Nowadays, various emerging applications, services, and user requirements require rethinking the future multiple access technique of 6G. First, 6G should fully enable high mobility, using high frequency bands with large bandwidths, thanks to a strong resilience to high Doppler spreads. Second, 6G should offer massive IoT-enabled applications, with reduced frequency synchronization and power control signaling overhead, thanks to a strong resilience to out-of-band emissions, time and frequency synchronization errors and received powers imbalance.
While traditional CP-OFDM, on which 4G and 5G are based, is still suitable for many future 6G use cases, it will suffer from severe self and multi-user interference, in such high Doppler spread and out-of-band emissions use cases. Therein lies the need for technological advancements, in terms of multiple access and, more specifically, on waveform design, to unlock the true value of 6G connectivity.
The need for a strong resilience to self and multi-user interference brought several proposals for new multiple access techniques, over the last decade. Among the recently published proposals are Orthogonal Time-Frequency Space (OTFS), Sparse Code Multiple Access (SCMA), and Non-Orthogonal Waveform (NOW). Moreover, given the ascending role to be played by Machine Learning (ML) in 6G, a proposal considered learning waveform design, with ML not only replacing the processing blocks, but also the physical layer and underlying waveform and modulation.
In this talk, will be concerned with the design of new waveforms, operating on the same time-frequency layout of 4G and 5G, as potential substitutes of CP-OFDM rectangular waveforms. While offering strong resilience to self and multiple-access interference, these waveforms have the merits of easily fusing with 4G and 5G technologies and enabling 6G to be rolled out faster.
We will start by a historical overview of the Ping-pong Optimized Pulse Shaping (POPS) algorithm, and its efficient application to different waveform optimization problems, within the single-access framework, whereby a single user makes use of the whole available bandwidth. We will examine the effectiveness of several time-frequency layouts (rectangular and hexagonal) and modulations (QAM and OQAM), in the light of best achieved Signal-to-Noise Ratio (SINR), through POPS waveform optimization. We will show that the additional gains in performance, in terms of SINR, brought hexagonal lattices and OQAM are sufficiently low and do not justify the complexity they entail.
We then move to waveform design in a multiple-access context, whereby different communications share the available frequency bandwidth. We will consider several radio interface impairments, such as time-frequency synchronization errors and received power imbalance, at the receiver, in both uplink and downlink, and time asynchronism at the transmitters, in uplink. We will introduce the notion of equivalent SINR, use the POPS algorithm for waveform optimization and the CP-OFDM as benchmark. We will show the effectiveness of the designed waveforms, with respect to CP-OFDM, in terms of resilience to self and multiple access interference, especially in the uplink, when full asynchronism is assumed.

Network-in-a-box (NIB) is one of envisaged enablers to eliminate the connectivity inequality and to enhance network resilience for ubiquitous networking in 6G. NIB suits for supporting one-off networks such as access on demand, emergency access, and traffic off-loading. The core idea of NIB is to pack the whole wireless network into a few or a single portable and reconfigurable box supporting multiple wireless standards for ease and speed of deployment. The advantages of NIB let us question how NIB can be used to sustain ubiquitous network connectivity in 6G. In this talk, the advantages and challenges of NIB will be summarized and NIB optimal deployment problem will be addressed as a case study.

We are witnessing a space renaissance. Tens of thousands of broadband low Earth orbit (LEO) satellites are expected to be launched by the end of this decade. These planned megaconstellations of LEO satellites along with existing constellations will shower the Earth with a plethora of signals of opportunity, diverse in frequency and direction. These signals could be exploited for navigation in the inevitable event that global navigation satellite systems (GNSS) signals (e.g., GPS) become unavailable (e.g., in deep urban canyons, under dense foliage, during unintentional interference, and intentional jamming) or untrustworthy (e.g., under malicious spoofing attacks).

This talk will present a framework, termed STAN: simultaneous tracking and navigation, for exploiting signals of opportunity from LEO megaconstellations, which are not intended as navigation sources. In this framework, specialized cognitive radios draw relevant positioning and timing information from unknown downlink signals of LEO satellites with poorly known ephemerides to build and continuously refine a spatiotemporal signal landscape map of the environment within which the radios simultaneously localize themselves in space and time. We will present an end-to-end research approach, spanning theoretical modeling and analysis of signals of opportunity, specialized cognitive software-defined radio (SDR) design, practical navigation algorithm development, and experimental demonstration of our system on ground and aerial vehicles, navigating with multi-constellation LEO satellite signals (Starlink, Orbcomm, and Iridium) to an unprecedented level of accuracy.

Mm-wave and massive MIMO technologies are some of the elements that have enabled the increase of the communication capacity and reliability, and the reduction of the latency in 5G systems. These technologies have also made possible the advent of new positioning solutions characterized by very high accuracy, provision of orientation information, and inclusion of sensing capabilities, with reduced network-side infrastructure. The term “5G Localization” was coined to encompass this class of solutions, which have already impacted the standardization process. Elements considered in the ongoing research towards 6G systems, such as reconfigurable intelligent surfaces (RIS), are also showing large potential for localization, as they extend the position availability to problems that were previously unfeasible. First, this talk will briefly present the foundations of radio-based positioning and an overview of the evolution of localization in cellular standards. Second, it will dwell on some specificities of 5G localization: single base-station and carrier-phase positioning, 6D problems, and signal design. Finally, it will address how the presence of RIS in the near or far field can be exploited for localization.

Global navigation satellite system (GNSS)-based attitude determination plays a critical role in various navigation, guidance, and control applications. GNSS attitude determination has been applied in various land, airborne, and maritime scenarios. The goal of GNSS attitude determination is to estimate a vehicle’s or platform’s orientation in a reference coordinate system utilizing multiple GNSS antennas/receivers rigidly mounted on the body frame. To estimate a platform’s attitude, it is critical to resolve the integer ambiguities of the carrier-phase observations correctly, an onerous mission that becomes even harder in urban areas with limited satellite visibility. This dramatically increases the ambiguity search time. In this talk, we explore ways of reducing the time complexity of GNSS ambiguity resolution by improving the optimization techniques and the search algorithms.

This talk presents the design of a high-accuracy spatial location estimation method using ultrasound waves by exploiting the fixed geometry of the transmitters. Assuming an isosceles triangle antenna configuration, where three antennas are placed as the vertices of an isosceles triangle, the spatial location problem can be formulated as a non-convex optimization problem whose interior is shown to admit a Riemannian manifold structure. Our investigation of the geometry of the newly introduced manifold (i.e., the manifold of all isosceles triangles in R3) enables the design of highly efficient optimization algorithms. Simulations are presented to compare the performance of the proposed approach with popular methods from the literature. The results suggest that the proposed Riemannian-based methods outperform the state-of-the-art methods. Furthermore, the proposed Riemannian methods require much less computation time compared to popular generic non-convex approaches.