Search

We introduce an open-source architecture for next-generation Radio-Access Network baseband processing: 1024 latency-tolerant 32-bit RISC-V cores share 4 MiB of L1 memory via an ultra-low latency interconnect (7-11 cycles), a modular Direct Memory Access engine provides an efficient link to a high bandwidth memory, such as HBM2E (98% peak bandwidth at 910GBps). The system achieves leading-edge energy efficiency at sub-ms latency in key 6G baseband processing kernels: Fast Fourier Transform (93 GOPS/W), Beamforming (125 GOPS/W), Channel Estimation (96 GOPS/W), and Linear System Inversion (61 GOPS/W), with only 9% data movement overhead.

Radio Access Networks (RAN) workloads are rapidly scaling up in data processing intensity and throughput as the 5G (and beyond) standards grow in number of antennas and sub-carriers. Offering flexible Processing Elements (PEs), efficient memory access, and a productive parallel programming model, many-core clusters are a well-matched architecture for next-generation software-defined RANs, but staggering performance requirements demand a high number of PEs coupled with extreme Power, Performance and Area (PPA) efficiency. We present the architecture, design, and full physical implementation of Terapool-SDR, a cluster for Software Defined Radio (SDR) with 1024 latency-tolerant, compact RV32 PEs, sharing a global view of a 4MiB, 4096-banked, L1 memory. We report various feasible configurations of TeraPool-SDR featuring an ultra-high bandwidth PE-to-L1-memory interconnect, clocked at 730MHz, 880MHz, and 924MHz (TT/0.80 V/25 {\deg}C) in 12nm FinFET technology. The TeraPool-SDR cluster achieves high energy efficiency on all SDR key kernels for 5G RANs: Fast Fourier Transform (93GOPS/W), Matrix-Multiplication (125GOPS/W), Channel Estimation (96GOPS/W), and Linear System Inversion (61GOPS/W). For all the kernels, it consumes less than 10W, in compliance with industry standards., Comment: 6 pages, 6 figures and 3 tables

While 5G networks are already being deployed for commercial applications, Academia and industry are focusing their effort on the development and standardization of the next generations of mobile networks, i.e., 5G-Advance and 6G. Beyond 5G networks will revolutionize communications systems providing seamless connectivity, both in time and in space, to a unique ecosystem consisting of the convergence of the digital, physical, and human domains. In this scenario, NonTerrestrial Networks (NTN) will play a crucial role by providing ubiquitous, secure, and resilient infrastructure fully integrated into the overall system. The additional network complexity introduced by the third dimension of the architecture requires the interoperability of different network elements, enabled by the disaggregation and virtualization of network components, their interconnection by standard interfaces and orchestration by data-driven network artificial intelligence. The disaggregation paradigm foresees the division of the radio access network in different virtualized block of functions, introducing the concept of functional split. Wisely selecting the RAN functional split is possible to better exploit the system resources, obtaining costs saving, and to increase the system performances. In this paper, we firstly provide a discussion of the current 6G NTN development in terms of architectural solutions and then, we thoroughly analyze the impact of the typical NTN channel impairments on the available functional splits. Finally, the benefits of introducing the dynamic optimization of the functional split in NTN are analyzed, together with the foreseen challenges.

Synchronization is likely the most critical performance killer in shared-memory parallel programs. With the rise of multi-core and many-core processors, the relative impact on performance and energy overhead of synchronization is bound to grow. This paper focuses on barrier synchronization for TeraPool, a cluster of 1024 RISC-V processors with non-uniform memory access to a tightly coupled 4MB shared L1 data memory. We compare the synchronization strategies available in other multi-core and many-core clusters to identify the optimal native barrier kernel for TeraPool. We benchmark a set of optimized barrier implementations and evaluate their performance in the framework of the widespread fork-join Open-MP style programming model. We test parallel kernels from the signal-processing and telecommunications domain, achieving less than 10% synchronization overhead over the total runtime for problems that fit TeraPool's L1 memory. By fine-tuning our tree barriers, we achieve 1.6x speed-up with respect to a naive central counter barrier and just 6.2% overhead on a typical 5G application, including a challenging multistage synchronization kernel. To our knowledge, this is the first work where shared-memory barriers are used for the synchronization of a thousand processing elements tightly coupled to shared data memory., Comment: 15 pages, 7 figures

