A Tight Channel-Capacity Lower Bound for the Simultaneous Wireless Information and Power Transfer Integrated Receiver

Konstantinos Ntontin and Symeon Chatzinotas · 2026

Read A Tight Channel-Capacity Lower Bound for the Simultaneous Wireless Information and Power Transfer Integrated Receiver by Konstantinos Ntontin and Symeon Chatzinotas (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2604-18986.txt.

Introduction

Ntontin and Chatzinotas derive a tight channel capacity lower bound for the integrated receiver in simultaneous wireless information and power transfer. Most prior work treats the separate receiver; this work targets the integrated design. They approximate the channel transition matrix in closed form using a fourth order Taylor expansion of the Schottky diode current voltage curve. Numerical tests show a gamma input distribution yields a tight bound, unlike Rayleigh or uniform inputs. Keeping the fourth order term notably beats the simpler second order model.

Catalog record

CREATORKonstantinos Ntontin and Symeon Chatzinotas
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE21161
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2604.18986

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

A Tight Channel-Capacity Lower Bound for the Simultaneous Wireless Information and Power Transfer Integrated Receiver Konstantinos Ntontin and Symeon Chatzinotas Abstract Contrary to the vast majority of works on simultaneous wireless information and power transfer that provide information-theoretic limits for the separate receiver architecture, in this work we focus on the integrated receiver and provide a channel-capacity lower bound. Towards this, we provide a closed-form tight approximation for the probability transition matrix of the channel by leveraging the 4th-order Taylor expansion of the current-voltage characteristic curve of a Schottky diode used for rectification. Numerical results reveal that the consideration of the gamma distribution as an input distribution leads to a tight channel-capacity lower bound, in contrast to other input distributions, such as the Rayleigh and uniform ones. Furthermore, the results reveal that the consideration of the 4th order term in the Taylor expansion leads to a notably higher capacity with respect to the overly simplified 2nd order term-based model. I Introduction I-A Background Over the last 15 years, there has been a large body of works that provide design guidelines for simultaneous wireless information and power transfer (SWIPT). Typical applications concern low-power Internet of Things (IoT) devices that need to simultaneously be energized and provided with information data that could for instance be used for actuation. The vast majority of works have considered a SWIPT receiver architecture that dedicates different receivers for radio frequency (RF)-to-direct current (DC) power conversion and information detection. Such an architecture in the power splitting case 1 1 1 A similar architecture also holds for the time-splitting case. is depicted in Fig. 1 [ 1 ] . n_{th}(t) is the antenna thermal noise, G_{LNA} is the gain of the low-noise amplification (LNA) stage, and \rho is the power splitting ratio. Fig. 1: SWIPT architecture with separate information and energy receivers. The components of each of the 2 receivers are depicted in Fig. 2. n_{rec} and n_{mix} are the noise random processes introduced by the rectifier and the mixer, in the energy and information receivers, respectively. With the separate receiver architecture the energy and communication channel paths are separated, which means that communication channel is the well-known linear and memoryless additive white Gaussian noise (AWGN) channel. Hence, according to information theory, the rate maximization of the communication channel in the separate receiver SWIPT architecture can be achieved by a circular symmetric complex Gaussian (CSCG) input distribution. [ 1 , 2 , 3 ] . In fact, it has been proved that the rate-energy bound of the ideal receiver that supplies the energy and information paths without power splitting can be achieved by the power-splitting architecture of Fig. 1 [ 1 ] . This is possible through a superposition of a waveform that is optimal for wireless power transfer with a waveform that is optimal for information transmission [ 4 ] . Fig. 2: Components of the SWIPT separate receiver architecture: a) Energy receiver; b) Information receiver. Despite