The Tech - Online EditionMIT's oldest and largest
newspaper & the first
newspaper published
on the web
Boston Weather: 73.0°F | Mostly Cloudy

FMX technology sparks dispute between Bose and developers

By David P. Hamilton

The effectiveness of a new FM broadcasting technology has become the flashpoint of a dispute between Amar B. Bose, MIT professor and founder of the Bose Corporation, and Broadcast Technology Partners, a Connecticut firm that has developed and is marketing the technology, known as FMX.

According to studies published by its inventors, FMX offers reduced noise levels, extended stereo separation, and a longer range than traditional FM broadcasting when received by special FMX equipment. Conventional FM receivers can still receive the FMX signal, although without any performance improvement.

Bose and an engineer from the Bose Corporation, William Short, reached the opposite conclusion. According to their results, FMX should seriously degrade the quality of stereo reception whether received by FMX equipment or not.

The disagreement erupted into public dispute when Bose and Short presented a theoretical and experimental analysis of FMX on Jan. 25. Before beginning his talk, Bose told the audience that he and Short had received messages from Emil Torick, president of BTP and co-inventor of FMX, which threatened "great personal liability" if they proceeded with their talk.

Bose, reached in Hawaii en route to Tokyo, said he has not heard from BTP or its lawyers since his talk. He suggested that BTP was using the threat of lawsuits to stifle criticism of the FMX system.

Bose's half of the study was based upon an original mathematical analysis of the FM phenomenon known as multipath. FM signals are limited in range by their tendency to bounce off objects between transmitter and receiver, a process which delays some signals and causes destructive interference at the receiver.

When multipath affects a stereo FM transmission, reception is likely to fade in and out or to switch over to monophonic FM, which is less susceptible to the effect because it is modulated at lower frequencies.

According to Bose's mathematical analysis, FMX should suffer from even worse multipath problems than traditional FM. Part of the trouble is that FMX, which includes an additional high-frequency signal to improve the signal-to-noise ratio, generates more high-frequency energy which in turn leads to more multipath distortion and noise.

The end result is that FMX, which outperforms traditional FM under laboratory conditions where no multipath is present, actually degrades broadcast coverage area and signal quality in actual use, Bose said.

To illustrate the point, he and Short conducted a simulation using the MIT radio station WMBR and a mobile receiver. By driving a test vehicle through Boston and surrounding suburbs, Short was able to collect data comparing the reception obtained with conventional FM against FMX on both standard and FMX receivers.

Short's results demonstrated that not only did multipath effects degrade the FMX signal far more than conventional FM, but that use of the FMX receiver distorted the signal even further.

In the question-and-answer session which followed the Bose-Short talk, Torick stood and denounced Bose's and Short's analysis. Drawing upon failed predictions in the history of science as examples, Torick claimed that Bose's "beautiful mathematics" failed to match reality and that WMBR was a "contaminated petri dish" for the broadcast experiments.

The WMBR transmission equipment was "improperly adjusted" and used an unapproved prototype FMX decoder, according to BTP literature.

The same literature also claims that over 100 radio stations nationwide had committed to FMX by 1988. Existing FMX broadcasters have achieved 125,000 hours of air time with "no listener complaints," Torick said at the lecture.

According to a UPI report, Torick told journalists after the presentation that Bose's and Short's work was supported by "a receiver manufacturer who opposed the new technology." The research was supported by the Bose Corporation, which manufactures quality speaker systems, amplifiers, and sound systems.

Bose discounted Torick's charges regarding WMBR, replying that the station was "checked out thoroughly" before the experiments began. He also claimed that three Chicago area radio stations had tried and rejected FMX because of listener complaints, although the owner of one of the stations, WNIB, said he'd received no customer complaints and had disconnected FMX only because few listeners in his area owned FMX receivers.

The Bose Corporation has no financial interest in seeing FMX fail, Bose said. In fact, Bose first became interested in studying FMX transmission when Short suggested that the Bose Corporation might wish to manufacture FMX receivers, he said.

The results of Bose's and Short's analysis will be submitted to a reputable journal in the near future, Bose said. He declined to identify the journal, citing worries that BTP might try to interfere in the publication process.