While 5G networks are being rolled out, the definition of the potential 5G-Advanced features and the identification of disruptive technologies for 6G systems are being addressed by the scientific and academic communities to tackle the challenges that 2030 communication systems will face, such as terabit-capacity and always-on networks. In this framework, it is globally recognised that Non-Terrestrial Networks (NTN) will play a fundamental role in support to a fully connected world, in which physical, human, and digital domains will converge. In this framework, one of the main challenges that NTN have to address is the provision of the high throughput requested by the new ecosystem. In this paper, we focus on Cell-Free Multiple Input Multiple Output (CF-MIMO) algorithms for NTN. In particular: i) we discuss the architecture design supporting centralised and federated CF-MIMO in NTN, with the latter implementing distributed MIMO algorithms from multiple satellites in the same formation (swarm); ii) propose a novel location-based CF-MIMO algorithm, which does not require Channel State Information (CSI) at the transmitter; and iii) design novel normalisation approaches for federated CF-MIMO in NTN, to cope with the constraints on non-colocated radiating elements. The numerical results substantiate the good performance of the proposed algorithm, also in the presence of non-ideal information., Comment: Submitted to IEEE Access

5G Radio access network disaggregation and softwarization pose challenges in terms of computational performance to the processing units. At the physical layer level, the baseband processing computational effort is typically offloaded to specialized hardware accelerators. However, the trend toward software-defined radio-access networks demands flexible, programmable architectures. In this paper, we explore the software design, parallelization and optimization of the key kernels of the lower physical layer (PHY) for physical uplink shared channel (PUSCH) reception on MemPool and TeraPool, two manycore systems having respectively 256 and 1024 small and efficient RISC-V cores with a large shared L1 data memory. PUSCH processing is demanding and strictly time-constrained, it represents a challenge for the baseband processors, and it is also common to most of the uplink channels. Our analysis thus generalizes to the entire lower PHY of the uplink receiver at gNodeB (gNB). Based on the evaluation of the computational effort (in multiply-accumulate operations) required by the PUSCH algorithmic stages, we focus on the parallel implementation of the dominant kernels, namely fast Fourier transform, matrix-matrix multiplication, and matrix decomposition kernels for the solution of linear systems. Our optimized parallel kernels achieve respectively on MemPool and TeraPool speedups of 211, 225, 158, and 762, 880, 722, at high utilization (0.81, 0.89, 0.71, and 0.74, 0.88, 0.71), comparable a single-core serial execution, moving a step closer toward a full-software PUSCH implementation., Comment: 6 pages, 9 figures, conference paper

Today, 5G networks are being worldwide rolled out, with significant benefits in our economy and society. However, 5G systems alone are not expected to be sufficient for the challenges that 2030 networks will experience, including, e.g., always-on networks, 1 Tbps peak data rate, <10 cm positioning, etc. Thus, the definition of evolutions of the 5G systems and their (r)evolutions are already being addressed by the scientific and industrial communities, targeting 5G-Advanced (5G-A) and 6G. In this framework, Non-Terrestrial Networks (NTN) have successfully been integrated in 3GPP Rel. 17 and it is expected that they will play an even more pivotal role for 5G-A (up to Rel. 20) and 6G systems (beyond Rel. 20). In this paper, we explore the path that will lead to 5G-A and 6G NTN communications, providing a clear perspective in terms of system architecture, services, technologies, and standardisation roadmap.

Satellite communication systems are a fundamental component in support of Europe's ambition to deploy smart and sustainable networks and services for the success of its digital economy. To cope with the 5G and beyond ever increasing demand for larger throughput, aggressive frequency reuse schemes (i.e., full frequency reuse), with the implementation of precoding/beamforming to cope with the massive co-channel interference, are recognised as one of the key technologies. While the best performance can be obtained with the knowledge of the Channel State Information (CSI) at the transmitter, this also poses some technical challenges related to signalling and synchronisation. In this paper, we focus on precoding solutions that only needs the knowledge of the users' positions at the transmitter side, namely the recently introduced Switchable Multi-Beam (MB) and Spatially Sampled MMSE (SS-MMSE) precoding. Compared to the vast majority of the studies in the literature, we take into account both the users' and the satellite movement in a Low Earth Orbit (LEO) mega-constellation, also proposing two system architectures. The extensive numerical assessment provides a valuable insight on the performance of these two precoding schemes compared to the optimal MMSE solution., Comment: Accepted for publication to EuCNC 2022

