Skip to content

Technologies

TrellisWare’s technologies and products all share a common foundation: using all the information available, in all of the processing, all of the time. While this may sound intuitively obvious, virtually every receive processing system designed in history breaks the process into pieces designed to operate independently or interconnected with simple feedback loops for sequential processing. TrellisWare applies likelihood techniques, such as joint maximum likelihood and adaptive iterative detection, to extract more information and use it globally, enabling our products to work where others do not.

TrellisWare Likelihood Processing: A Revolutionary Approach to Communication Design

When environments rarely change or they change slowly, sequential processing provides receive processing results that are good and may be sufficient. However when environments change abruptly – whether due to multipath, obscurants in the path, dynamics or the operating environment – traditional sequential feedback can actually make things worse, since the information is often out-of-date before it is used.

TrellisWare’s likelihood processing technologies (e.g., Per Survivor Processing [PSP] and Adaptive Iterative Detection [AID]) essentially remove the feedback while applying the information globally throughout the processing and avoid the potentially catastrophic failure modes of traditional sequential feedback approaches. TrellisWare has extended this philosophy across all of our products, in our FEC designs, our waveform designs and in our cross-layer MAC architectures.

The result is solutions that work where everything else fails, especially when degradation sources such as high dynamics, interference and multipath are severe and the communication system needs to adjust seamlessly to rapid changes in the operating environment. 

The reasons joint processing has not been traditionally applied are twofold:

  1. digital silicon capabilities, fueled by Moore’s Law, have only made such approaches conceivable, much less practical, over the past half decade, and as a result,
  2. most of today’s lead engineers have been trained to break the problem into pieces, processed sequentially, and have yet to embrace the full paradigm shift.

On the first point, it is still true that it is often impractical to fully implement a “brute force” global likelihood receive design and part of what sets TrellisWare apart is that we have developed critical insight and trade secrets into where and how to apply these techniques. On the second point, from its foundation TrellisWare has focused exclusively on likelihood processing approaches and techniques, and TrellisWare’s academic co-founders have played a significant role in bringing these concepts to light and keeping TrellisWare well ahead of the curve for commercial application.

PCCC Design by TrellisWare for FPGA - 10Mbps, Rate 1/3 Click to enlarge image


TrellisWare has utilized its Likelihood Processing Techniques to develop several unique processing implementations that significantly extend the performance of a wide variety communication products including:

Click to enlarge image

Harshness in an operating environment can take on many forms: increased multipath reflections, obstructions in the comms path line of sight, motion dynamics from walking around a corner to driving at high speed or moving from open to dense urban comms.

PrintClose

Harshness in an operating environment can take on many forms: increased multipath reflections, obstructions in the comms path line of sight, motion dynamics from walking around a corner to driving at high speed or moving from open to dense urban comms. Conventional receive solutions utilizing sequential processing or simple feedback mechanisms can perform well in benign environments, but degrade rapidly in the face of more difficult channel environments.

Shannon's First Lesson:
"Never discard information prematurely that may be useful in making a [signal processing] decision until after all decisions related to that information have been completed"
"Advances in digital communication in the latter half of this century were guided by the lessons of information teh cry but fueled by teh spectacular pregress in solid state electronics."
--Andrew Vitervi, 1996
"There is nothing more powerful than an idea whose time has come."
--Victor Hugo