The March 8-9, 2023 Workshop, co-organized by LINCS and ERC NEMO, dealt with a central problem for the new generations of wireless networks which is that of real-time type guarantees. These guarantees are necessary for industrial applications of 5G and 6G. Deterministic guarantees are essential for critical real-time. This workshop showed that the problem required fundamental contributions from several areas.
The first area is that of information and coding theory. Shannon’s theory, which is based on codes using blocks of length tending towards infinity, cannot be used in this context of time constraints and must be replaced by a theory allowing the treatment of the short block length regime. This can be combined with new forms of coding, based on a differentiated way of treating information with stronger latency or reliability constraints.
The second area is that of the sharing of time, frequency and spatial resources, allowing to obtain guarantees on the end-to-end delays in radio networks. Classical network calculus, which makes it possible to give guarantees on latency in wireline networks by means of time flow regulation, must be complemented by new forms of regulated access to space and to frequencies and by the development of distributed algorithms allowing to implement these regulations in very large networks.
The integration of joint advances in these areas is expected to allow for e.g. the design of new multiple access methods meeting the requirements of mass access and guaranteed latencies for objects connected to the new generations of cellular networks.
This line of research is one of those retained in the project “Foundations of communication networks” of the PEPR on “networks of the future” which is about to be launched. This axis is a good illustration of the nature of the long-term scientific challenges that are those of the French community. It also illustrates the new multidisciplinary interactions that are currently being initiated in this field in the framework of this PEPR.
You can find the recordings for all the talks on the LINCS YouTube channel : https://www.youtube.com/playlist?list=PLlzbq4-mnm8TjbRRXiZ_lhnYtC-3YY59y and access the slides of the presentations : https://drive.google.com/drive/folders/1KjQMED32OtTsf6WauTR4OgGZr1-SqrWk
Invited speakers (4 sessions):
1. Fundamental limits
JEAN-MARIE GORCE (INSA LYON / INRIA LYON) – Performance guarantees in a wireless cell with massive access: beyond capacity with non asymptotic information theory
In the context of machine-to-machine communications, the information transmitted in a cell differs from the information modeled in conventional communications. The M2M paradigm relies on distributions of many sources or destinations with sporadic transmissions. The information transmitted in a the network is a superposition of small information quantities with strong latency and reliability constraints (URLLC). What is the price of reliabilty/latency ? We propose in this presentation a survey of the guaranties that can be provided in an information theory framework, especially in the non asymptotic regime and in mumlti-user scenarios. Open questions and challenges will be drawn.
DERYA MALAK (EURECOM) – The Interplay of Spectral Efficiency, User Density, and Energy in Random Access Protocols
The fifth-generation of wireless communication networks is required to support a range of use cases such as enhanced mobile broadband (eMBB), ultra-reliable, low-latency communications (URLLC), massive machine-type communications (mMTCs), with heterogeneous data rate, delay, and power requirements. The 4G LTE air interface uses extra overhead to enable scheduled access, which is not justified for small payload sizes. We employ a random access communication model with retransmissions for multiple users with small payloads at the low spectral efficiency regime. The radio resources are split non-orthogonally in the time and frequency dimensions. Retransmissions are combined via Hybrid Automatic Repeat reQuest (HARQ) methods, namely Chase Combining and Incremental Redundancy with a finite buffer size constraint Cbuf . We determine the best scaling for the spectral efficiency (SE) versus signal-to-noise ratio (SNR) per bit and for the user density versus SNR per bit, for the sum-optimal regime and when the interference is treated as noise, using a Shannon capacity approximation as well as for the finite blocklength regime. Numerical results show that the scaling results are applicable over a range of ?, T, Cbuf , J, at low received SNR values. The proposed analytical framework provides insights for resource allocation in general random access systems and specific 5G use cases for massive URLLC uplink access.
LAURENT CLAVIER (IMT NORD EUROPE) – Some words about interference
Interference is a strong limiting factor in the deployment of wireless networks. The very nature of this interference, analysed as noise, can vary greatly depending in particular on channel access strategies. We will focus on the modelling of this interference. In particular, we will discuss environments where network coordination is very limited and dynamicity can be high.
