Skip to main content

RF Sampling Architecture

Published Date
Author Cadence PCB Solutions

Key Takeaways

  •  When the signals involved in analog to digital conversion are RF signals, then the sampling refers to RF sampling.

  • IF and Zero-IF are the traditional RF sampling architectures and direct RF sampling  involves direct conversion without using mixers and local oscillators. 

  • Direct RF sampling architecture consists of low noise amplifiers (LNA), filters, and ADCs.

RF Sampling

        When the signals involved in analog to digital conversion are RF signals, then the sampling refers to RF sampling

Most modern communication and radar technologies operate at RF frequencies. RF sampling is necessary for such systems to achieve the desired performance within the expected dynamic range of frequencies. Analog-to-digital converters are employed for direct digitization of the RF signal in communication systems and radar circuits. Let’s read more about RF sampling and various architectures in this article.

RF Sampling

We are in a digital world; digitalization offers greater advantages than analog signals, and most systems have evolved into digital over the past years. In most cases, the original signals generated–for example data or voice–are analog in nature. We convert these analog signals to digital ones to make them suitable for transmission without loss. The analog to digital conversion involves sampling, quantization, and coding.

Sampling is the process in which continuous-time signals are transformed into discrete time signals. When the signals involved in analog to digital conversion are RF signals, then the sampling refers to RF sampling.

RF Sampling Architecture

RF sampling architecture can be classified into:

  1. Intermediate frequency (IF) architecture
  2. Zero IF architecture
  3. Direct RF sampling architecture

The first two are traditional RF sampling architectures and the last one is direct conversion without using mixers or local oscillators. 

IF Architecture

The signal to be sampled goes through the following processes:

  1. The signal received from the antenna (to be sampled) is amplified.
  2. The amplified signal is down-converted to IF using a mixer.
  3. The IF signal is amplified using a variable gain amplifier.
  4. The amplified IF signal is filtered and given to an analog-to-digital converter for signal digitization.

IF architecture

Zero IF Architecture

  1. The RF signal is amplified and down-converted to the baseband within analog quadrature  demodulation.
  2. The baseband signal is filtered.
  3. A dual ADC is employed for converting the RF analog signal to digital.

Zero IF architecture

Direct RF Sampling Architecture

Even though the end outcome of direct RF sampling is the same as that of traditional IF and zero-IF architecture, the former is in demand due to the replacement of mixers and local oscillators with digital processing. Let’s see the steps involved in direct RF sampling.

Direct RF Sampling

Direct RF sampling architecture consists of low noise amplifiers (LNA), filters, and ADCs.

  1. The signal received from the antenna is filtered and amplified.
  2. The amplified signal from LNA is filtered again and sent to ADC.
  3. The direct digitization of RF signals is performed by ADC, which sends the digitized signals to the processor.
  4. Other operations such as mixing and filtering the sampled signal received from ADC are implemented using digital signal processors.

Advantages of Direct RF Sampling

  1. Direct RF sampling replaces traditional components in IF and zero-IF sampling architectures such as mixers, local oscillators, etc. with digital signal processing. RF sampling simplifies the RF signal chain.

  2. Direct RF sampling enables the development of low-power and small-form-factor systems in cellular communication, thereby reducing the count of remote radio head (RRH) boxes at each cell site in cellular infrastructure.

  3. Wideband operation is possible by keeping to giga samples per second (GSPS).

  4. In traditional IF and Zero IF architecture, additional signal chains must be incorporated to include extra bands. The bandwidth limitation of traditional RF sampling architecture is overcome in direct RF sampling. 

  5. In direct RF sampling architecture, one ADC can handle multiple bands of frequency, only at the expense of additional digital down converters for baseband signal conversion.

  6. Reduced cost per channel.

  7. Less channel density.

  8. Smaller in size and power efficient.

Where to Use Direct RF Sampling Architecture

With the application of CMOS technology, direct RF sampling offers better power dissipation and performance than traditional RF and zero-IF architecture. Direct RF sampling architecture is used in the following systems:

  1. Developing systems with high channel count, as it reduces the footprint and cost.
  2. For applications requiring the removal of local oscillator leakage and quadrature impartments.
  3. For systems requiring simplified synchronization.

Direct RF sampling avoids the need for synchronizing the internal clocking of local oscillators and RF instruments. The only synchronization in direct RF sampling is the clock synchronization of RF devices. Whichever system requires RF sampling with SWAP-C (size, weight, power, cost) reduction, direct RF sampling can be implemented.

Cadence offers a full suite of analysis tools where you can simulate the behavior of a direct RF sampling signal chain. Leading electronics providers rely on Cadence products to optimize power, space, and energy needs for a wide variety of market applications. If you’re looking to learn more about our innovative solutions, talk to our team of experts or subscribe to our YouTube channel.