Post-Quantum Cryptography in Practice: Sector-Specific Deployment and Integration Patterns

From Theory to Sector-Specific Engineering

Standards are finalised. Algorithms are selected. Nobody is deployed.

That gap is what this article maps. The theoretical foundations and migration pathways are documented elsewhere in this series, but documentation is not deployment. An IoT sensor running on 64 KB of RAM, a cloud CSPM platform processing terabytes of security telemetry, an energy grid engineered to operate for 30 years, a vehicle OTA channel pushing firmware at motorway speeds: each of these operates under constraints so different from the others that “deploy PQC” is not one engineering problem. It is a family of problems sharing an algorithm portfolio.

What follows is a sector-by-sector examination of how those constraints shape integration decisions. One finding needs to be stated upfront, because it colours everything that follows: no reviewed paper reports a live production deployment. Every result is from a simulation, a prototype, or a conceptual framework. The publication volume is impressive. The operational footprint is zero.

Convergent Pattern: Kyber/Dilithium Dominance

One pattern is universal across every applied paper reviewed here: CRYSTALS-Kyber (ML-KEM) for key encapsulation, CRYSTALS-Dilithium (ML-DSA) for digital signatures. Full stop. No applied paper evaluates or deploys a non-lattice primary PQC algorithm. The uniformity makes sense (NIST standardisation, library availability, solid performance characteristics), but it also creates a concentration risk. None of the papers acknowledge this.

This convergence is pragmatic. Kyber and Dilithium offer the best performance-to-security ratio on commodity hardware, and NIST endorsement removes procurement and compliance friction. But there is a systemic risk that none of the reviewed papers acknowledge: the entire applied PQC deployment ecosystem sits on a single mathematical family. Lattice problems. If a lattice-breaking technique emerges, whether from quantum or classical cryptanalysis, the entire portfolio is compromised at once. Not just one sector. All of them.

Strategic implication: Crypto-agility at the deployment architecture level, not just the library level, is the mitigation. None of the reviewed implementations demonstrate it.

IoT and Blockchain: Quantum-Resistant Distributed Trust

IoT systems face a compound security problem. The devices are resource-constrained. The communication channels are wireless and often unauthenticated at the physical layer. The trust models tend toward centralisation, which creates single points of failure that a sufficiently motivated attacker can target. Several research groups have proposed blockchain as a decentralised trust anchor, with PQC protecting both on-chain and off-chain operations [1] [6] [2].

Fahomida et al.: CertiSense-PQ

CertiSense-PQ provides a quantum-resistant certificate management framework for IoT [1]. Certificates are issued and revoked on a distributed ledger (removing the CA single-point-of-failure), device certificates are signed with Dilithium, and session keys are exchanged via Kyber. The consensus mechanism is adapted for IoT networks where full Proof-of-Work would be computationally absurd. CertiSense-PQ provides a quantum-resistant certificate management framework for IoT [1]. Certificates are issued and revoked on a distributed ledger (removing the CA single-point-of-failure), device certificates are signed with Dilithium, and session keys are exchanged via Kyber. The consensus mechanism is adapted for IoT networks where full Proof-of-Work would be computationally absurd.

The performance cost is not negligible: certificate verification latency increased by roughly 40% compared to ECDSA-based certificates. For non-real-time IoT applications (environmental monitoring, asset tracking), that overhead sits within acceptable bounds. For anything with hard latency requirements, it might not. The performance cost is not negligible: certificate verification latency increased by roughly 40% compared to ECDSA-based certificates. For non-real-time IoT applications (environmental monitoring, asset tracking), that overhead sits within acceptable bounds. For anything with hard latency requirements, it might not.

Rajkumar et al.: Blockchain Authentication

Rajkumar et al.: Blockchain Authentication

Rajkumar et al. combine lattice-based PQC with blockchain for IoT device authentication and report 97.8% accuracy in device identity verification [6]. The paper focuses on the authentication protocol design; the underlying PQC performance characteristics are not benchmarked independently. Rajkumar et al. combine lattice-based PQC with blockchain for IoT device authentication and report 97.8% accuracy in device identity verification [6]. The paper focuses on the authentication protocol design; the underlying PQC performance characteristics are not benchmarked independently.

