Q-Cyber Magazine
Secure communications for defence: Do not trust the path by default
Defence communications should be designed so that compromise of a relay, ground station or bearer does not, by itself, compromise the message, the key or the command.

A defence message is only as trustworthy as the protection it carries when the bearer is hostile, misconfigured or simply unknown. The path should be assumed untrusted; the cryptographic and operational design should make that assumption survivable.
The path is not the perimeter
Defence communications have always crossed ground they do not own: leased circuits, shared satellites, coalition networks, commercial ground stations, and, increasingly, software-defined infrastructure. NIST SP 800-207 was written for enterprise architectures, but its central claim travels well. Location on a "closed" network is not a sufficient reason to trust a session. Authentication and authorisation should be established for the parties and the resource, not inferred from the route.
CERT-In's space guidance makes the same point in a physical form. The threat list is not limited to stolen files. It includes spoofed signals, unauthorised commands and compromised ground stations. Segment-wise controls — for the spacecraft, the ground, the link and the terminal — exist because a failure in any one segment can undermine the others. Encryption, authentication, access control, intrusion detection and signed firmware are listed as the complementary measures.
Exhibit 1
Trust decisions in defence communications
Source: Qualitative framework derived from NIST SP 800-207 (August 2020); NSA, Post-Quantum Cryptography: A Digital Armor, 1 October 2026; CERT-In space and SatCom framework, February 2026. No incident counts are claimed.
Design consequences, kept at the right altitude
Three design consequences follow, without any need to describe a particular product.
First, keys and identities must belong to the communicating parties, not to the transport. A relay that cannot read the traffic and cannot mint a credible command is a relay that can fail without becoming a command-and-control breach.
Second, the cryptographic plan must be dated. Harvest-now-decrypt-later is already a reason to prefer post-quantum key establishment for traffic with a long confidentiality life. Trust-now-forge-later is a reason to plan the replacement of vulnerable signature schemes that underwrite devices, updates and operators. NIST's FIPS 203, 204 and 205, and the 2026 transition report, exist for that work.
Third, operations must assume degraded bearers. Jamming and spoofing are availability and integrity problems. They are addressed by rehearsed fallback paths, authenticated sources and the refusal to treat an unexpected signal as authoritative.
Implications
- Write the trust boundary around the parties and the message, not around the network diagram.
- Inventory key establishment and signatures separately, and sequence migration by confidentiality life and command criticality.
- Extend the same discipline to satellite and commercial capacity, using CERT-In's segment model and reporting duties.
- Rehearse loss of a bearer. A path that has never been failed over is still a single point of failure.
What this means for defence
Secure communications are a systems problem: cryptography, identity, bearer diversity and incident reporting have to be designed together. Applications such as Knowlcom's Naad Secure sit in that landscape as one way to carry messaging, voice and video with contemporary cryptographic protection. The standard of judgement remains the same: if the path is compromised, does the command still hold?


