Modulation Techniques For Wireless Data Links

Despite a slow start, the wireless data industry has shown strong growth over the past few years. To continue this trend, wireless companies must work hard to lower operating and system costs. Re-evaluation of modulation schemes is one successful approach to meeting these goals.

Data transmission across a noisy wireless communications channel requires some manner of separating valid data from background noise. To accomplish this, engineers generally modulate data during transmission and demodulate data during reception.

In a wireless system, the efficiency of the modulation/demodulation process determines the accuracy of the data coming from the receiver. Therefore, careful consideration must be given to the selection of an appropriate modulation scheme during the design and development of wireless data systems.

Many modulation techniques are available for today's design engineers. Some, such as on-off shift keying (OOSK), are very simple and inexpensive to implement, while others, like Gaussian minimum shift keying (GMSK), efficiently use bandwidth and offer higher data rates. Two common modulation methods employed in wireless data systems are carrier present, carrier absent (CPCA) and frequency shift keying (FSK).

CPCA Modulation
Amplitude modulation (AM) is perhaps the oldest technique for impressing intelligence (or information) onto an RF carrier. With normal AM, the amplitude of a carrier frequency is raised and lowered in unison with the modulation source. If this source happens to be voice, then the carrier amplitude will follow the amplitude of the voice input.

In the case of AM data transmission, the modulation source is a voltage generator that has only two states: "on" and "off." When the modulation source is "on" (representing a logic high or a 1 in binary terms), the carrier amplitude is at its maximum. On the contrary, when the modulation source is "off" (representing a logic low or a binary 0), the carrier amplitude is at its minimum. This method of AM modulation is referred to by multiple designations including OOSK, on-off-keying (OOK), and CPCA. For the purposes of this article, CPCA will be used (Figure 1).

Figure1: In CPCA-based devices, the carrier amplitude is at its maximum when the modulation source is in an "on" state. On the contrary, when the modulation source is in an "off" state, the carrier amplitude is at its minimum.

Modulation depth is defined as the difference in output power between the "on" and "off" states of the modulation source. Modulation depth is an important parameter for any CPCA transmitter since it determines the possible sensitivity of the receiver, and, therefore, the ultimate range over which the data link will operate.

If a CPCA receiver has a sensitivity of -105 dBm, it will see any carrier with an amplitude above -105 dBm as an indication of a transmitted "on" condition. The receiver's modulation depth must be greater in magnitude than the difference between this sensitivity and the output power capability of the transmitter. For example, if the transmitter has an output power of 0 dBm and a modulation depth of 60 dB, then the transmitter features -60 dBm output power in its "off" state. But the receiver sees the carrier at this power level and considers it an "on" condition since the power of the carrier is above the sensitivity of the receiver. Therefore, if the modulation depth is not great enough, the carrier may never drop below the sensitivity of the receiver and, in turn, the receiver will always indicate an "on" condition.

Another important parameter of a CPCA transmitter is its output power during an "on" condition. When the modulation source is in an "on" state, output power is delivered into a 50 W load. Federal Communications Commission (FCC) regulations place limits on this output power. These limits are based on the center frequency and operational parameters of the transmitter.

Under FCC regulations, CPCA-based devices can transmit at higher peak output power levels since the carrier is not always present. As a result, a CPCA transmitter with a 50 percent duty cycle can output twice the power of a frequency modulation (FM) transmitter.

CPCA Benefits
CPCA modulation has many benefits. One of the biggest is cost. By accommodating the inherent inaccuracies of surface acoustic wave (SAW) resonators, CPCA transmitters and receivers can take advantage of the low cost of a SAW-based design.

Power consumption is another plus. When the modulation source is "off," the transmitter draws virtually no power. In the "on" condition, a SAW-based design typically dissipates one-half to one-third the power of a synthesized design. Thus, a CPCA transmitter with a 50 percent duty cycle can consume as little as 2 to 3 mA current during operation.

When developing any wireless system, size is always a concern. CPCA is a very simple modulation technique. As a result, it requires fewer components to be implemented, which, in turn, reduces the overall system size.