Huang et al.: Confidential Smart Contracts

Huang et al. address a more specific problem: executing smart contracts on quantum-vulnerable blockchains while maintaining input and output confidentiality [2]. Their framework uses lattice-based encryption for contract confidentiality, threshold PQC for multi-party execution, and PQC-compatible zero-knowledge arguments for on-chain verification.

The practical motivation is straightforward. If smart contracts manage property titles, derivatives, or insurance policies with lifetimes measured in decades, quantum resistance is not a future design consideration; it is a present one. Whether this particular framework survives contact with production smart contract platforms remains to be seen.

Energy Grid: The Most Urgent Case

Krishnan et al. make a case that the power grid is uniquely vulnerable to quantum attack, and it is a persuasive one [4]. Three factors converge. Grid infrastructure deployed today must remain secure for 30 years or more, which puts it well inside any plausible quantum timeline. SCADA and ICS protocols use authentication mechanisms designed in an era when computing threats to physical infrastructure seemed remote. And compromise in a power grid is not a data breach; it cascades into physical damage across interconnected networks.

The STRIDE-Guided Assessment

Krishnan et al. apply Microsoft’s STRIDE threat model (Spoofing, Tampering, Repudiation, Information disclosure, Denial of service, Elevation of privilege) to map quantum threats against power grid operations:

Smart Grid Architecture Recommendations

The paper proposes PQC integration at three grid layers: PQC-authenticated SCADA commands between control centres and generation units at the generation layer; PQC-encrypted inter-substation communication replacing IEC 62351 classical profiles at the transmission layer; and PQC certificate management for smart meters and distributed energy resources at the distribution layer. The paper proposes PQC integration at three grid layers: PQC-authenticated SCADA commands between control centres and generation units at the generation layer; PQC-encrypted inter-substation communication replacing IEC 62351 classical profiles at the transmission layer; and PQC certificate management for smart meters and distributed energy resources at the distribution layer.

Gap noted: Krishnan et al. provide no performance benchmarks for PQC on actual SCADA hardware. This is a significant omission. SCADA systems typically run on embedded processors with limited computational headroom, and the question of whether Kyber key encapsulation or Dilithium signature verification can complete within real-time control loop deadlines remains unanswered.

Automotive: Firmware Security Over Vehicle Lifetimes

Nakka et al. focus on a specific, high-consequence problem: securing over-the-air firmware updates for connected vehicles [5]. A compromised OTA channel is the highest-severity attack in automotive cybersecurity because it enables remote vehicle takeover at scale. Current OTA channels depend on RSA/ECDSA code signing and TLS key exchange. Both are quantum-vulnerable, and vehicles designed today will be on roads for 15 to 20 years.

Automotive-Specific Constraints

Key finding: Kyber + Dilithium fit within automotive ECU constraints. SPHINCS+ (larger signatures) is acceptable as a conservative backup for vehicles with longer planned lifespans, at the cost of increased OTA bandwidth consumption.

Cloud Security: Quantum-Resistant Posture Management

Joseph et al. take a different angle entirely [3]. Rather than replacing encryption in transit, they propose integrating PQC into cloud security posture management (CSPM) systems: Kyber-encrypted threat intelligence sharing between agents across multi-cloud environments, PQC-authenticated security events for quantum-resistant audit trails, and LWE-derived feature extraction for detecting cryptographic downgrade attacks.

Cloud environments face fewer resource constraints than IoT or automotive, so the mechanical difficulty of deploying PQC is lower. The challenge is operational. Coordinating PQC across multi-cloud CSPM agents means getting multiple cloud vendors to agree on implementation details, interoperability testing, and timeline. That kind of coordination does not exist at scale today.

Multi-Cloud Integration Architecture

Unconventional Applications

Two papers push PQC into domains that rarely appear in migration discussions.

