Global Sources
EE Times-Asia
Stay in touch with EE Times Asia
EE Times-Asia > RF/Microwave

Near-field magnetic comms emerges

Posted: 17 May 2004 ?? ?Print Version ?Bookmark and Share

Keywords:nfmc? wireless? headphone? rf? wi-fi?

Reliable delivery of wireless streaming audio and voice in the rapidly growing market for portable devices is a demanding technical problem. The solution will not be achieved by simple customization of one of the many wireless RF technologies in the market today. Instead, it will require looking at the unique challenges of this application, and developing solutions that fully meet customer and market requirements.

Let us consider the example of a wireless headphone for a portable audio player. The ubiquitous wired headphone delivers near-perfect audio quality, requires no batteries, is lightweight and is virtually free. While the wire is widely recognized as a significant inconvenience, it is accepted by the consumer because no competitive wireless solution currently exists that meets the criteria of excellent audio quality, charging convenience, light weight and low cost.

Near-field magnetic communication (NFMC) is a wireless physical layer with a number of unique attributes that make it ideal for high-density, short-range streaming applications. While the advantages of lower cost and power are frequently cited, what makes NFMC a compelling alternative to RF in these applications is the unique attenuation property of the magnetic field.

To achieve good sound quality in portable wireless applications, QoS must be guaranteed. This term, when applied to real-time delivery of voice and music, basically means that the device must meet a user's expectation for quality and reliability of real-time content delivery. At a more technical level, it means that the transmission characteristics, such as bandwidth, latency and error rates required to transmit the content, can be guaranteed in advance.

In an audio application, any deterioration in QoS will typically be perceived by the user as degraded sound quality in the form of noise, pops, static and dropouts. Conversely, when non-time-critical data rather than audio is sent through a channel with poor QoS, the resulting data loss can be corrected through various data retransmission techniques that often go unnoticed by the user.

Magnetic vs. RF

It is illustrative to compare conventional RF and NFMC as physical-layer solutions for short-range communications. NFMC communicates wirelessly by coupling a low-power, non-propagating, quasistatic magnetic field between devices. The range for NFMC, at 1.5m to 2m, is both predictable and highly reliable due to a strong 1/r6 attenuation in magnetic-field energy. On the surface this would appear to be a considerable disadvantage for NFMC systems. In the context of short-range communication, however, the roll-off behavior is a substantial advantage.

While a small amount of RF energy invariably flows in free space, the majority of the energy is stored in the quasistatic magnetic field, which forms a tight communication "bubble." It is this fact that allows a modest bandwidth generated with very low power to be spatially allocated among multiple users, even in high-user-density situations. Each user owns the entire bandwidth in his physical space, and the small number of rarely occurring "overlaps" can easily be accommodated by using simple frequency-allocation techniques. Furthermore, the field characteristics are relatively unaffected by the surroundings, as magnetic fields reliably follow the 1/r6 behavior regardless of the presence of metal objects, conductive materials or people.

RF communication is substantially different in its field shape and propagation characteristics. Modulated RF plane waves propagate through free space with theoretical attenuation with a distance approaching 1/r2. Due to the longer range of propagation, virtually all systems must be able to share the allocated bandwidth using time or frequency allocation. In practice, the RF signals are also quite unpredictable, especially indoors, where fades and blocking are prevalent. Further, the human body significantly attenuates and reflects RF signals through an effect known as the body shadow. The signal loss due to these combined effects can usually be restored to some degree by transmitting with more power, which requires still more sharing of bandwidth.

The issue of bandwidth allocation among users is particularly acute in the 2.4GHz RF band--which is getting more crowded as an increasing number of Bluetooth, Wi-Fi and assorted other consumer devices try to share the same limited bandwidth. The result is steadily worsening interference and interoperability problems that simply cannot be overcome by transmitting with yet again more power or moving to more complex and power-intensive frequency-management schemes.

NFMC suffers from virtually none of these problems. The tight communication bubbles afforded by the 1/r6 roll-off of the magnetic-field energy allow for a high density of NFMC systems to be co-located with guaranteed QoS. Further, because the systems transmit at extremely low power, they largely fall below regulatory limits in much of the world, allowing further flexibility in the frequency of operation. Today's NFMC devices, such as wireless headsets for mobile phones, operate in the worldwide ISM band at 13.56MHz, where the only known interferers, RFID readers (not tags), experience the same 1/r6 roll-off and thus interfere only when within a 1m to 2m range.

Real-world products

Much of the discussion on guaranteed bandwidth illustrates problems encountered by Bluetooth or similar RF systems in voice applications. Voice requires only 64Kbps at a bit-error rate (BER) of 10-3. In comparison, high-quality music in an NFMC system using a 4:1 ADPCM codec requires a data rate of 384Kbps at a BER of 10-5. The interference issues for RF worsen when targeting these specifications. While 384Kbps may appear excessive for music, especially since high quality can be attained using 128Kbps MP3, there are other factors to consider that are generally applicable to high-QoS wireless applications.

First, if highly compressed music content is transmitted to a wireless headphone, a complex decoder, such as that found in an MP3 player, is likely required in the headphone. Also, depending on the format of the audio, a complex encoder may also be required on the transmitting end. These additional devices consume more power and occupy more ASIC die area, thereby decreasing battery life while increasing the cost of the system.

Second, highly compressed information requires a much higher-quality channel for the reliable delivery of content. This is where the robustness of the coding scheme becomes important. A 384Kbps ADPCM encoding scheme that requires a BER of 10-5 is relatively tolerant of small data errors. Comparatively, even a modest number of channel errors introduced into a 128Kbps MP3 audio stream creates an unacceptable level of audible disturbances. Thus, a wireless-streaming channel necessitates an opposing set of trade-offs, namely higher bandwidth for robustness in the presence of interference.

In a high-user-density RF wireless situation, it is not possible to allocate 384Kbps of low-BER capacity to each user while simultaneously achieving very low power consumption. Moving to a higher-frequency spectrum, such as the 5GHz band, may improve things somewhat but at the expense of higher power. NFMC's compact communication bubbles give all users uninterrupted access to this desired bandwidth. Multiple access to a defined spectrum is accomplished not by dividing frequency or time, but by dividing space-an approach that is not possible for an RF short-range system.

- Vincent Palermo

Chief Technology Officer

Aura Communications Inc.

Article Comments - Near-field magnetic comms emerges
*? You can enter [0] more charecters.
*Verify code:


Visit Asia Webinars to learn about the latest in technology and get practical design tips.

Back to Top