Long Range (LoRa) technology is widely adopted in Low Power Wide Area Networks due to its robustness, energy efficiency, and long-range capabilities. However, its lightweight ALOHA-based medium access scheme makes it vulnerable to packet collisions, especially in dense deployments with multiple simultaneous transmissions. Efficient multiuser detection strategies are therefore essential to enhance system scalability and reliability. In this paper, we propose a novel multiuser detection scheme for LoRa systems that exploits the time offset information of received packets, obtained through an effective synchronization method, to mitigate mutual symbol interference. The proposed approach incorporates successive interference cancellation, performed at symbol level, that jointly leverages both power and delay characteristics of overlapping signals. Such combined use of temporal and power-domain knowledge enhances the capability to separate superposed transmissions, thereby improving detection performance in multiuser scenarios.

Handling multiple access represents a key challenge of indoor optical wireless communications, even though rarely addressed in the uplink case. Typical solutions in the literature rely on orthogonal access schemes, providing interference-free links to users but suffering from low performance when the number of access requests grows. An alternative is represented by non-orthogonal multiple access, even though its feasibility and effectiveness are generally assessed only for simple scenarios with few users. In this paper, we introduce a space-time division multiple access framework where access points equipped with an angle diversity receiver allow the channel resources to be managed in both space and time domain. Moreover, by resorting to successive interference cancellation, it is possible to serve users in a non-orthogonal fashion. Herein, we propose two novel resource allocation algorithms aimed to maximize the users access rate. Performance results demonstrate the capability of the proposed solution to serve up to 9 users, outperforming the other benchmark orthogonal and non-orthogonal strategies. Additionally, we introduced two mechanisms that, combined with successive interference cancellation, guarantee a reliable detection of superposing signals, with bit error rate below 10-3. Summarizing, the presented multiple access scheme allows users to transmit simultaneously and reliably, thus realizing channel resources saving as well.

Underwater optical communication (UWOC) is a promising technology for effective data transmission, nevertheless challenges like multipath interference, turbulence, and strong attenuation persist. Additionally, multiple access management represents a crucial issue to address for the realization of the so called Internet of Underwater Things networks (IoUT). In particular, uplink medium access control is a largely unexplored field, with performance of both known orthogonal and non orthogonal strategies being assessed only for basic scenarios. This paper introduces a novel uplink space division multiple access framework for underwater optical camera communication (UOCC), integrating image segmentation techniques to distinguish and process multiple users simultaneously. Spatial diversity enables initial users separation by position relative to the receiver, while segmentation enhances signal decoding efficiency at the pixel level. The proposed system is evaluated under realistic underwater conditions, addressing propagation impairments such as Gaussian blurring, intensity saturation and quantization artifacts. Simulation results validate the robustness and scalability of the approach, highlighting its potential to enable reliable multiuser communication in uplink scenarios.

The simultaneous task of data transmission and illumination achievable with Visible Light Communication is continuing to raise the interest of research and industry community. The joint use of the same infrastructure for illumination and communication may enable new location-based services, by implementing sensing functionalities. In this paper, we propose a new multi-dimensional modulation scheme combining different techniques, in order to increase the spectral and energy efficiency, while meeting the optical power emission constraints. Specifically, Pulse Position Modulation Pulse Amplitude Modulation and Frequency Shift Keying are effectively combined in a joint fashion, in order to exploit time, amplitude and frequency information. The innovation of using a smart combination of those modulation formats is based on the exploitation of specific features of the different approaches. In particular, we combine time, amplitude and frequency features to enhance the robustness of the system, without sacrificing the data rate performance as for coded systems. The major implication of such approach is twofold. In this way, we grant a constant illumination level per symbol and, moreover, we outperform in terms of reliability coded modulations for the same level of spectral efficiency. Theoretical results, validated through experimental evaluations, demonstrate that the combined approach achieves very good performance.

Long Range (LoRa) defines a particular chirp spread spectrum modulation employed for signaling in the LoRaWAN network. LoRa is recognized as one of the most effective technologies for the Internet of Things communication scenario, providing high reliability, wide coverage and low energy consumption for devices. The achievement of such performance is tied to the capability of LoRa receivers to accurately detect the signal even at low signal-to-noise ratio conditions. Dealing with signal demodulation, coherent and non-coherent detection represent the most widespread approaches. In this contribution, we present a new detection mechanism that exploits the geometrical properties of linear chirp to perform symbol demodulation, without requiring the energy consuming operations related to coherent and non-coherent detection. Furthermore, we demonstrate how the proposed approach can be fruitfully exploited not only for symbol demodulation, but also for LoRa packet synchronization and spreading factor recognition. By showing some numerical results, we compare the reliability level of the presented scheme through bit error rate with respect to the one offered by conventional detection methods.

Long Range (LoRa) represents an efficient low power solution for the Internet of Things. Specifically, LoRa defines a physical layer technology, while access control and network issues are handled by the LoRaWAN protocol. Channel access is essentially unslotted ALOHA, so LoRaWAN performance suffer from packet collision events. Such problem becomes more penalizing in dense network scenarios, where the large number of devices simultaneously connected makes the collision probability significantly grow. The largest part of solutions proposed to overcome LoRaWAN inefficiency is oriented to collision avoidance. Differently, we firstly present an algorithm for mitigating the interference among superposing LoRa signals, allowing collided packets to be detected anyway. Then, we propose a novel combining mechanism, implemented at the LoRaWAN network server, that exploits the information carried by the same packet, but received by different gateways, to achieve a more robust decoding. Hence, packet detection is reliably performed even in the presence of interference, thus reducing the need of retransmission and providing energy saving for end devices. Furthermore, simulation results show that the proposed solutions allow the end users to transmit exploiting low spreading factors, thus reducing the channel use that will be fundamental when dealing with larger scale network scenarios.

A novel cooperative least mean square adaptation rule is proposed with application to fractionally spaced blind channel equalization schemes used in quadrature amplitude modulation data links. A novel cost function based on second-order statistics only of the received signal is defined and jointly used with another chosen among classical blind cost functions such as decision directed and other Bussgang-type ones. Numerical results show that severe misconvergence observed in classical blind equalization algorithms is drastically mitigated using the here described technique.