SAMIR M. PERLAZA (INRIA SOPHIA) – An Upper Bound on the Error Induced by Saddlepoint Approximations—Applications to Wireless Communications
In this talk, an upper bound on the absolute difference between: (a) the cumulative distribution function (CDF) of the sum of a finite number of independent and identically distributed random variables with finite absolute third moment; and (b) a saddlepoint approximation of such CDF is introduced. This upper bound, which is particularly precise in the regime of large deviations, is used to study the dependence testing (DT) bound and the meta converse (MC) bound on the decoding error probability (DEP) in point-to-point memoryless channels. Often, these bounds cannot be analytically calculated and thus lower and upper bounds become particularly useful. Within this context, the main results include, respectively, new upper and lower bounds on the DT and MC bounds. A numerical experimentation of these bounds is presented in the case of the binary symmetric channel, the additive white Gaussian noise channel, and the additive symmetric ?-stable noise channel. The same type of result is also presented for the case of finite sums of real-valued independent and identically distributed random vectors. In such a case, the approximation error is upper bounded and thus, as a byproduct, an upper bound and a lower bound on the CDF of the sum are obtained.
2. Short Codes
EMMANUEL BOUTILLON (UNIVERSITÉ DE BRETAGNE SUD) – Quasi Cyclic Short Packets for unslotted aloha transmission.
Massice IoT applications are characterized by an expected high density of connected devices, small data payloads, as well as stringent constraints on the device energy consumption and cost. Maximizing the spectral efficiency of an IoT network is a key pre-requisite for providing massive connectivity. At the link level, it can take advantage of powerful error control codes such as Non-Binary (NB) codes. At the system level, reducing “meta-data” throughput, i.e., exchange of information linked to signaling, synchronization and identification is the new paradigm of massive IoT network. This requires encompassing those functions in a single well protected frame.
In this talk, we will present a new coded modulation scheme based on the association of a non-binary code combined with a Cyclic Code Shift Keying (CCSK) modulation. This new coded modulation scheme, called Quasi Cyclic Short Packet (QCSP), provides several advantages compared to state-of-the-art waveforms: it avoids the use of a preamble for detection and synchronization and allows transmission of small payload at Signal-to-Noise Ratios (SNR) far below 0 dB with performance rather close (1.5 dB) to the theoretical limit. Some real-time experimentation results are shown for the mobile channel and a maritime channel. Simulation of multi-users transmission shows that QCSP frame are an efficient solution for the unslotted aloha channel.
MAXIME GUILLAUD (INRIA LYON) – Towards Practical Waveforms for Massive Random Access
In multi-user wireless communications, random access departs from the classical set-up of multiple access in the fact that the number and identity of the active transmitters are unknown to the receiver (typically because the sporadic nature of the traffic does not allow for coordination between transmitters). This is a significant departure from the well-understood multiple access schemes (including non-orthogonal multiple access, NOMA). Random access arises e.g. in massive internet-of-things, ultra-reliable and low-latency communications, and non-terrestrial networks applications. This talk will outline why, compared to multiple access, multi-user decoding in random access scenarios is markedly more difficult, and requires to revise some of the basic assumptions that underpin modern multi-user communications systems, such as the pre-existence of synchronization and timing offset compensation, or centralized assignment of pilot sequences. Scenarios of practical interest will be discussed, including fading and multiple-antenna channels, and synchronization impairments. I will focus on massive random access, which explicitly considers the high-contention regime, i.e. the case where the number of simultaneously active transmitters can be very large, and discuss some of the practical waveforms and coding approaches that have been proposed in practice to solve this problem. In particular, I will discuss tensor-based modulation, a class of waveforms adapted to massive random access, and discuss its applicability over practical wireless channels.
PHILIPPE MARY (CNRS RENNES) – Increasing the performance of polar codes at short blocklength
Polar codes, the first provably asymptotically capacity achieving error correcting codes over binary input memoryless channels with explicit construction, are currently used over the control channels of 5G networks. They also are envisioned for ultra-reliable low-latency communications and massive machine-type communications thanks to their low complexity successive cancellation based decoder. Unfortunately, standard polar codes do not show outstanding performance at short-to-moderate block lengths due to their poor minimum distance and a non-complete polarization. Several techniques have been proposed over the years to improve the distance spectrum at short block length in order to counter the poor channel channel polarization at these block lengths. In this talk, we go through the main techniques used in literature to construct good polar codes at short to moderate block length and we present a new precoding scheme that allows to increase the information length of at least one bit of the underlying Reed-Muller code. The proposed technique is shown to perform better than the polarization adjusted convolutional polar code with the same parameters.
