Didier Colle

Photonic Network on Wafer (PNoW) for multi-chip computing systems is a promising solution for providing high inter-chip bandwidth while maintaining low energy consumption. This paper introduces a novel approach for routing optical waveguides in PNoWs on a 2-dimensional wafer plane, eliminating the need for any waveguide crossings. Our proposed method guarantees zero optical waveguide crossings, therefore high signal-to-noise ratio (SNR) and low signal loss in the photonics circuits. Notably, our proposed routing method is applicable to PNoWs with significantly higher inter-chip bandwidth, surpassing current commercial products by one magnitude. In contrast, we demonstrate that classical routing heuristics that allow waveguide crossings might result in a substantial degradation in terms of signal qualities, which is evaluated with certain approximations in this work. Furthermore, our grid-less routing method imposes minimal computational requirements, offering an efficient and practical solution for the implementation of high-performance PNoWs.

Hybrid beamforming has received significant attention as a solution to the thermal issues, costs, and implementation complexities associated with fully digital mmWave Extremely Large MIMO (XL-MIMO) systems. The Hybrid approach offers a balance between system flexibility and performance. However, even with optimal beamforming techniques, having a large number of User Equipments (UEs), if the data rate is not effectively shared among them, the overall network delay and throughput will remain unsatisfactory. Considering multiple temporal characteristics of an indoor wireless channel, we propose a multi-layer scheme to optimize resource allocation to the UEs in a cooperative multi-Base Station (BS) setup. We studied the performance of different resource allocation techniques to share the bandwidth effectively among UEs, rather than merely maximizing the total downlink bitrate. Our findings indicate that queue length-based allocation strategies yield the lowest delay under different downlink traffics and UE/BS deployment densities. Moreover, instead of serving all UEs simultaneously in each channel block, selectively serving a subset of UEs at each interval significantly enhances network delay and throughput. To further improve resource utilization and overall performance, we introduce an extension called sub-coherence time allocation. This technique considers early downlink queue exhaustion and speculatively calculates multiple digital precoders to be activated sequentially within a channel block. Simulation results demonstrate that this approach improves delay and jitter with minimal computational overhead, achieving a 25\% and 15\% decrease in digital and hybrid modes respectively.

Next-generation Internet will require strict end-to-end delay guarantees to support upcoming latency-sensitive applications. The IEEE 802.1 Time-Sensitive Networking (TSN) standard has become the de-facto solution for Ethernet-based L2 networks to support applications with strict latency, jitter and packet loss requirements. The IETF DetNet Working Group tries to expand on TSN to support real-time applications over larger-scale L3 networks. This paper proposes control and routing strategies that provide latency guarantees in L3 networks without requiring time synchronization among nodes. The proposed strategies include a link-state routing protocol and several exploration-based protocols that exploit queue-level information and network calculus to provide latency guarantees. Additionally, the use of queueing delay budgets enables independence among flows, while enabling fine-grained routing. This allows to make better routing decisions and to support applications with diverse latency requirements. Moreover, traffic shaping is only required at the network ingress. The strategies are evaluated extensively and compared in a simulation environment in multiple large-scale scenarios, considering acceptance rate, network utilization, path dissemination time, control overhead, and memory consumption, as well as how these metrics evolve w.r.t. different network scales. Experimental results demonstrate that representative delay-constrained traffic demands can be accommodated adequately by queue-level link-state routing protocols only in smaller-scale networks. In larger-scale network scenarios, breadth-first exploration-based protocols are required to provide stable performance w.r.t. acceptance rate and path dissemination times at the cost of only linearly increasing control overhead and memory footprint.