Using FM to Improve Wireless Networks
November 10, 2015 | Northwestern UniversityEstimated reading time: 3 minutes
One minute your wireless Internet is working fine. The next minute, it takes an infuriatingly slow five seconds to load a single Web page. You paid for the fastest Internet speed available but during these all-too-frequent times, it seems like the connection barely works at all.
"Most people think it's a mystery," said Aleksandar Kuzmanovic, associate professor of electrical engineering and computer science at Northwestern University's McCormick School of Engineering. "They get upset at their routers. But what's really happening is that your neighbor is watching Netflix."
Most people don't realize how much their neighbors' Internet networks interfere with their own, heavily affecting speed and performance. Unless a home is located in the middle of nowhere, it is likely that neighboring homes' Wi-Fi networks will bump into each other and prevent data from getting through. This is particularly true in large, urban apartment buildings where many people reside within a smaller area.
Kuzmanovic and his PhD students Marcel Flores and Uri Klarman have found that problems caused by competing networks can be mitigated by using an already-existing, extremely cheap medium: FM radio. Flores will present this work Tuesday, November 10 at the 23rd annual IEEE International Conference on Network Protocols in San Francisco.
"Our wireless networks are completely separate from each other," said Flores, the lead author of the study. "They don't have any way to talk to each other even though they are all approximately in the same place. We tried to think about ways in which devices in the same place could implicitly communicate. FM is everywhere."
Called "Wi-FM," the team's technique enables existing wireless networks to communicate through ambient FM radio signals. The team agreed that using FM was attractive for several reasons. For one, most smartphones and mobile devices are already manufactured with an FM chip hidden inside. FM is also able to pass through walls and buildings without being obstructed, so it's very reliable. Minor upgrades to software would allow devices to take advantage of Wi-FM.
Using Wi-FM prevents a person's network data from fighting with his or her neighbor's data. When network data are sent at the same time, they bump into each other. Then both data packets back off and stop moving toward their destinations. This is what causes those unexpectedly slow Internet speeds. Wi-FM works by allowing the device to "listen" to the network and select the quietest time slots according to FM radio signals.
"It will listen and send data when the network is quietest," Flores said. "It can send its data right away without running into someone else or spending any time backing off. That's where the penalty happens that wastes the most time."
This is a problem that Klarman knows all too well. Living in a large, urban apartment building with more than 30 different networks, he regularly experiences slow Internet speeds.
"Even if I configure my Internet to choose a channel that is least likely to overlap with my neighbors, the problem cannot be avoided," Klarman said. "You can't find a quiet channel when there are 30 other networks in the same building. My speed is 10 percent of what it should be."
Wi-FM identifies the usage patterns of other networks in order to detect times with lightest and heaviest traffic, helping to harmonize Wi-Fi signals that are transmitting on the same channel. And it can adapt as those patterns change with very little effort.
"Our system can solve these problems without involving real people," Kuzmanovic said. "Because are you going to knock on 30 doors to coordinate your wireless network with your neighbors? That is a huge management problem that we are able to bypass."
Testimonial
"The I-Connect007 team is outstanding—kind, responsive, and a true marketing partner. Their design team created fresh, eye-catching ads, and their editorial support polished our content to let our brand shine. Thank you all! "
Sweeney Ng - CEE PCBSuggested Items
Soaring Inference AI Demand Triggers Severe Nearline HDD Shortages; QLC SSD Shipments Poised for Breakout in 2026
09/16/2025 | TrendForceTrendForce’s latest investigations reveal that the massive data volumes generated by AI are straining the global infrastructure of data center storage.
Advanced Packaging-to-Board-Level Integration: Needs and Challenges
09/15/2025 | Devan Iyer and Matt Kelly, Global Electronics AssociationHPC data center markets now demand components with the highest processing and communication rates (low latencies and high bandwidth, often both simultaneously) and highest capacities with extreme requirements for advanced packaging solutions at both the component level and system level. Insatiable demands have been projected for heterogeneous compute, memory, storage, and data communications. Interconnect has become one of the most important pillars of compute for these systems.
Procense Raises $1.5M in Seed Funding to Accelerate AI-Powered Manufacturing
09/11/2025 | BUSINESS WIREProcense, a San Francisco-based industrial automation startup developing cutting-edge AI and remote sensing technologies for process manufacturers has raised $1.5 million in a seed funding round led by Kevin Mahaffey, Business Insider’s #1 seed investor of 2025 and HighSage Ventures, a Boston-based family office that primarily invests in public and private companies in the global software, internet, consumer, and financial technology sectors.
Zuken Announces E3.series 2026 Release for Accelerated Electrical Design and Enhanced Engineering Productivity
09/10/2025 | ZukenZuken reveals details of the upcoming 2026 release of E3.series, which will introduce powerful new features aimed at streamlining electrical and fluid design, enhancing multi-disciplinary collaboration, and boosting engineering productivity.
AI Infrastructure Boosts Global Semiconductor Revenue Growth to 17.6% in 2025
09/09/2025 | IDCAccording to the Worldwide Semiconduct o r Technology and Supply Chain Intelligence service from International Data Corporation (IDC), worldwide semiconductor revenue is expected to reach $800 billion in 2025, growing 17.6% year-over-year from $680 billion in 2024. This follows a strong rebound in 2024, when revenue grew by 22.4% year-over-year.