LTE-U utilizes time sharing using periodic time slots. Since the transmissions are scheduled, the issue is whether LTE-U will add interference and increase collisions for WiFi. It would seem to ignore WiFi's transmission window by taking control over the channel. This ignores the principle of fair-sharing, ABI Research said recently.
At SCTE Cable-Tec Expo in New Orleans last week, Belal Hamzeh, CableLabs' director of network technologies, summed up the LTE-U interference problem neatly: "It's like the bully at the playground. Nobody likes the bully."
The cellular industry, led by Verizon (NYSE:VZ), formed the LTE-U Forum to work on co-existence of the technologies, and the FCC opened a Public Notice on the topic. But is there any credence to the claims of interference? Joey Padden, lead architect at CableLabs, says yes. In a blog post, he detailed tests he ran with the help of 7Signal to determine whether the impact of a co-channel LTE signal on WiFi could be detected.
The first round of testing took place in the controlled environment of a CableLabs anechoic chamber, which is an RF environment used for wireless testing. After choosing a number of key performance indicators for the WiFi, it was determined that these KPIs were affected when LTE-U was present.
The second test was in the real-world environment of the CableLabs offices. Padden chose two duty cycles, 50% of 80 ms and 50% of 200 ms that served a channel with a single WiFi access point. Two power levels were used.
"We were able to discern a clear LTE 'signal' from many other KPIs," Padden said. "Some notable examples were channel noise floor and the rate of client retransmissions .... I (also) have a couple of anecdotal results to share. First, our WiFi controller was continuously changing the channel on us, making testing a bit tricky. I guess it didn't like the LTE much."
Padden also noted that several CableLabs employees, upon finding out about the test after the fact, said, "Oh, that's why my WiFi has been acting up." He also pointed to a Google (NASDAQ:GOOG) white paper that similarly found a problem with WiFi when the LTE power level at an access point is below the WiFi energy detection level.
"It is looking pretty clear that we can detect and measure the impact of LTE on WiFi in the wild if need be," Padden said. "But again, with continued efforts by the WiFi community to help develop fair sharing technologies for LTE to use, it won't come to that."
Upon the release of a technical report evaluating use of LTE-U, Verizon says the LTE-U Forum "looks forward to working with the unlicensed community to ensure consumers can choose the best connectivity available to meet their needs without negatively impacting other unlicensed users." However, Jennifer Anreoli-Fang, principal architect at CableLabs, said standards bodies haven't been included in the LTE-U "club."
There are different approaches to enabling LTE in the unlicensed band. The cellular industry standards body (3GPP) has been working on Licensed-Assisted Access (LAA) and likely will adopt the coexistence measure, "listen before talk" (LBT), which is commonly used elsewhere in the world. "(This) would ensure compatibility between LAA and WiFi - at least in Europe and Japan. We welcome the collaborative process within 3GPP," Anreoli-Fang said, noting that LTE-U, on the other hand, is the non-standard U.S. version and is being developed outside of the 3GPP standards.
LTE-U Forum coexistence requirements calls for duty cycling when other operators are present, but does not include sharing fairly to avoid interrupting WiFi mid-stream. "The fundamental problem here is that LTE-U does not 'listen before talk,' a most basic politeness protocol; instead LTE-U 'listens and talks anyway,' regardless of whether somebody else is talking or not," Anreoli-Fang said.