SDR Architectures--Key Features

The advances in SDR technologies require a new architectures to support the reconfigurability of those technologies. In this paper the authors show several architecture setups. Key features of any architecture includes variable bandwidth, sufficient dynamic range, fast retuning and reconfiguration, minimized cost, and energy efficient.



Best regards,
Hall T.

Primer on SDR continued--Cognitive Radio

In continuing last week's post on SDR primer, this week we look at Cognitive Radio which is "a radio that senses or is aware of its operational environment and can dynamically and autonomously adjust its radio operating parameters accordingly" according to Joe Mitola. Basically, the Cognitive radio senses its envionment and learns how to adapt to its environment. While SDR and Cognitive Radio can be a potential panacea for every wireless problem, it appears to be ideally suited for interoperability of radios and reducing interference. A software defined radio can come in various forms such as reconfigurable (can be changed by the user), to policy -based (change is based on a predetermined set of configurations), to cognitive radio (can be changed based on the environment).

Best regards,
Hall T.

SDR Advantages and Disadvantages

For those unfamiliar with software defined radio and related topics, here's a great tutorial by VT. The slide set does a good job in defining SDR by showing how the physical layer is moved from hardware into software and how the radio adapts to its environment although that part leans more to the Cognitive Radio side. The advantages of SDR include reduced component cost because hardware specific components are replaced by DSPs and FPGAs. The number of components tends to be lower. DSP components can compensate for problems in other areas of the system. Disadvantages include power consumption, security, and overall cost.

The key difference between traditional radios and software defined radios is that the latter senses their environment and adapt to it. This is important particularly to government agencies such as the FCC who is in the process of reallocating spectrum usage in the TV bands. There's a shift away from rigid, spectrum allocations to a demand-based approach to maximum the usage in that band. Currently, spectrum usage in most bands in the USA range from .3 to 3%. As the need for more spectrum grows, a new paradigm will be needed to supply this bandwidth.

Best regards,

Hall T.

Using SDR in Pico Satellites

Software defined radio is not just for mobile phone communications. It finds applications in numerous other places and for different reasons. One of the more intriguing applications is that in pico satellites. Pico satellites are small satellites developed by university groups for research and education purposes which catch a ride on rocket and space shuttle flights. Once out of the earth's atmosphere, they are tossed out into the ether and start their mission. In this application SDR is used for positioning and navigation. By using SDR techniques, the positioning system was made lighter, smaller, and most importantly with lower power consumption.

Best regards,

Hall T.

Power Management in SDR

SDR brings a great deal of resource-intensive processing to the application in order to handle the variety of waveforms, modulation schemes, and other RF functions. This additional processing comes at the expense of greater power requirements. This drives the designer to make performance/power tradeoffs in the design of SDR systems. Power management in Software Defined Radio is a major concern since most SDR applications use more power than hardware radios. Also, field deployed units need to conserve power since they typically run on batteries. The RF front ends are typically overpowered as they need to generate RF patterns across a wide range. Finally, SDR applications tend to generate excessive heat which needs to be dissipated.

Power management is a key success factor in software-defined radio applications due to the portable nature of most target systems. In this paper the authors propose a horizontal layering of the hardware along with software-specific APIs to provide component-level control over power management. By dividing the system into components, power consumption can be customized for the application. For example, a signal processing intensive application could shut down other functions while the FPGA/Processor works. The user can turn off functions to increase the battery lifetime in a mission critical operation.

Best regards,
Hall T.

Dynamic Spectrum Access Regulatory Models

The current model of RF Frequency management is to assign frequency bands to groups and applications. This mitigates the problem of interference and controls to some extent the usage of the RF spectrum. In many of the allocated bands, the usage of the alloted spectrum runs from 0.03 to 3%. As spectrum becomes scarce due to a growing use of existing applications and new applications requiring additional bandwidth come on the market, the spectrum access model is coming under scrutiny. A newer Dynamic Spectrum Access Regulatory model is coming into focus. In this model, the spectrum is not allocated but rather users are allowed to make use of a frequency band if they can do so without interfering with other users in that band.

The key conference in this area is the IEEE Dyspan (Dynamic Spectrum Access Network ). The purpose of the group is

" to expand collective understanding of complex next-generation wireless systems focused on using RF spectrum more efficiently and dynamically. This includes advancing both cutting edge technical and multidisciplinary research as well as practical experience related to building a healthy industry/regulatory ecosystem for the commercialization of smart radio system technologies.

As networks and devices increasingly gain intelligence and "cognitive" capabilities, and as regulators around the world seek to enhance spectrum utilization through exploiting areas such as "white spaces", dynamic decentralized access is becoming one of the most important, but most complex topics in wireless communications development. IEEE DySPAN 2010 will continue investigation of decentralized spectrum access and focus on approaches for highly scalable dynamic optimization of wireless spectrum use. "

There are several initiatives to further the Dynamic Spectrum Access effort. One is the Potential Alliance for World-wide Dynamic Spectrum Access by the New America Foundation. The group seeks to foster better working relationship between military, consumer, and public-safety groups.



Best regards,

Hall T.

What will the "IBM PC" of SDR Look Like?

The IBM PC revolutionized the personal computer industry by bringing standards to a low-cost personal computer platform. It became the dominant design drawing from several industry-proven technical solutions and covering many market segments. It held the right tradeoff balance between technical performance and market requirements. Up until the entry of the IBM PC the market was flush with competing standards and methodologies. An "IBM PC" solution will eventually come to the SDR world. Software defined radio will see much greater success if it can generate a dominant design solution rather than a splintered approach.

So what are the key elements that need to be synthesized into a dominant design? In this paper the authors argue that it will bring portability of waveforms, maintainability, and allow specialization of waveforms.

Another key factor will be size. It will be greatly reduced from what we see today. Just as the IBM PC took computing from the mainframe/mincomputer world down to the desktop so the next generation of SDR will go down in size. Here's one example using Gumstix.

Finally, the cost must be low and the tools widely available. Defacto standards available at low cost will drive SDR applications into new areas and create a richer set of tools for more sophisticated applications.



Best regards,

Hall T.