CHARLY POULLIAT (CNRS TOULOUSE) – Power efficient Communications: asymptotic versus finite length error correcting schemes performance
In this talk, we discuss the performance of several power efficient communications schemes for both the asymptotic and the finite length regimes, including schemes such as LoRA or code shift keying modulations. To this end, we investigate on taylored sparse graph based binary and non binary schemes for the Gaussian and Rayleigh fading channels, for both the coherent and non coherent regimes. We will see that the use of binary coding schemes is only viable if bit-interleaved coded modulation with iterative decoding is considered or if we adopt a multi-level coded modulation approach. The former induced an increase of complexity compared to the classical linear modulation case and some extra latency, while the latter induced a high number of layers, as reduced layer strategies cannot be as efficient as in the linear modulation case. Considering non binary coding schemes, we will show that if they exhibit good performance at finite length in the Gaussian channel and can operate close to the capacity, they suffer from a great penalty in the Rayleigh channel case if not designed properly, showing that simple non binary cycle codes are from far not sufficient for this kind of channels. In addition, as the modulation order increases, we need to consider simple coding strategies to avoid a too high field order to limit the complexity at the receiver. Finally, we will compare performance of such schemes to existing low-rate coding schemes designed for the binary shift keying modulation.
3. URLLC 1
JEAN-YVES LEBOUDEC (EPFL) – Time Sensitive Networks and Network Calculus and Clock Non-idealities
Time Sensitive Networks offer guarantees on worst-case delay, worst-case delay variation and zero congestion loss; in addition, they provides mechanisms for packet duplication in order to hide residual losses due to transmission errors. They find applications in many areas such as factory automation, embedded and vehicular networks, audio-visual studio networks, and in the front-hauls of cellular wireless networks. In this talk we will describe how network calculus can be used to analyze time sensitive networks with components such as packet ordering and duplicate removal functions, schedulers, regulators and dampers. We will also explain why clock non-idealities matter, and will describe how to take them into account.
KE FENG (INRIA PARIS) – Spatial Network Calculus and Performance Guarantees in Wireless Networks
Network calculus is initially a methodology allowing one to provide performance guarantees in queuing networks subject to regulated traffic arrivals and service guarantees. It is a key design tool for latency-critical wireline communication networks where it allows one to e.g. guarantee bounds on the end-to-end latency of all transmitted packets. In wireless networks, service guarantees are more intricate as electromagnetic signals propagate in a heterogeneous medium and interferer each other. In this talk, we present a novel approach toward performance guarantees for all links in arbitrarily large wireless networks. We introduce spatial regulation properties for stationary spatial point processes, which model transmitter and receiver locations, and develop the first steps of a calculus for this type of regulation. This can be seen as an extension to space of the classical network calculus developed with respect to time. Using this approach, we derive reliability, rate and latency guarantees for all links in various types of wireless architectures. Such guarantees do not exist in networks without spatial regulations, e.g., Poisson networks.
MARIOS KOUNTOURIS (EURECOM) – Goal-oriented Communication for Distributed Intelligent Systems
Wireless networks are evolving to cater to cyber-physical and mission-critical interactive systems, such as swarm robotics, self-driving cars, and smart Internet of Things. As we are entering the era of networked intelligence, fundamental advances are necessary to satisfy the pressing requirements for real-time communication, timely decision-making, and effective distributed processing. In this talk, we introduce goal-oriented semantic communication; a paradigm shift that aims at redefining data importance, timing, and effectiveness. First, taking on a rate-distortion approach, we introduce a variant of a robust description source coding with two individual single-letter separable distortion constraints. We provide several theoretical results that allow us to highlight the cardinal role of context-dependent fidelity criteria in goal-oriented communication. Second, we consider the problem of semantic filtering and timely source coding, and we determine the optimal codeword lengths assigned to packet arrivals so as to maximize a weighted sum of semantics-aware utility functions in both single-user and multiuser systems. Third, we present new joint sampling and information transmission policies for real-time tracking and reconstruction of an information source with the purpose of actuation, as a means to drastically reduce the generation and transmission of unimportant packets. We conclude this talk by discussing the potential and the technical challenges associated with this promising avenue of research.
MICHÈLE WIGGER (TÉLÉCOM-PARIS) – Capacity-Tradeoffs in Networks with Mixed-Delay Traffics and Random Arrivals
In this talk we present the information-theoretic fundamental limits of regular cellular networks with mixed-delay traffics, and we discuss the tradeoffs between the performances of the different types of traffics. In particular, we shall describe a coding strategy that can mitigate the penalty on the overall performance of the system caused by stringent delay-constraints on part of the traffic. Benefits and challenges related to random arrivals of low-latency traffics are also discussed, as well as finite-blocklength analyses of our models.
