Chirps, Clicks, and Neural Nets: Securing Underwater Operations With LPI/LPD ACOMMS

In modern military operations, the ability to communicate without being detected, intercepted, or exploited is paramount. Whether coordinating movements, issuing commands, or transmitting time-sensitive data, unsecured communication can compromise entire missions, with consequences that ripple from the tactical to the strategic level of war. For decades, painstaking effort has been devoted to developing low probability of intercept (LPI) and low probability of detection (LPD) techniques to mask the signatures of radar transmissions and radio frequency (RF) communication, and these have become fundamental to operations across the land, surface, air, and space domains. Today, as strategic competition increasingly extends into the undersea and seabed environments, similar approaches are being adapted for secure underwater communication.

The undersea domain is a profoundly challenging communication environment. Standard electromagnetic RF transmissions, like those underpinning land, air, space and ocean surface operations, attenuate rapidly in seawater, preventing them from crossing the air-sea interface or propagating effectively through the water column. With the proliferation of uncrewed underwater vehicles and distributed undersea systems, there is now a rapidly growing need for robust, reliable, and secure subsea communication technologies. Two primary modes are currently in use: optical communication (OCOMMS) which uses blue-green laser light to achieve extremely high bandwidth (megabits or even gigabits per second) but requires short-range line-of-sight between transmitter and receiver, and acoustic communication (ACOMMS), which encodes information onto sound waves that can travel over long distances, often tens of kilometers or more, and does not require line-of-sight. While it offers much lower bandwidth (typically kilobits per second), ACOMMS can function across a range of depths and environmental conditions where OCOMMS is impractical, making it the most widely used method for underwater communication in defense, scientific, and commercial applications.

But as with RF, without proper safeguards, ACOMMS transmissions operate “in the clear,” and are therefore vulnerable to detection and interception by eavesdropping adversaries. This could jeopardize a vital mission, not only by revealing the transmission’s contents, but also by exposing the location or presence of a submarine, UUV, diver, or seabed installation. LPI/LPD ACOMMS strategies are specifically tailored to the unique undersea operating environment, employing techniques that blend in, evade, or deceive to make acoustic signals difficult to detect, locate, or decode. But even though ACOMMS and RF transmissions operate in fundamentally different environments, their shared tactical requirement—minimizing detection, recognition, and exploitation—has led to the adoption of parallel LPI/LPD techniques. Both utilize irregular, burst transmissions, as well as beamforming—the directional control of signal transmission—to reduce their signature. With RF, directional antennas or phased arrays limit signal exposure, while in ACOMMS, directional transducers and narrow-beam acoustic projectors serve the same purpose, confining transmission to a specific volume of water. Power control, another common strategy, involves adjusting transmission strength to maintain signal links while minimizing leakage.

The acoustic domain has also adapted other RF-derived LPI/LPD methods as well. Spread-spectrum techniques, long used in military RF systems to distribute a signal’s energy across a wide range of frequencies to make it harder to detect, are implemented in ACOMMS using underwater transducers that modulate signals across multiple frequency bands or pseudorandom sequences. Chaotic Direct Sequence Spread Spectrum (Chaotic DSSS) spreads a signal across a wide range of frequencies using a sequence that looks random but is generated from shared starting conditions, similar to an encryption key, so only a receiver with the same setup can decode it. Frequency Hopping Spread Spectrum (FHSS) involves rapidly switching between different frequencies in a pseudo-random pattern, a technique that is widely used in military communications across warfare domains to prevent jamming and interception. Chirp Spread Spectrum (CSS) changes a signal’s frequency over time to help the receiver distinguish between direct and reflected signals, helping to reduce the effects of multipath delay spread—the distortion caused when signals take multiple paths through the water column. Another method, Time-Reversal Mirroring (TRM), involves retransmitting a received signal backwards into the water column, focused on the original point of transmission. Repeating this process creates an increasingly narrower beam that improves signal-to-noise ratio (SNR) at the receiver, and minimizes sound propagation in unintended directions.

While traditional spread-spectrum techniques dominate LPI/LPD strategies, artificial intelligence (AI) and machine learning (ML) are driving the development of adaptive, “cognitive” communication systems. AI/ML-enabled ACOMMS can dynamically adjust waveforms in response to jamming, interference, or changes in sound propagation conditions. Neural networks can be trained to identify the unique acoustic properties of underwater channels and mitigate their effects during signal reception. They can also identify patterns in time-series data and optimize transmission parameters, such as power, timing, and modulation, to maintain covert, reliable links in complex environments.

One of the most intriguing and promising areas of LPI/LPD ACOMMS is biomimicry—emulating the natural communication strategies of marine life to enhance stealth, security, and efficiency. Cetaceans such as dolphins and whales use highly directional, frequency-modulated clicks, whistles, and chirps that vary in timing and structure. Fish produce bursts of random, irregular choruses for communication or mating, while invertebrates like snapping shrimp produce chaotic, broadband pulses that dominate the ambient soundscape. These natural patterns have inspired biomimetic signaling schemes that hide in plain sight, blending with the biotic background noise of the ocean and benefitting from acoustic camouflage. Chaotic DSSS was inspired by the unpredictability of biological signals, mimicking dolphin clicks or whale calls. Time-hopping and randomized transmission techniques mimic irregular call patterns found in nature, preventing predictable signal structures. And adaptive modulation schemes can shift in response to environmental noise, much as marine mammals adjust their calls in noisy waters.

Although promising, LPI/LPD ACOMMS faces significant challenges. Its performance depends heavily on environmental factors which influence how sound travels underwater. Bathymetry, salinity, and temperature can cause spreading loss and absorption loss. Motion-induced Doppler shifts degrade signal coherence, and dynamic ocean environments can create time-varying multipath interference. Natural and man-made ambient noise can mask or corrupt transmissions, and random transient noises can cause random signal fluctuations. Matching biomimetic signals to complex ocean environments also requires sophisticated modeling and advanced signal processing. And ACOMMS in general suffers from inherent latency and low bandwidth, though research and development efforts continue to push these limits.

In the rapidly evolving undersea battlespace, those who can transmit clearly, stealthily, and securely will hold a decisive advantage. LPI/LPD ACOMMS is a critical enabler of survivability and mission success in the world’s most challenging communication environment. Continued advances in signal processing, AI-driven adaptation, and biomimetic design will push the boundaries of what is possible in underwater communication, offering ever more sophisticated ways to blend into the ocean’s complex acoustic landscape.