There are over five billion mobile phone subscribers, and an even greater number of wireless devices competing for access in the unlicensed bands. Every tech giant--from Amazon to ZTE--wants to be the leading force in mobile platforms, networks, and cloud services in what could be a winner-takes-all contest.
Unfortunately, spectrum crowding threatens dreams of a future filled with ubiquitous access to content and services. As connectivity is extended to “things” (both indoors and outdoors) and consumers get their first taste of ultra-high definition TV and broadcasts in virtual reality, the growth in demand for wireless access could overwhelm spectrum resources.
The rapid rise of devices with multiple radios compounds the problem. Smart phones, whole-home/small enterprise Wi-Fi systems, and smart home hubs are just three examples of devices that support multiple wireless standards and frequency bands--and are routinely expected to operate two or more radios at the same time. (For instance, when you are talking on your cell phone via a Bluetooth headset.) In today’s environment, wireless gadgets not only interfere with each other, they interfere with themselves.
Controlling interference is the key to wireless growth
Finding new spectrum is almost impossible. However, a new technology called “self-interference cancellation” (SIC) that emerged from Stanford University and is being developed by companies including Huawei, Kumu Networks, and Qualcomm could enable more flexible use of existing spectrum.
Studies show that most of the radio spectrum is idle at any given time. However, due to the limitations of conventional radio filters, users at or near the same location are kept far apart in frequency. Perfectly good channels are left unused, serving as “guard bands” between users.
SIC was originally developed to do what seemed impossible: enable radios to transmit and receive on the exact same frequency at the exact same time. That could effectively double the capacity of the radio spectrum. However, it gradually became apparent that there are more compelling applications, such as eliminating the need for guard bands.
Here’s a closer look at how SIC could enhance spectrum use in the two biggest wireless industries:
The growing use of unlicensed devices is a tragedy of the commons waiting to happen.
Users today want Wi-Fi coverage throughout small enterprises and homes. However, wireless routers often don’t have sufficient range. Fortunately, there are now multi-unit Wi-Fi systems that employ mesh technology to achieve broader coverage. The best performance is provided by “tri-band” units with one 2.4 GHz radio for talking to older Wi-Fi devices, one 5 GHz radio for talking to the latest Wi-Fi devices, and a second 5 GHz radio reserved for communication between units strategically-located around the home or small enterprise.
Conventional radio filters impose limitations. Today’s tri-band Wi-Fi units typically use the two channels at opposite ends of the 5 GHz band (with the frequencies in-between serving as a guard band). Unfortunately, when there are only two channels to choose from and two radios, everyone ends up using the same pair of channels, making it hard to avoid interference between neighboring networks.
SIC would enable tri-band Wi-Fi units to use any pair of the six channels in the 5 GHz band. When you have six channels to choose from, it’s possible to coordinate use to prevent interference between neighboring networks.
A group has been formed within IEEE 802.11 to discuss the technology--they refer to it as “full duplex”--and how it might be added to Wi-Fi standards.
Smart homes and buildings are finally gaining market traction. They are frequently built around smart hubs with multiple radios to talk to different wireless standards. Today, the radios must take turns communicating so they don’t interfere with each other. SIC would enable them to communicate simultaneously. That’s important: a security sensor should not have to wait for a status report from a thermostat.
The wireless industry is always looking for more spectrum, and spectrum sharing schemes are often the best hope. For instance, “LTE-Unlicensed” has been developed to give mobile operators the option of using unlicensed spectrum in specific locations.
Spectrum sharing has traditionally depended on the “Listen before talk” (LBT) protocol. LBT prevents a device from transmitting on a channel that is already in use. However, if two devices decide at the same time to transmit on a quiet channel, a collision occurs. Detecting and resolving collisions wastes time.
SIC enables an improved protocol: “Listen while talking.” Collisions can be detected immediately and resolved that much faster. Listen-while-talking could be a boon for not only LTE-Unlicensed but the new 3.5 GHz Citizens Broadband Radio Service (CBRS) and the “White Spaces” spectrum being reclaimed from over-the-air TV broadcasting.
Wireless operators could achieve a dramatic increase in capacity by upgrading to 5G, enabling them to compete with cable operators and telcos providing fixed broadband services to homes. However, this will require deploying hundreds of thousands of small cells, each requiring a high-speed link back to the network. Most small cells are expected to be connected via fiber optic cable, but running fiber isn’t always practical.
With SIC, operators can reuse the frequencies employed by the small cell to communicate with local customers for backhaul. The backhaul radio us equipped with a directional antenna pointed at the appropriate network node. If nothing more were done, the backhaul and small cell radios would interfere with each other. Using SIC, the backhaul radio’s transmit signal is cancelled out at the small cell’s receiver, and the small cell’s transmit signal is cancelled out at the backhaul radio’s receiver. The interference vanishes.
The reuse of access frequencies for backhaul is planned for Release 15 of the 3rd Generation Partnership Project’s 5G standards, with SIC an allowed implementation. SIC was successfully field-tested in a 4G LTE system by Telecom Italia Mobile.
The above are just a few examples of how SIC technology can overcome the limitations of conventional filters and the listen-before-talk protocol. They aren’t making any more spectrum, so eliminating guard bands and enabling more dynamic spectrum sharing would be major advances.
Some engineers dream of “cognitive radios” that automatically detect and move to vacant channels. Self-interference cancellation may be the missing piece to that puzzle.
This article is published as part of the IDG Contributor Network. Want to Join?