4. URLLC 2
MARCO DI RENZO (CENTRALE SUPELEC) – Analysis of the Delay Distribution in Cellular Networks by Using Stochastic Geometry
In this talk, with the aid of the mathematical tool of stochastic geometry, we introduce analytical and computational frameworks for the distribution of three different definitions of delay, i.e., the time that it takes for a user to successfully receive a data packet, in large-scale cellular networks. We also provide an asymptotic analysis of one of the delay distributions, which can be regarded as the packet loss probability of a given network. To mitigate the inherent computational difficulties of the obtained analytical formulations in some cases, we propose efficient numerical approximations based on the numerical inversion method, the Riemann sum, and the Beta distribution. Finally, we demonstrate the accuracy of the obtained analytical formulations and the corresponding approximations against Monte Carlo simulation results, and unveil insights on the delay performance with respect to several design parameters, such as the decoding threshold, the transmit power, and the deployment density of the base stations. The proposed methods can facilitate the analysis and optimization of cellular networks subject to reliability constraints on the network packet delay that are not restricted to the local (average) delay, e.g., in the context of delay sensitive applications.
MARCEAU COUPECHOUX AND BENOÎT-MARIE ROBAGLIA (TÉLÉCOM-PARIS) – Deep Reinforcement Learning for Multiple Access in uplink URLLC networks
Distributed Medium Access Control (MAC) protocols have been studied for many decades and are incorporated in industry standards and commercial products. However, traditional protocols based on random access fall short in many ways as they cannot ensure any latency guarantees and cannot adapt well to dynamic environments, missing out strategic opportunities for optimal transmission. Due to the requirement for supporting Ultra Reliable Low Latency Communications (URLLC), they become incompatible with 5G and 6G wireless networks. Deep Reinforcement Learning (DRL) is a framework where decision makers learn to make optimal actions by interacting with an environment. Due to their capacity to learn effective transmission protocols that can adjust to the network dynamics and unique characteristics of many users, this family of algorithms holds promise for overcoming this new constraint. In this talk, we are going to explore how DRL can be applied to multiple access, both from a single-agent and a multiple-agent perspective, what are the challenges that traditional DRL methods have with the multiple access problem and see potential solutions.
SALAH ELAYOUBI (CENTRALE SUPELEC) – Rediscovering Aloha for latency-critical services: the blind and the far-sighted
Contention-based channel access is the oldest wireless data communications technology, where transmitters send packets on a shared medium in a completely non-coordinated way. For a long time, Aloha seemed surpassed by protocols that ensure a certain level of coordination, such as TDMA centralized scheduling or CSMA-like communications. However, with the integration of Industrial Internet of Things (IIoT) services within the wireless networks, Aloha protocols are becoming popular again, as stringent delay constraints make coordination and scheduling almost impossible. We explore the Aloha-like protocols for critical services, where transmitters are completely blind (do not receive any feedback about their actions) or far-sighted (receive a delayed feedback). Repetitions are key in this context, hoping that the repetition rate ensures reliability without much aggressively. We derive the optimal repetition rates for different scenarios and apply them to the main 5G radio configurations.
INBAR FIJALKOW (CNRS CERGY) – NOMA uplink networks under statistical delay constraints
We study the performance of an uplink non-orthogonal multiple access (NOMA) network under statistical quality of service (QoS) delay constraints, captured through each user’s effective capacity (EC). We first propose novel closed-form expressions for the EC in a two-user NOMA network and show that in the high signal-to-noise ratio (SNR) region, the “strong” NOMA user, referred to as U 2 , has a limited EC, assuming the same delay constraint as the “weak” user, referred to as U1. We demonstrate that for the weak user U1, OMA and NOMA have comparable performance at low transmit SNRs, while NOMA outperforms OMA in terms of EC at high SNRs. On the other hand, for the strong user U 2 , NOMA achieves higher EC than OMA at small SNRs, while OMA becomes more beneficial at high SNRs. Furthermore, we show that at high transmit SNRs, irrespective of whether the application is delay tolerant, or not, the performance gains of NOMA over OMA for U 1 , and OMA over NOMA for U 2 remain unchanged. When the delay QoS of one user is fixed, the performance gap between NOMA and OMA in terms of total EC increases with decreasing statistical delay QoS constraints for the other user. Next, by introducing pairing, we show that NOMA with user-pairing outperforms OMA, in terms of total uplink EC. The best pairing strategies are given in the cases of four and six users NOMA, raising once again the importance of power allocation in the optimization of NOMA’s performance.
