decibel units. We show that Schwartz & Yeh's method provides better accuracy than Wilkin- son's method and is virtually invariant with the differ- ence of the mean values and standard deviations of the summands, and the number of summands.

Communications Paulo Cardieri

Papers

Statistics of the sum of lognormal variables in wireless