Massive MIMO for communication and radar coexistence

Abstract

In the evolving landscape of wireless communications, spectrum sharing, integrated sensing and communication (ISAC), and massive multiple-input multiple-output (mMIMO) technologies emerge as pivotal enablers for the 5G and forthcoming 6G networks. This thesis delves into the increasingly pertinent topic of frequency spectrum sharing between radar and communication systems, with a specific focus on the coexistence of mMIMO systems and MIMO radar. The research encompasses four comprehensive studies, each offering unique insights and innovative solutions to this challenging domain.

The first study investigates the coexistence between a multiuser mMIMO downlink system and MIMO radar. It reveals that increasing the number of antennas at the base station (BS) enhances the mMIMO performance while maintaining the interference levels to the radar system. A significant contribution of this study is the derivation of closed-form expressions for the probability of detection of the radar system and the downlink SE of the mMIMO system. Furthermore, the study introduces an innovative power allocation scheme that enables optimal transmit power selection in closed-form. This scheme maximizes the radar detection probability without compromising the performance of the mMIMO system.

Building on the foundational insights established in the first study, the second chapter expands the scope to include a nuanced analysis of linear precoding designs and their impact under channel estimation errors. This chapter explores the coexistence of a downlink multiuser mMIMO communication system and MIMO radar, with a focus on power control. The study characterizes the performance of the mMIMO system and the probability of detection with maximum ratio ($\MR$), zero-forcing ($\ZF$), and protective $\ZF$ ($\PZF$) precoding designs. The $\PZF$ precoding design is highlighted for its ability to protect radar operations by projecting the communication signals onto the null space of the radar channel. The study derives closed-form expressions for the detection probability under these precoding designs and explores the detection probability in multiple target scenarios and correlated fading environments. A power control problem is efficiently solved using a linear programming approach and the bisection method, aiming to maximize the radar detection probability while satisfying the per-user SE requirements. The analysis shows that the $\PZF$ design achieves the highest radar detection probability among all designs, with an intermediate SE performance that improves significantly with optimized power control.

Transitioning from the detailed exploration in the second chapter, the third study extends the discussion to the broader domain of cell-free massive MIMO (CF-mMIMO) systems within the integrated sensing and communication (ISAC) framework. It addresses the problem of access point (AP) operation mode selection, where some APs are dedicated to downlink communication and others to sensing. The research derives closed-form expressions for the individual SE and mainlobe-to-average-sidelobe ratio (MASR), assessing communication and sensing performances. A max–min fairness problem is formulated and solved, optimizing the minimum SE of users while adhering to per-AP power constraints and sensing MASR requirements. The numerical results demonstrate that the proposed AP operation mode selection with power control improves significantly the communication performance while meeting the specified sensing requirements. This approach not only enhances the overall network performance but also paves the way for future wireless networks where integrated communication and precise sensing are paramount. The findings and methodologies presented offer valuable insights for the development of advanced ISAC systems.

Building upon the sophisticated analysis presented in the third study, the fourth chapter advances the discourse into the more intricate realm of CF-mMIMO systems within the ISAC framework, with a special emphasis on the innovative concept of multi-zone sensing. This chapter not only builds upon the foundational insights from the third study but also explores the effectiveness of advanced precoding strategies—MR, ZF, and partial zero forcing (PZF)—in downlink communication for CF-mMIMO ISAC systems. It provides a comprehensive comparative analysis of each strategy, evaluating their unique advantages and potential limitations to identify the most beneficial approach. PZF is spotlighted as a transformative strategy that skillfully integrates the strengths of both ZF and MR precoding. The strategic essence of $\PZF$ is its refined interference management technique, which focuses on minimizing interference for users with the highest channel gains, effectively prioritizing the network's most robust users while tolerating some interference for its weakest users. This approach marks a significant leap in optimizing the network efficiency and enhancing user experience. Moreover, the study embarks on a thorough examination of system scalability and efficiency as the number of APs designated for both sensing and communication approaches infinity, while each AP's antenna count remains fixed. The analysis of the dynamics when the antenna count per AP increases to infinity, while the total number of APs remains constant. This detailed analysis is crucial for comprehending the scalability and adaptability of CF-mMIMO ISAC systems in complex wireless scenarios. Additionally, this chapter addresses the complex challenge of AP mode selection optimization. It ambitiously seeks to refine the operation mode selection and power control coefficients, aiming to maximize the minimum per-user SE within the boundaries set by specified MASR levels for target detection and transmit power constraints at the APs. Together, these studies provide comprehensive theoretical and practical insights into the coexistence of mMIMO communication systems and MIMO radar. They make significant contributions to the field of ISAC, paving the way for more efficient spectrum sharing in the future.

Date of Award	Jul 2024
Original language	English
Awarding Institution	Queen's University Belfast
Sponsors	Libyan Embassy (London)
Supervisor	Hien-Quoc Ngo (Supervisor) & Michalis Matthaiou (Supervisor)