CPCA modulation does, however, have its limitations. Data rate is one of the largest drawbacks. Data rate in a CPCA-based system is a direct function of the start-up time of the oscillator. Since SAW resonators have a fairly high-loaded quality factor (Q), their start-up time can be as high as 20 µs. This limits a typical CPCA transmitter to data rates of less than 10 kb/s.

Another limitation is poor noise immunity. Any noise in the receiver passband above the receiver's sensitivity will interfere with data transmission.

FSK Modulation
FSK modulation is a simplified form of FM. In true FM, an analog signal is represented by a linear frequency deviation from the center frequency. FSK is a binary form of FM that uses hard shifts between deviant frequencies to represent the data originally impressed on the carrier. The magnitude of frequency shift is directly related to the magnitude of the modulation source voltage.

The FSK modulation source is allowed two states: "on" and "off." When the modulation source is in an "off" state, the carrier frequency is shifted down from the center frequency. On the other hand, when the modulation source is in an "on" state, the carrier frequency is shifted up from the center frequency. The amount of carrier frequency shift is referred to as the frequency deviation (Figure 2).

Figure 2: In an FSK scheme, the amount of carrier frequency shift is referred to as the frequency deviation.

Unlike CPCA, a carrier is always present with FSK modulation. This provides several benefits to the design engineer. First, the carrier will load the receiver at all times--providing greatly increased noise immunity. Secondly, the strength (or amplitude) of the carrier can be used to determine the quality of the incoming signal. A received strength signal indicator (RSSI) circuit is used to make this determination. This circuit outputs a voltage that corresponds to signal strength and has a typical dynamic range of 70 to 90 dB.

But there are drawbacks to having a continuous carrier. One drawback is power consumption. Since the carrier is continuously operating, it will require higher supply current to operate than CPCA-based systems.

FCC regulations also provide limitations. Under FCC rules, transmitters employing continuous carriers are subject to more stringent output power regulations. As a result, FSK-based systems transmit power levels are more restricted than CPCA-based devices.

Non-Return To Zero
FSK is a non-return to zero modulation method. As a result, the carrier should never be at the center frequency when modulation is present.

The benefit of a non-return to zero approach is noise immunity. Hysteresis can be applied to the detector, eliminating the effect of spurious frequency modulation generated from sources other than the data stream.

Since FSK relies on frequency change, and not amplitude change, to indicate data states, an FSK-based receiver is inherently immune to amplitude noise. This is of great importance in bands that are extremely crowded and feature a high potential for near-band interference, such as the Industrial, Scientific, and Medical (ISM) bands. This increased noise immunity suggests a potential for higher data rates. In fact, data rates up to 100 kb/s can be readily achieved with FSK-based systems.

Although FSK systems are immune to amplitude noise, they are very sensitive to frequency noise. Unwanted frequency changes caused by in-circuit sources will ultimately lead to bit errors in the data stream.

As mentioned, simple hysteresis can be applied to the FSK detector to remove some of the frequency noise. But a stable frequency source must still be used to ensure good noise immunity. While SAW resonators work extremely well for low baud rate applications at lower frequencies, their inherent frequency inaccuracies make them poorly suited for FSK applications. Thus, a synthesized source based on a crystal reference must be used.

It is a well-known fact that crystals are superior to SAW resonators with regard to loaded Q and frequency accuracy. But crystals cannot be operated in their fundamental mode at UHF. Instead, a crystal is used with a phase locked loop (PLL) to synthesize a high frequency. Although this technique is expensive and requires additional board space, it is the best method for attaining the tight frequency control necessary to achieve high data rates and noise immunity. It also provides the added benefit of channelization.

By using a divide-by-n PLL, the synthesized frequency can be set by changing the values of the internal counters. This allows an engineer to select a transmit or receive frequency from multiple channels. As a result, one transmitter or receiver can operate on many separate channels.

Linx Technologies, 1089 Medford Center, Bldg. 137, Medford, OR 97504. Phone: 800-736-6677; Fax: 541-471-6251.