Raju et al. apply Kyber-1024 encryption to data encoded in synthetic DNA sequences [7]. The use case has its own logic. DNA storage offers extreme longevity (thousands of years) and extreme density (1 exabyte per cubic millimetre). If the storage medium will outlast the encryption, and it will, the data is eventually exposed unless the encryption is quantum-resistant from the start. Their approach encrypts data before DNA encoding and stores decryption keys in a separate PQC-protected key management system.

Zhao et al. apply lattice-based PQC to covert communications: legitimate-looking transmissions that conceal hidden messages [8]. They replace classical Diffie-Hellman key exchange with a lattice-based equivalent, preserving covertness properties while adding quantum resistance. The applications (censorship circumvention, diplomatic channels, intelligence) are obvious. Less obvious is that a quantum adversary breaking classical covert channels would expose not just the message content but the very existence of the hidden communication. That is a different threat profile from ordinary encryption failure.

Cross-Sector Analysis

What Works

Kyber and Dilithium as default selection is a pragmatic choice backed by NIST standardisation, library availability, and competitive performance characteristics. Hybrid migration strategies (dual classical + PQC during transition) maintain backward compatibility without forcing a cliff-edge cutover. Blockchain-anchored PKI removes certificate authority single points of failure in IoT architectures where centralised trust is a design liability.

What Is Missing

The Prototype-to-Production Gap

This is the single most important finding across the sector-specific literature, and it deserves direct language. Every paper reviewed here presents a conceptual framework, a simulation, or a small-scale prototype. None reports a production deployment. No real users, no real adversaries, no real operational constraints.

The gap is not unique to PQC; the field received its first formal standards only in 2024. But the implication for practitioners is that performance claims, security assurances, and integration architectures from these papers have not been stress-tested in environments where things actually break.

Takeaways

Kyber and Dilithium dominate every sector examined here. The reason is straightforward: NIST endorsement, library availability, and competitive performance on commodity hardware. But universal adoption of a single mathematical family creates a concentration risk that the applied literature has not confronted. If lattice assumptions weaken, every sector loses its PQC portfolio at once. Crypto-agility at the architecture level, not merely the library API level, is the only hedge, and none of the reviewed implementations provide it.

The prototype-to-production gap is the primary bottleneck of the field. IoT and blockchain generate the most proposals. Energy grids face the most acute urgency, with 30-year asset lifetimes that already exceed Mosca’s inequality for many data categories, yet they have the least validated benchmarks on actual SCADA hardware. Automotive OTA channels fit within ECU memory and latency constraints on paper, but vehicle lifespans of 15 to 20 years mean that PQC decisions made in the next few model years will determine whether those vehicles are quantum-vulnerable or quantum-resistant for their entire service life.

Algorithm selection is the solved problem. Integration architecture, constrained-device performance, and cross-vendor interoperability are not.

Applied Questions on PQC Deployment

Can IoT devices with limited memory run post-quantum cryptography?

Kyber-768 needs roughly 2.4 KB for public and private keys combined, which fits on most 32-bit microcontrollers. But “fits” is not the same as “works well.” No peer-reviewed benchmark validates PQC on production IoT hardware with real workloads. Devices below 32 KB of available RAM may need hardware-accelerated PQC, lighter constructions, or offloaded key operations.

Which sector faces the most urgent quantum threat?

Energy grid infrastructure. The combination of 30-year-plus operational lifetimes, cascading failure potential, and legacy SCADA authentication makes the case almost self-evident. Encrypted SCADA traffic captured today may well be decryptable within the operational lifespan of grid equipment being installed right now. Mosca’s inequality is already in the red for this sector.

Why does every applied PQC paper default to Kyber and Dilithium?

Incentive alignment. NIST standardisation brings library support, regulatory compliance, and interoperability expectations. Kyber and Dilithium also happen to perform well on commodity hardware. The downside is systemic: one mathematical family, one failure mode. If lattice assumptions weaken, every sector loses protection simultaneously.

Are there any production deployments of PQC?

Not among the papers reviewed here. Broader industry experiments exist (Google’s CECPQ2, Cloudflare’s post-quantum TLS rollout), but they sit outside the scope of this academic review. The applied research community is still at the conceptual and prototype stage. Production deployment with real adversaries and real operational constraints has not been reported in the peer-reviewed literature.

Technical Appendix