With the first commercializations of 5G networks, Beyond 5G (B5G), or 6G, systems are starting to be defined. In this context, Internet of Things (IoT) services will be even more impactful with respect to 5G systems. In order to cope with the huge amount of IoT devices, and the potentially large capacity requirements for the most advanced of them (e.g., live camera feeds from first responders in emergency scenarios), non-terrestrial systems will be pivotal to assist and complement the terrestrial networks. In this paper, we propose a multi-layer non-terrestrial architecture for NarrowBand-IoT (NB-IoT) services based on Geosynchronous Orbit (GSO) and Non GSO (NGSO) nodes. To cope with both the number of devices and the potentially large capacities, we propose to implement Minimum Mean Square Error (MMSE) beamforming in an aggressive full frequency reuse scenario. The performance assessment shows the significant benefits of the proposed solution., Comment: Submitted to IEEE ICC'21 Workshop - SatMegaConst 2021

Current State-of-the-Art High Throughput Satellite systems provide wide-area connectivity through multi-beam architectures. Due to the tremendous system throughput requirements that next generation Satellite Communications (SatCom) expect to achieve, traditional 4-colour frequency reuse schemes are not sufficient anymore and more aggressive solutions as full frequency reuse are being considered for multi-beam SatCom. These approaches require advanced interference management techniques to cope with the significantly increased inter-beam interference both at the transmitter, e.g., precoding, and at the receiver, e.g., Multi User Detection (MUD). With respect to the former, several peculiar challenges arise when designed for SatCom systems. In particular, multiple users are multiplexed in the same transmission radio frame, thus imposing to consider multiple channel matrices when computing the precoding coefficients. In previous works, the main focus has been on the users' clustering and precoding design. However, even though achieving significant throughput gains, no analysis has been performed on the impact of the system scheduling algorithm on multicast precoding, which is typically assumed random. In this paper, we focus on this aspect by showing that, although the overall system performance is improved, a random scheduler does not properly tackle specific scenarios in which the precoding algorithm can poorly perform. Based on these considerations, we design a Geographical Scheduling Algorithm (GSA) aimed at improving the precoding performance in these critical scenarios and, consequently, the performance at system level as well. Through extensive numerical simulations, we show that the proposed GSA provides a significant performance improvement with respect to the legacy random scheduling., Comment: To be submitted to ASMS 2018

Next generation multi-beam SatCom architectures will heavily exploit full frequency reuse schemes along with interference management techniques, e.g., precoding or multiuser detection, to drastically increase the system throughput. In this framework, we address the problem of the user selection for multicast precoding by formulating it as a clustering problem. By introducing a novel mathematical framework, we design fixed-/variable-size clustering algorithms that group users into simultaneously precoded and served clusters while maximising the system throughput. Numerical simulations are used to validate the proposed algorithms and to identify the main system-level trade-offs., Comment: Submitted to International Journal of Satellite Communications and Networking. arXiv admin note: text overlap with arXiv:1706.09482

3GPP is finalising the first release of the 5G New Radio physical layer. To cope with the demanding 5G requirements on global connectivity and large throughput, Satellite Communications might be a valuable resource to extend and complement terrestrial networks. In this context, we introduce an integrated architecture for 5G-based LEO mega-constellations and assess the impact of large Doppler shifts and delays on both the 5G waveform and the PHY/MAC layer procedures., Comment: Accepted for publication on IEEE 5G Tech Focus

Current State-of-the-Art High Throughput Satellite systems provide wide-area connectivity through multi-beam architectures. However, due to the tremendous system throughput requirements that next generation Satellite Communications expect to achieve, traditional 4-colour frequency reuse schemes, i.e., two frequency bands with two orthogonal polarisations, are not sufficient anymore and more aggressive solutions as full frequency reuse are gaining momentum. These approaches require advanced interference management techniques to cope with the significantly increased inter-beam interference, like multicast precoding and Multi User Detection. With respect to the former, several peculiar challenges arise when designed for SatCom systems. In particular, to maximise resource usage while minimising delay and latency, multiple users are multiplexed in the same frame, thus imposing to consider multiple channel matrices when computing the precoding weights. In this paper, we focus on this aspect by re-formulating it as a clustering problem. After introducing a novel mathematical framework, we design a k-means-based clustering algorithm to group users into simultaneously precoded and served clusters. Two similarity metrics are used to this aim: the users' Euclidean distance and their channel distance, i.e., distance in the multidimensional channel vector space. Through extensive numerical simulations, we substantiate the proposed algorithms and identify the parameters driving the system performance., Comment: Submitted to IEEE Transactions on Wireless Communications

