Nokia (Belgium)

We demonstrate super-rated PON downstream transmission at 100Gb/s net rate using PAM2 and PAM4 line rates ranging between 109 and 124Gb/s, exploiting an optimized LOPC code set. We showcase loss budgets >30dB for transmission equivalent to 20km at 1320nm.

The emerging Video-based dynamic mesh coding standard is designed on the concepts of subdivision surfaces and the lifting scheme applied to a simplified variant of a dynamic mesh. Those processes create displacement data, i.e., residual information that is hierarchically predicted through each subdivision iteration, that is then coded using traditional 2D video codecs. The current draft of the standard defines the packing of the displacement data in a video frame in a way that does not support adaptations based on Level-of-Detail (LoD). In this paper, we describe and evaluate an experimental LoD adaptive displacement packing method that could be used to optimize bandwidth consumption or to save computational complexity and battery at the client side. We show that a modified packing of the displacement video data and a LoD-adaptive selection of lifting parameters, combined with HEVC motion constrained tile sets provide this functionality with a low impact on compression performance. The LoD adaptive displacement packing method described in this paper has been recently adopted into the V-DMC standard specifications.

To deal with the excessive transport capacity and power consumption challenges faced by Cell-free massive MIMO systems, we propose a fronthauling solution with MIMO-cluster-aware schedulable coherent digital subcarriers. The system power consumption is modelled, and simulation results show significant power saving compared to P2P optics.

This letter proposes a methodology for phase-normalization of the complex-valued I/Q inputs of a real-valued time delay neural network (RVTDNN). The normalization enables modeling of the nonlinear behavior of a radio frequency (RF) power amplifier (PA) in a more efficient way, by complying with the physical characteristics of the distortions at RF. The presented digital predistortion (DPD) linearization experiments with a Doherty GaN PA at 3.5 GHz show a 4-dB improvement in the output linearity compared to state-of-the-art neural network (NN) and polynomial-based DPD models, allowing linearization to below <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$-$</tex-math> </inline-formula> 50 dBc adjacent channel leakage ratio (ACLR) levels with feasible processing complexity.

Radio Access Network (RAN) openness is a vision for 5G and beyond to avoid unnecessary vendor lock-in effects and introduce new business models. To expand the Open RAN (O-RAN) potentials, such as real-time control and data-driven intelligence, we propose a cross-domain 5GECO network architecture to let flourish the new Intelligent Neutral Host (INH) model in the value chain. Also, to realize this new model, we consider the role of a INH Service provider as having to define a Service Level Agreement (SLA) with its customers, called tenants, and a scheme to map said SLA into 5G technology (termed the 5GECO multiplet) is elaborated. It is noted that this 5GECO architecture considers sharing, control, and orchestration across both the RAN and the Transport Network (TN) domains. Finally, three key challenges are identified for an INH Service Provider, along with a preliminary solution analysis, ranging from Radio Resource Management (RRM) optimization, cross-domain low-latency, and flexible virtualized resource scalability.

Reinforcement learning (RL) is a powerful machine learning technique to learn optimal actions in a control system setup. An important drawback of RL algorithms is the need for balancing exploitation vs exploration. Exploration corresponds to taking randomized actions with the aim to learn from it and make better decisions in the future. However, these exploratory actions result in poor performance, and current RL algorithms have a slow convergence as one can only learn from a single action outcome per iteration. We propose a novel concept of Inference-based RL that is applicable to a specific class of RL problems, and that allows to eliminate the performance impact caused by traditional exploration strategies, thereby making RL performance more consistent and greatly improving the convergence speed. The specific RL problem class is a problem class in which the observation of the outcome of one action can be used to infer the outcome of other actions, without the need to actually perform them. We apply this novel concept to the use case of dynamic resource allocation, and show that the proposed algorithm outperforms existing RL algorithms, yielding a drastic increase in both convergence speed and performance.

Mobile network operators are facing the rollout of 5G New Radio small cells in the next few years. High cell density will require intensive support of optical transport, and there are multiple options depending on cell configuration and capacity. Comparing these optical technologies in terms of capacity and cost is important to maximize return on the likely limited deployment budget. We developed a model that can be used by operators in the strategy and decision-making phase and present example results on outdoor 5G small cells using the sub-6-GHz spectrum bands.

The promise of future generations of mobile communication, both 5.5 and 6G, is that of continued increases in capacity and coverage of the wireless communication network together with novel applications demanding higher levels of performance between the mobile devices and their applications on servers in the cloud. In addition, the growing shift towards virtualization of network function itself offers perhaps the most stringent requirements on the network. To continue to offer ever higher capacities and speeds, future networks will be required to further densify addressing capacity and coverage demands through the use of small cells or new radio spectral bands, such as 7-20GHZ or sub-THZ, with inherently shorter range. Therein lies the dilemma. A denser and ever more complex network of access points, with higher speeds and performance, but continued pressure on the deployment and costs. To address this challenge solutions which address all aspects of the End to End network must be considered together &#x2013; the mobile access points, transport networks, deployment options, and edge cloud all contribute equally to the success of the new consumer and industrial applications promised for next generation networks as well as the operation of the networks themselves. In this paper we will outline the expected requirements of next generation networks as well as new technologies and methods which address the simultaneous challenges of higher capacity, lower costs, and reliable deterministic performance.