[Paper Review] FPGA Based Implementation of Distributed Minority and Majority Voting Based Redundancy for Mission and Safety-Critical Applications

Padmanabhan Balasubramanian, Nikos E. Mastorakis|arXiv (Cornell University)|Nov 29, 2016

Radiation Effects in Electronics8 references4 citations

TL;DR

This paper presents the first FPGA-based implementation of a distributed minority and majority voting redundancy (DMMR) scheme, which reduces area and delay overhead compared to conventional N-modular redundancy (NMR) in mission and safety-critical systems. DMMR achieves 44.5% and 56.5% lower area-delay product (ADP) than NMR for 3- and 4-module fault tolerance, respectively, across multiple FPGA platforms including radiation-tolerant devices.

ABSTRACT

Electronic circuits and systems used in mission and safety-critical applications usually employ redundancy in the design to overcome arbitrary fault(s) or failure(s) and guarantee the correct operation. In this context, the distributed minority and majority voting based redundancy (DMMR) scheme forms an efficient alternative to the conventional N-modular redundancy (NMR) scheme for implementing mission and safety-critical circuits and systems by significantly minimizing their weight and design cost and also their design metrics whilst providing a similar degree of fault tolerance. This article presents the first FPGAs based implementation of example DMMR circuits and compares it with counterpart NMR circuits on the basis of area occupancy and critical path delay viz. area-delay product (ADP). The example DMMR circuits and counterpart NMR circuits are able to accommodate the faulty or failure states of 2, 3 and 4 function modules. For physical synthesis, two commercial Xilinx FPGAs viz. Spartan 3E and Virtex 5 corresponding to 90nm and 65nm CMOS processes, and two radiation-tolerant and military grade Xilinx FPGAs viz. QPro Virtex 2 and QPro Virtex E corresponding to 150nm and 180nm CMOS processes were considered for the NMR and DMMR circuit realizations which employ the 4-by-4 array multiplier as a representative function module. To achieve a fault tolerance of 2 function modules, both the DMMR and the NMR schemes provide near similar mean ADPs across all the four FPGAs. But while achieving a fault tolerance of 3 function modules the DMMR features reduced ADP by 44.5% on average compared to the NMR, and in achieving a fault tolerance of 4 function modules the DMMR reports reduced ADP by 56.5% on average compared to the NMR with respect to all the four FPGAs considered.

Motivation & Objective

• To develop a more area- and cost-efficient redundancy scheme for mission and safety-critical systems compared to conventional NMR.
• To minimize design metrics such as area and delay while maintaining equivalent fault tolerance.
• To evaluate the performance of the proposed DMMR scheme across commercial and radiation-tolerant FPGAs.
• To demonstrate that DMMR can achieve comparable or better fault tolerance with significantly reduced hardware overhead.

Proposed method

• The DMMR scheme uses distributed voting logic to detect and correct faults across multiple function modules, replacing centralized redundancy with localized, distributed decision-making.
• The 4-by-4 array multiplier is used as the representative function module in the design and implementation.
• Physical synthesis is performed on four FPGA platforms: Spartan 3E (90nm), Virtex 5 (65nm), QPro Virtex 2 (150nm), and QPro Virtex E (180nm).
• The DMMR and NMR circuits are compared based on area occupancy and critical path delay, with the area-delay product (ADP) used as the primary metric.
• Fault tolerance is evaluated for up to 4 failed function modules, with voting logic designed to identify and mask faulty outputs using minority/majority logic.
• The implementation uses Xilinx ISE and Vivado tools for synthesis, fitting, and timing analysis across all FPGA types.

Experimental results

Research questions

• RQ1 Can the DMMR scheme achieve equivalent fault tolerance to NMR with reduced hardware resource usage in FPGA-based systems?
• RQ2 How does the area-delay product (ADP) of DMMR compare to NMR across different FPGA technologies and process nodes?
• RQ3 What is the performance gain of DMMR over NMR in terms of ADP when tolerating 3 or 4 faulty function modules?
• RQ4 Does the DMMR scheme maintain reliability and low delay in radiation-tolerant and military-grade FPGAs?
• RQ5 Can the distributed voting architecture reduce system weight and cost while preserving fault tolerance in safety-critical applications?

Key findings

• For 2-module fault tolerance, DMMR and NMR exhibit near-identical mean area-delay product (ADP) across all four FPGAs.
• For 3-module fault tolerance, DMMR reduces ADP by an average of 44.5% compared to NMR across all FPGA platforms.
• For 4-module fault tolerance, DMMR achieves an average ADP reduction of 56.5% compared to NMR on all four FPGAs.
• The DMMR scheme maintains consistent performance improvements across both commercial (Spartan 3E, Virtex 5) and radiation-tolerant (QPro Virtex 2, QPro Virtex E) FPGAs.
• The implementation confirms that DMMR is a viable and more efficient alternative to NMR for safety-critical systems, especially as fault tolerance requirements increase.
• The results demonstrate that distributed voting reduces redundancy overhead without compromising fault detection or correction capability.