Fault Adaptive Routing in Metasurface Controller Networks

Taqwa Saeed, Constantinos Skitsas, Dimitrios Kouzapas, Marios Lestas, Vassos Soteriou, Anna Philippou, Sergi Abadal, Christos Liaskos, Loukas Petrou, Julius Georgiou, Andreas Pitsillides

Abstract

HyperSurfaces are a merge of structurally reconfigurable metasurfaces whose electromagnetic properties can be changed via a software interface, using an embedded miniaturized network of controllers, thus enabling novel capabilities in wireless communications. Resource constraints associated with the development of a hardware testbed of this breakthrough technology necessitate network controller architectures different from traditional regular Network-on-Chip architectures. The Manhattan-like topology chosen to realize the controller network in the testbed under development is irregular, with restricted local path selection options, operating in an asynchronous fashion. These characteristics render traditional fault-tolerant routing mechanisms inadequate. In this paper, we present work in progress towards the development of fault-tolerant routing mechanisms for the chosen architecture. We present two XY-based approaches which have been developed aiming to offer reliable data delivery in the presence of faults. The first approach aims to avoid loops while the second one attempts to maximize the success delivery probabilities. Their effectiveness is demonstrated via simulations conducted on a custom developed simulator.

Introduction: The HyperSurface Challenge

Metasurfaces are planar structures capable of manipulating electromagnetic waves for applications like beam steering or perfect absorption. The HyperSurface (HSF) represents the next evolution: a metasurface where these properties are software-controllable via an embedded network of miniaturized Application-Specific Integrated Circuits (ASICs).

Figure 1: The HyperSurface architecture, showing the integration of the control layer (ASIC network) with the electromagnetic function layer.

Building this control network (CN) presents significant engineering hurdles. Unlike standard Networks-on-Chip (NoC), HSF networks are inherently asynchronous to prevent electromagnetic interference from a global clock. They are also resource-constrained, as the tiny meta-atoms leave minimal space for complex buffers or routing tables. Furthermore, they utilize an irregular Manhattan-like topology and are highly fault-prone due to the density and scale of the controller array. Traditional fault-tolerant routing is too "heavy" for this environment, necessitating lightweight alternatives.

Proposed Routing Methodologies

The research proposes two fault-adaptive routing schemes derived from classic XY routing, where packets move horizontally then vertically. These schemes are designed to fit the specific pin-out and communication constraints of the HSF controllers.

Figure 2: Controller chip architecture including pin allocation for asynchronous communication handshakes.

1. Loop-Free Algorithm (LFA)

LFA is a reactive approach. Packets initially follow XY routing but switch to YX routing upon encountering a faulty link. To prevent infinite loops (live-locks), the algorithm employs "turn prevention" rules that restrict specific 90-degree turns near faults, effectively breaking potential cycles.

2. Reliable Delivery Algorithm (RDA)

RDA is a proactive approach focused on high reliability. It identifies two disjoint paths—one clockwise and one counter-clockwise—between nodes. If the primary path is obstructed, the packet is rerouted via the alternative non-overlapping path.

To evaluate these, the authors developed the HSF CN Asynchronous Simulator (HCNAS) on the AnyLogic platform, which models the four-way handshake protocol required for asynchronous ASIC communication.

Key Insights and Results

  • The Deadlock Dilemma: Standard XY routing can cause deadlocks in HSF topologies due to the interaction between data and Acknowledgement (ACK) packets. The authors resolved this using an "orientation-aware" variant of XY routing.
  • RDA Performance: The Reliable Delivery Algorithm achieved over 98% successful delivery at a node failure probability ($P_f$) of 0.01. Success remained above 90% until $P_f$ exceeded 0.04.
  • LFA Effectiveness: In single-fault scenarios, LFA achieved a 100% success rate, navigating isolated failures without entering loops.
  • Spatial Sensitivity: Success rates drop for nodes further from the Gateway (the "Fourth Quarter") because the probability of encountering a fault increases with path length.

Limitations and Future Work

While promising, the current Loop-Free Algorithm is optimized primarily for single-fault scenarios; its performance in high-density fault clusters requires further investigation. Future work will focus on more sophisticated fault models, hardware validation on physical HSF prototypes, and providing mathematical guarantees for deadlock and livelock freedom in increasingly irregular topologies.

Summary updated

Abstract

HyperSurfaces are a merge of structurally reconfigurable metasurfaces whose electromagnetic properties can be changed via a software interface, using an embedded miniaturized network of controllers, thus enabling novel capabilities in wireless communications. Resource constraints associated with the development of a hardware testbed of this breakthrough technology necessitate network controller architectures different from traditional regular Network-on-Chip architectures. The Manhattan-like topology chosen to realize the controller network in the testbed under development is irregular, with restricted local path selection options, operating in an asynchronous fashion. These characteristics render traditional fault-tolerant routing mechanisms inadequate. In this paper, we present work in progress towards the development of fault-tolerant routing mechanisms for the chosen architecture. We present two XY-based approaches which have been developed aiming to offer reliable data delivery in the presence of faults. The first approach aims to avoid loops while the second one attempts to maximize the success delivery probabilities. Their effectiveness is demonstrated via simulations conducted on a custom developed simulator.
BibTeX
@inproceedings{saeed2018fault,
  author = {Taqwa Saeed and Constantinos Skitsas and Dimitrios Kouzapas and Marios Lestas and Vassos Soteriou and Anna Philippou and Sergi Abadal and Christos Liaskos and Loukas Petrou and Julius Georgiou and Andreas Pitsillides},
  title = {{Fault Adaptive Routing in Metasurface Controller Networks}},
  booktitle = {NoCArc@MICRO},
  year = {2018},
  url = {https://openreview.net/forum?id=0kIY4xwjhf},
}
Figure