Your browser is out-of-date!

Update your browser to view this website correctly. Update my browser now


Technique Aims to Reduce Multipath

Jeff Keith offers his findings on FM L+R/L–R multipath control

The first image is a graph of data extracted from the paper ‘The Psychoacoustics of Multichannel Audio’ by J. Robert Stuart. Jeff Keith commented, ‘I found this paper online about two years ago and was glad to see that his data pretty much matched what I discovered back in 1998.’ The second shows data gathered by Keith from 62 male and female test subjects in the summer of 1998. SSBSC on FM was a hot topic at last year’s NAB Show. Single-sideband suppressed carrier generation of FM stereo instead of the conventional DSBSC has shown promise in a number of markets as an effective means of reducing multipath distortion. Omnia, Orban and Wheatstone all offer the SSB option in their top models of FM audio processors. However field results have not been universally judged as providing meaningful improvement for many stations.

Jeff Keith, senior product design engineer at Wheatstone Corp., offers an alternative solution to the challenge of multipath reduction in his planned April 9 presentation at the Broadcast Engineering Conference.

Keith had evaluated and added SSBSC as a feature in the Wheatstone AirAura FM audio processor. “Field experience and feedback from customers using SSB-equipped processors from both Wheatstone and our competitors has been mixed,” he says. “A few claim it to be a miracle, a few claim it to be a detriment and the rest seemingly notice no difference between conventional DSBSC and SSBSC.”

Although Wheatstone still offers SSBSC as an option, Keith decided there was an entirely different FM processor design approach that provides more consistent multipath mitigation.

“The issue of minor receiver compatibilities aside, what stations really want in a multipath solution is something that is beneficial to all listeners and is compatible with all stereo receivers regardless of stereo decoder design. Combining intelligent stereo separation management with conventional DSBSC modulation achieves that goal.”

New perceptions

Keith built and marketed the SMO-900 Stereo Enhancer when stereo wars were heating up in the late 1980s. In 1998, as chief engineer at WMJI(FM) in Cleveland, he designed and built a processing device to preserve mono loudness when songs with ping-pong stereo were aired.

“Having had success with the SMO-900, I knew that excessive stereo enhancement reduced mono loudness. However it had not occurred to me until WMJI that the early ping-pong stereo recordings were in fact the same thing as ‘excessive stereo separation,’ a problem that I knew quite well how to solve.

“The Mono Compatibility Controller that I built for WMJI was designed to dynamically adjust stereo separation on the fly according to the program material being fed to it. Its sole purpose was to even out loudness on mono radios, nothing more,” he said.

He discovered that the “normal” stereo sound field could be reduced significantly without listeners noticing it.

“It didn’t take very long for the station’s staff to discover two new effects of the processor being in line. First, there was no noticeable change in stereo separation even though it was aggressively reducing separation on quite a lot of program material. Second was the effect on our perceived multipath — it was far less noticeable in ‘bad’ multipath areas and was virtually gone in others. This was a complete surprise to all of us and something that I never expected or even considered.”

Wheatstone has included the algorithm from Keith’s original Mono Compatibility Controller in the stereo generators of their FM processor models from the AP-2000 forward. “Over time, and so as not to give away exactly what it does or how it works to competitors, we’ve hidden its function under several different control names and have never explained in our documentation exactly what’s going on inside it.”

Keith cites feedback from many hundreds of customers over the past few years that reinforced what he found at WMJI. “Intelligent stereo separation management apparently works much better at reducing the audible effects of multipath-induced stereo blending than any other method except perhaps transmitting in mono,” he concludes. Further, “No special permission is needed from the FCC, and it is 100 percent compatible with every FM stereo receiver that was ever made.”

Tom McGinley is technical adviser and a longtime contributor to Radio World.