Next-Generation IoT Networks: Integrated Sensing Communication and Computation

Abstract

To enable the exponential expansion of Internet of Things (IoT) applications, IoT devices must gather and transmit massive amounts of data to the server for further processing. By employing the same signals for both radar sensing and data transmission, the integrated sensing and communication (ISAC) approach provides simultaneous data gathering and delivery in the physical layer. Over-the-air computation (AirComp), which leverages the analog-wave addition property in multi-access channels, is a communication method that also supports function computation. In order to leverage the individual benefits of ISAC and AirComp, this work focuses on Integrated Sensing Communication and Computation (ISCCO) framework for the IoT network. Since the IoT sensors are small size low cost devices and each is equipped with single antenna, and hence to make the processing of received echo simple this work assume that the waveform transmitted by each sensor is orthogonal to each other. Furthermore, joint optimal power allocation for each sensor in the IoT network and the combining vector at the EC is designed such that the signal-to-noise (SNR) ratio at the EC is maximized. However, the design challenge lies in the non-convex joint optimal power allocation for each IoT device and the combining vector at the server. To address this, an iterative algorithm is proposed which provides closed-form solution for each quantity in each iteration. Results show that the proposed optimal power allocation and orthogonal waveform design scheme outperforms the equal power allocation-based design.