Paper discussion 17a: Revisiting Software Defined Radios in the IoT Era

[grahamschock]

No description provided.

[grahamschock]

Review Type: Skim
Reviewer: Graham Schock

Problem Being Solved
Low Power IoT devices currently make use of dedicated wireless internet chips. Although this choice allows for more generality in IoT it also sacrifices a little bit of performance compared to a software radio. However, as IoT devices are simple and low powered this trade off might no longer be true. Another problem being addressed is the idea of cross technology interference, this is when “technologies share the same RF spectrum in an un-coordinated way”.

Contributions
In order to meet this challenge the paper proposes GalioT which is an architecture for low-power IoT devices. GalioT's main contribution is that it can decode cross-technology interference when multiple devices are on the same RF spectrum. It achieves this through unique software filters that help identify, segregate and decode radio signals. The paper also outlines somes of these algorithms as well as show evidence that GailoT is effective.

[nikorev]

Reviewer: Niko Reveliotis
Review Type: Comprehension (Skim Edition)

Problem Being Solved
Traditionally, the market has shifted towards using dedicated wireless internet chips over software radios for their performance benefits. As the IoT-space continues to mature, a case where software radios may be more fitting has appeared. Specifically, in software radios having the ability to "decode any wireless technology with a simple software update." With IoT devices not being as network-constrained in needing top-tier performance (minimal latency, high throughput, etc), software radios can be a cheaper alternative to the current solution of "multi-technology gateways that support many radio technologies ... through multiple radio chips each decoding dedicated wireless technologies".

Main Contributions

• "GalioT, a cloud-assisted IoT gateway" that is able to decode multiple different wireless technologies simultaneously.
  • Also has the ability to segregate different connections while they're interfering on the same frequency band.

[jacobcannizzaro]

Reviewer: Jacob Cannizzaro
Type: Comprehend/Skim

Problem:

Software radio was once seen as the future of radio technology. They are programmable, easily updateable to new radio technologies, and can simultaneuously read multiple types of wireless communications.The problem with them is that to optimize throughput for different wireless technologies adds a lot of cost that made it more feasible to just optimize hardware of each individual technology. Many IoT devices using many of these technologies at once causes current multi-technology gateways to experience collisions of radio waves from different devices that can be very hard or impossible to decode which leads to a loss in throughput as well as excess battery drain.

Main Contributions:

This paper introduces GalioT, a framework for a software multi-technology gateway that can handle concurrent decoding of multiple wireless technologies at once using cloud technologies. The prototype in this paper can handle Bee, Z-Wave, and LoRa wireless technologies, three of the most popular used in IoT devices everywhere. It is shown throughout that because IoT devices don't have high needs for throughput in most cases, and can deal with high latency bounds, software radios are enough for their low power needs. Whereas achieving the highest possible throughputs on any one technology with software radios is expensive, having one that can achieve results good enough for the IoT devices is actually cheaper than any existing multi-technology gateway. Throughput is shown to increase on average by 745.96%, fully supporting the goal of the paper. This technology is amazing because not only is it cheaper and more efficient, as new radio technologies come, the software radios can just be updated to keep up with the times instead of having a need for constant hardware replacement.

[searri]

Reviewer: Rick Sear
Review Type: Skim

Problem being solved

Internet-enabled devices pretty much all use specialized chips (capable only of transmitting/receiving over a particular protocol) to communicate because they have very high throughput. However, there is a cheaper, more general alternative: software defined radios.

Important contributions

• Since SDRs are more general and since IoT devices don't need massive throughput, it makes sense that IoT devices should use SDRs to communicate instead of specialized wireless chips
• GalioT is an architecture that includes a cloud and gateway component which can decode multiple types of wireless comms for an attractively low price

[s-hanna15]

Reviewer: Sam Hanna
Review Type: Skim

Problem Being Solved:
This paper talks about how many IoT devices use dedicated chips for wireless infrastructure. This paper proposes using software-defined radio in order to assist with decoding IoT signals. The current solution is very expensive and many low-power IoT devices could use software-defined radio as a cheap solution.