The standardization of 5G systems is entering in its critical phase, with 3GPP that will publish the PHY standard by June 2017. In order to meet the demanding 5G requirements both in terms of large throughput and global connectivity, Satellite Communications provide a valuable resource to extend and complement terrestrial networks. In this context, we consider a heterogeneous architecture in which a LEO mega-constellation satellite system provides backhaul connectivity to terrestrial 5G Relay Nodes, which create an on-ground 5G network. Since large delays and Doppler shifts related to satellite channels pose severe challenges to terrestrial-based systems, in this paper we assess their impact on the future 5G PHY and MAC layer procedures. In addition, solutions are proposed for Random Access, waveform numerology, and HARQ procedures., Comment: Submitted to IEEE Global Communications Conference (GLOBECOM) 2017

In this paper, we propose two novel schemes to solve the problem of finding a quasi-optimal number of coded packets to multicast to a set of independent wireless receivers suffering different channel conditions. In particular, we propose two network channel virtualization schemes that allow for representing the set of intended receivers in a multicast group to be virtualized as one receiver. Such approach allows for a transmission scheme not only adapted to per-receiver channel variation over time, but to the network-virtualized channel representing all receivers in the multicast group. The first scheme capitalizes on a maximum erasure criterion introduced via the creation of a virtual worst per receiver per slot reference channel of the network. The second scheme capitalizes on a maximum completion time criterion by the use of the worst performing receiver channel as a virtual reference to the network. We apply such schemes to a GEO satellite scenario. We demonstrate the benefits of the proposed schemes comparing them to a per-receiver point-to-point adaptive strategy.

In this paper, we propose two novel physical layer aware adaptive network coding and coded modulation schemes for time variant channels. The proposed schemes have been applied to different satellite communications scenarios with different Round Trip Times (RTT). Compared to adaptive network coding, and classical non-adaptive network coding schemes for time variant channels, as benchmarks, the proposed schemes demonstrate that adaptation of packet transmission based on the channel variation and corresponding erasures allows for significant gains in terms of throughput, delay and energy efficiency. We shed light on the trade-off between energy efficiency and delay-throughput gains, demonstrating that conservative adaptive approaches that favors less transmission under high erasures, might cause higher delay and less throughput gains in comparison to non-conservative approaches that favor more transmission to account for high erasures., Comment: IEEE Advanced Satellite Multimedia Systems Conference and the 14th Signal Processing for Space Communications Workshop (ASMS/SPSC), 2016

Multi-user Multiple-Input Multiple-Output (MU-MIMO) has allowed recent releases of terrestrial LTE standards to achieve significant improvements in terms of offered system capacity. The publications of the DVB-S2x standard and particularly of its novel superframe structure is a key enabler for applying similar interference management techniques -such as precoding- to multibeam High Throughput Satellite (HTS) systems. This paper presents results resulting from European Space Agency (ESA) funded R&D activities concerning the practical issues that arise when precoding is applied over an aggressive frequency re-use HTS network. In addressing these issues, the paper also proposes pragmatic solutions that have been developed in order to overcome these limitations. Through the application of a comprehensive system simulator, it is demonstrated that important capacity gains (beyond 40%) are to be expected from applying precoding even after introducing a number of significant practical impairments.

We consider a windowed decoding scheme for LDPC convolutional codes that is based on the belief-propagation (BP) algorithm. We discuss the advantages of this decoding scheme and identify certain characteristics of LDPC convolutional code ensembles that exhibit good performance with the windowed decoder. We will consider the performance of these ensembles and codes over erasure channels with and without memory. We show that the structure of LDPC convolutional code ensembles is suitable to obtain performance close to the theoretical limits over the memoryless erasure channel, both for the BP decoder and windowed decoding. However, the same structure imposes limitations on the performance over erasure channels with memory., Comment: 18 pages, 9 figures, accepted for publication in the IEEE Transactions on Information Theory