How WLAN superheterodyne receiver works?

The practical design of radios has been largely dominated by the superheterodyne architecture, commonly referred to as superhet. This architecture employs two frequency conversion stages. Firstly, the received frequency is converted to an intermediate frequency (IF). Then, after undergoing amplification and filtering, the IF is further converted to the baseband frequency. In the case of a WLAN application, where the RF frequency is 2.4GHz, it may be converted to a shared IF of 374MHz and subsequently down-converted to DC.

It is important to note that the final down-conversion is performed twice, resulting in an I and Q output. Here, I and Q represent the in-phase and quadrature components of the signal, respectively. This dual conversion process is necessary because the down-conversion operation cannot differentiate between frequencies above and below the carrier. As a result, these frequencies overlap when the carrier is converted to zero frequency. By performing two conversions using different carrier phases, we can preserve information about the sidebands.

The block diagram below illustrates the configuration of a WLAN superheterodyne receiver. For the sake of simplicity, only one final down-converter is shown. In this setup, RF mixers are employed to convert the RF signal to an intermediate frequency (IF), and subsequently, the IF signal is converted to a baseband output at a frequency of '0' (zero).

WLAN receiver

The band select filter is designed to encompass the entire RF band of interest, such as the complete 2.4 to 2.483 GHz Industrial, Scientific, and Medical (ISM) band. The amplifiers following the filter are intended to operate effectively across this entire frequency range.

The first local oscillator (LO) must possess tunability. The IF is determined by the difference between the desired RF frequency and the frequency of the LO. To receive a specific channel, like channel 1 at 2412 MHz, the LO needs to be set to (2412 - 374) = 2038 MHz. However, once this initial conversion is performed, the IF remains constant, and the DC component remains the same. Consequently, the remaining components of the radio can be fixed at their respective frequencies. It's worth noting that in modern radios, the baseband filters may incorporate adaptive bandwidths to enhance performance under varying conditions.

Certain key elements significantly impact the performance of a superhet receiver. The low-noise amplifier (LNA) is the primary contributor to excess noise and typically determines the sensitivity of the radio. The cumulative distortion produced by all the components (including amplifiers and mixers) prior to the channel filter plays a crucial role in determining the selectivity of the receiver. If an interfering signal becomes distorted, it may acquire power at the desired frequency and bypass the channel filter. Additionally, sufficient gain adjustment must be provided to ensure that the output signal falls within the operational limits of the analog-to-digital converter (ADC), which are typically narrower than the range of input RF powers encountered.

On the other hand, a superhet transmitter functions in the opposite manner compared to a receiver. The in-phase (I) and quadrature (Q) baseband signals are mixed to an intermediate frequency (IF), combined, and then, after amplification and filtering, converted to the desired RF signal. In a WLAN radio, the IF is generally consistent for both transmit and receive operations, allowing the use of a single channel filter for both directions and facilitating the sharing of local oscillators (LOs) between transmit and receive functions.

 Similar to the superhet receiver, the components preceding the final mixer in the transmitter function at fixed frequencies, either in the baseband or the intermediate frequency (IF) range. However, the final mixer and amplifiers in the transmitter must operate across the entire desired RF bandwidth. The performance of the transmitter is often dictated by the power amplifier. Distortion in the power amplifier can result in the emission of unwanted spurious signals, both within neighboring channels and in other frequency bands. To mitigate distortion, the output power can be reduced while maintaining a constant DC power level, but this inevitably decreases the efficiency of the power amplifier. As the power amplifier holds significant influence over transmitter performance, the advantages offered by the superhet architecture are comparatively fewer in transmission than in reception.

Despite the relative ease of implementing intermediate frequency (IF) filters compared to radio frequency (RF) filters, IF filters still contribute to the overall cost of a superhet radio. Additionally, IF amplifiers tend to be more expensive than baseband amplifiers. This raises the question: why not eliminate the IF stage altogether? Two increasingly popular architectural alternatives, known as direct conversion and near-zero IF (NZIF), aim to do just that.

In a direct conversion radio, the name itself implies the process. RF signals are directly converted to the baseband in a single step. The following shows a direct conversion radio receiver.

direct conversion receiver

The channel filter in this case is a low-pass filter that spans from zero frequency to a few megahertz. Since these frequencies are relatively low, active filters utilizing amplifiers can be employed, allowing for flexible digital adjustment of bandwidth and gain. This not only simplifies the filtering process but also reduces costs, as inexpensive low-pass filters can be utilized in place of complex IF filters. Furthermore, at these lower frequencies, digital adjustment of the filter characteristics becomes more versatile and manageable.

Direct-conversion radios face significant challenges as well due to the absence of an intermediate frequency (IF) stage. The gain that was traditionally provided by the IF stage needs to be compensated elsewhere, often at the baseband level because it is a more cost-effective option. However, this shift introduces certain issues.

One such issue is the occurrence of DC offset voltages, which can arise from various sources, including distortion in the local oscillator (LO) signal. These offset voltages can be amplified to such an extent at the baseband that they drive the voltage of the analog-to-digital converter (ADC) to its limits, overpowering the desired signal. Similarly, low-frequency noise stemming from factors like electrical noise in metal-oxide-semiconductor field-effect transistors (MOSFETs) and microphonic noise (resulting from mechanical vibrations within the radio) can be amplified to an undesirable level due to the substantial baseband gain.

Moreover, in direct conversion systems, the LO operates at the same frequency as the RF signal. Consequently, it cannot be effectively filtered out and may inadvertently radiate during the receive operation, causing interference with other users. This radiated signal can interact with external objects in a time-varying manner, resulting in reflections that interfere with the desired signal.

To address some of these challenges, a solution is to convert the received signal to a very low frequency instead of zero frequency, just enough to accommodate the entire signal. In the case of a WLAN signal, an intermediate frequency (IF) of approximately 8 MHz can be chosen, resulting in what is known as a near-zero IF (NZIF) receiver.

By selecting a low IF, there is no need for additional frequency conversions, and the IF signal can be directly accepted by the analog-to-digital converter (ADC). As there is no signal at DC in this configuration, any DC offsets can be easily filtered out without affecting the performance of the receiver. Furthermore, low-frequency noise below a few megahertz, which includes electrical noise and microphonic noise, can also be effectively filtered out.

However, due to the small IF value in this setup, the image of the desired frequency is very close to the actual desired frequency and cannot be filtered out conventionally. Therefore, alternative methods of image rejection must be implemented to ensure unwanted images do not interfere with the receiver's operation.


[1] Coherent Detection of DSB-SC AM wave

[2] AM demodulation with AD633 as Square Law Detector 

[3] Balanced Slope Detector - FM Demodulation

[4] Zero-Crossing FM Detector 




Post a Comment

Previous Post Next Post