Important Areas:
In order to implement their software-defined radio, they created GalioT. This is a cloud-assisted gateway which works to decode signals and help to prevent collisions. GalioT focuses on low-power IoT devices and works to filter conflicting signals in all environments. They created a universal preamble to detect signals, then they send the signals to the cloud in order to do collision detection.

[samfrey99]

Reviewer: Sam Frey
Review Type: Skim

Problem
Software-defined radio decoding was previously explored as a potential replacement for having multiple radio chips for WiFi, Bluetooth, etc. While combining multiple chips into one and decoding the signals in software offered simpler hardware, software-defined radios were ultimately abandoned in favor of the performance benefits of individual chips. With IoT devices generally having more relaxed requirements for latency, software-defined radios may be able to provide increased compatibility with new radio technologies with just a software update.

Contributions
The authors propose GalioT, an IoT framework where devices use one chip for receiving radio signals and a cloud application for decoding. The cloud decoder also provides increased tolerance to radio signal collisions.

[ericwendt]

Review Type: Skim
Reviewer: Eric Wendt

Problem Being Solved
This paper talks about the software radio technology. They argue that software radios have been thrown by the wayside in favor of the faster multiple radio chip strategy that has taken over. They introduce GalioT to assist with a revitilization of software radios.

Contributions

• GalioT is the first contribution, listed as intending to "decouple" collisions of multiple radio signals. This works even when there are not enough antenna to receive all the different types.
• kill filters allow GalioT to block out radio signals that are not relevant for the receiver.
• results: the kill filters results in a gain of 818.36% in high SNR and 532.4% in low SNR.

[bushidocodes]

Review Type: Skim

Reviewer: Sean McBride

Problem Being Solved

The proliferation of competing wireless IoT standards has resulting in the development of multi-protocol hubs with dedicated chips for speaking each of the dominant protocols. However, these different wireless technologies interfere with each other, and the gateway relies on the dedicate chips to handle interference and retransmission. How might a software defined radio that could receive and interpret these different protocols help dis-aggregate overlapping transmissions and reduce interference?

Contributions

Created a Raspberry Pi-powered software defined radio receiver that is able to generate a least-common-denominator "universal preamble" that can detect transmissions of any of the supported wireless protocols. These transmissions are captured and uploaded to a cloud-based system that is able to apply kill filters and successive interference cancellation to separate out interfering instructions. This resulted in an increase in average throughput of 745.96%.

[anguyen0204]

Review Type: Skim
Reviewer: Andrew Nguyen

Problem Being Solved
Software radio technology is something that can be revisited and applied to IOT. Since chips have been dominant lately, these piece of technology may provide more ingenuity than meets the eye. GalioT is a cloud assisted gateway that could aid in this idea.

Contributions
IoT should use SDR's as a way of better communications and such. GalioT can leverage edge computing to process I/Q samples from collisions. It can also be an effective replacement in cost, flexibility and scalability against the current existing multi-tech gateway platforms. GalioT can create unique single preambles that captures unique signatures and data of the radio technologies it intends to decode.

[AkinoriKahata]

Reviewer: Akinori Kahata
Review type: Skim

1. The problem being solved.

• Software radio technique, which can respond a wide variety of wireless transmission by program, is important for the era the number and type of IoT devices are increased. For IoT gateways, to connect to many IoT devices, such as a wide variety of sensors, need to be more cost and energy-efficient. In terms of those requirements, a gateway does not need to awake all time but wake up at the timing they detect the transmission. For implementing such a detection system, avoiding collisions of many IoT devices transmission and interference is essential.

2. The main contributions.

• The researchers suggest the IoT gateway “GalioT”, which are programmable and can avoid collision of multiple transmissions of IoT edge devices. GalioT consists of a gateway device and cloud computing, and GalioT adopts the multiple filters, which can separate transmission by frequency, Chirp Spread Spectrum, and code. Compared to a conventional technique for canceling interference of multi transmissions “SIC”, GalioT shows high throughput, high flexibility and cost-efficiency.