the advantage of the separate SWIPT receiver in having disjoint communication and energy harvesting paths, its main shortcoming originates from the need for the mixing operation for downconversion that involves a local oscillator (LO). Although the technology has been notably progressing over the years and the power consumptionof LOs has been substantially reduced, it is still in the order of hundreds of \mu Ws for typical sub-6 GHz 4G/5G bands [ 5 ] . Such figures make it very challenging to accommodate the LO energy needs through the harvested energy from wireless power transfer. This is why several types of Ambient-IoT devices include passive envelope detectors, where only the amplitude of the signal is detected and its phase is sacrificed for achieving very low complexity [ 6 ] . The envelope detection inherently downconverts the RF signal to baseband and can be achieved by the same rectification circuitry that harvests energy. This is the motivation behind the SWIPT integrated receiver design, which is depicted in Fig. 3 [ 7 ] . y_{DC}(t) is the DC signal at the output of the rectifier and ADC is the analog-to-digital converter. As it can be observed from Fig. 3, in the integrated receiver there is a common channel path for the energy harvesting and information detection processes. Fig. 3: SWIPT integrated receiver. The fact that in the SWIPT integrated receiver the output of the rectifier not only supplies the battery, but it is also the information signal to be decoded, creates fundamental differences with respect to the well-studied separate receiver architecture of Fig. 1. More specifically, the communication channel becomes non-linear, due to the diode used for rectification, and also subject to memory effects due to the low-pass filter. Regarding the former, a diode, normally a Schottky one, is highly non-linear circuit, with a characteristic i_{d}(t)-v_{d}(t) response that is given by i_{d}(t)=i_{s}(e^{\frac{v_{d}(t)}{nv_{t}}}-1). (1) where v_{d}(t) is the voltage drop across the diode and i_{d}(t) is the current that flows through the diode. Moreover, i_{s} , v_{t} , n , are the diode reverse bias saturation current, the thermal voltage, and the ideality factor, respectively. Through a Taylor approximation of i_{d}(t) , up to the 4th order term, the DC signal y_{DC} is in the form [ 1 ] \displaystyle y_{DC} \displaystyle\approx k_{2}|\sqrt{{G_{LNA}}{PL}}x+\sqrt{G_{LNA}}n_{th}|^{2} (2) \displaystyle+k_{4}|\sqrt{{G_{LNA}}{PL}}x+\sqrt{G_{LNA}}n_{th}|^{4}+n_{rec}, where x is the transmitted signal, PL is the path loss, and k_{2} and k_{4} are constants 2 2 2 The noise introduced by the ADC is not considered since its contribution is expected to be notably lower than n_{rec} due to the LNA stage at the output of the receive antenna and a potentially another LNA stage before the ADC input, which is a common practice. . The literature is very limited with respect to works on the SWIPT integrated receiver [ 7 , 8 , 9 , 10 , 11 ] . In particular, [ 7 ] provides the rate-energy characteristic curve, but only the 2nd-order Taylor expansion term is used for approximating y_{DC} . Such a rate-energy tradeoff is analytically computed also in [ 8 ] but again only the 2nd-order Taylor expansion term is considered for y_{DC} . [ 9 ] proposes a novel modulation and demodulation method for the SWIPT integrated receiver based on pulse position modulation, where information is encoded in the position of the pulse. The error rate performance and harvested DC power are numerically evaluated by considering the 4th- order Taylor expansion for y_{DC} . Furthermore, for a pulse position modulation the authors in [ 10 ] optimize the constellation design and show performance improvements with respect to a uniform design. Similar to [ 7 ] and [ 8 ] , the 2nd-order Taylor expansion term is considered for y_{DC} . Finally, the authors in [ 11 ] apply circuit analysis to obtain a model that captures the memory effects of the low-pass filter. They then focus on the error rate and apply maximum likelihood sequential detection for reducing it. I-B Motivation and Contribution I-B 1 Motivation The maximum information rate that can be transmitted over the communication channel given by (2) is unknown. The mentioned literature works that study the information-theoretic limits of the SWIPT integrated receiver consider the simplistic 2nd-order Taylor expansion for y_{DC} , which means that y_{DC}\approx

Read the complete text