Working with high-frequency signals and selecting components for a signal chain is challenging enough. Amplifiers are an important part of a signal chain for radio systems as they provide the boost a signal needs in order to reach its destination. In these systems, there tend to be two types of amplifiers that will make an appearance: a low noise amplifier (LNA) and a power amplifier (PA). Both types of amplifiers perform a similar function, but in different places in the signal chain.
The differences between LNA vs. PA components illustrate something more fundamental about amplifier selection: what aspect of the signal is being manipulated by the component prior to delivery to a load. In radio systems, these amplifiers will both appear in the RF front-end as part of signal broadcasting and reception, so these components must be chosen carefully and should operate within the correct signal power range to provide best results. In this article, I’ll examine the differences between these two types of components and provide some examples of advanced parts for RF systems operating in many frequency ranges.
In RF front end, an LNA and a PA are typically used on the RX and TX sides, respectively. This is generally the case in many RF systems that require wireless communication; the PA and LNA sections are often built into application processors or highly integrated RF transceivers. A similar use case appears in audio, where the power amplifier is driving a speaker and an LNA could be used on a microphone to collect faint voices from the nearby environment.
The image below shows where amplifiers typically appear in an RF front-end and how these amplifiers are implemented in the TX and RX sides of the signal chain. This type of TX/RX architecture is typical in chips that have an integrated transceiver block, as well as in systems that use discrete components running at higher power. The switch on the output is optional and is being used to implement time division multiplexing (TDD) with a single antenna so that TX and RX are separated into different time windows. However, this is not required and the RX/TX lines can be connected directly to their own antennas.
On the RX side, the LNA input is fed directly to a demodulator/down converter to extract data from a received modulated signal. The LNA only handles the input received by the RX antenna and is meant to provide just enough gain to ensure the signal exceeds the receiver’s threshold sensitivity. This effectively extends the receive range with just a small amount of gain applied on the RX signal chain.
On the TX side, the power amplifier takes the output from the modulation/up-conversion stage and amplifies it to deliver maximum power to the load. In the case of direct connections to an antenna, the power given the antennas or any other components in the system could require matching to a reactive impedance. This will require conjugate impedance matching with a nonlinear component to hit maximum power transfer as described below.
With these points in mind, let’s look closer at each type of amplifier.
The purpose of a power amplifier is very simple: to deliver maximum power to a load with minimal distortion of the signal. In terms of signal level, the power amplifier should maximize the signal-to-noise ratio in terms of power compared to the noise floor within the signal chain’s bandwidth. This should sound rather simple and an obvious function of an amplifier, but as I’ve discussed in articles on other types of amplifiers, different amplifiers involve different signal inputs and will attempt to accommodate different types of loads in the signal chain.
To deliver maximum power to a load, conjugate impedance matching is needed in the signal chain. Power amplifiers operating in the MHz to GHz range for radio systems can operate with 50 Ohm output impedance, so the antenna could be designed to 50 Ohm impedance to provide real impedance matching. In the case where the antenna impedance is reactive, an impedance-matching network is needed with passives, or a cascaded impedance transformer is needed. The latter is only feasible in physically large systems when working at MHz frequencies, but this can be done at high GHz frequencies without making the board too large.
The other important point about impedance matching is that simple conjugate matching will not actually deliver maximum power transfer to the TX antenna in most situations. This is because it is common to run a power amplifier very close to saturation (near the 1 dB compression point). In this state, the power amplifier’s transfer function starts to become nonlinear, as shown below.
In this state, maximum power transfer will occur when there is some very slight impedance mismatch left between the power amplifier and its load. This is because the maximum power transfer value will be a function of the input power level, which requires solving a transcendental equation in an optimization problem to determine the optimal impedance matching. A simulation technique called load-pull analysis can be used to determine the optimal mismatch that provides maximum power transfer.
Power amplifiers are available in any of the standard amplifier classes, and components are available in many frequencies ranges spanning from audio to microwave.
Some of the important specifications used to select a power amplifier include:
Gain @ required frequency - The gain value that is given in the amplifier’s specifications will be valid for a specific operating frequency or frequency range.
Driving mechanism - Usually analog driving is required for higher frequencies, while lower frequencies (e.g., audio) can run with PWM driving.
Gain-bandwidth product - The overall bandwidth will be limited as the gain in the amplifier is ramped higher. Make sure you can get the gain and bandwidth you need with this specification.
Thermal resistance - Power amplifiers can get hot, so it is important to note the thermal resistance to get a rough estimate of the component’s operating temperature.
1 dB Compression and 3OIP points - The former value tells you when the amplifier starts to saturate, while the latter tells you when 3rd-order intermodulation product powers are equal to the main signal power. This limits the input power you can use in the amplifier.
The HMC455LP3 from Analog Devices is a 2.5 GHz power amplifier based on a GaAs-InGaP heterojunction bipolar transistor. This amplifier provides a high 3OIP point (see the transfer function curves below) with approximately 12 dB gain at up to approximately 15 dBm input power. This component can be useful in low-frequency microwave systems operating from 1.7 GHz to 2.5 GHz.
For audio systems, the TPA2012D2RTJR from Texas Instruments is a Class D audio amplifier offering selectable gain with up to 2.1 W power output in the audio range. The component can provide power delivery to 4 Ohm or 8 Ohm speakers at 5 V or 3.6 V with a selectable gain of up to 24 dB. This component comes in a very small BGA package making it appropriate for use in mobile devices, including phones, tablets, and portable media players.
Low-Noise Amplifiers
A low noise amplifier is intended to amplify the voltage of an incoming signal without significantly amplifying the accompanying noise in the system, thereby increasing the SNR value for the signal. These components must have very low inherent noise to provide such amplification characteristics. They must also be able to sufficiently reject noise sources within their operating bandwidth, requiring high PSRR and narrow roll-off in their transfer function curve. Finally, to minimize distortion when applying high gain, these components must have high linearity to prevent the generation of harmonics and intermodulation products.
In order to provide very high gain with minimal noise amplification, one of the important specifications is the noise figure, or more specifically the gain-to-noise-figure ratio. Some very sensitive receiver applications could require these ratios to range from 20 to 30 (e.g., 1 dB noise figure with 20 to 30 dB gain).
One very simple example of LNA is the MBC13720NT1 from NXP Semiconductors. This LNA component has a very wide operating frequency range, spanning from 400 MHz to 2.4 GHz. This component can provide selectable controlled current up to 11 mA with high gain reaching 20 dB at 900 MHz. The noise figure is also low, with a gain-to-noise-figure value of approximately 15. This type of component would be useful on the RX side of sub-GHz radio transceiver modules operating at high power.
From the diagram shown above, it should be clear that there are many other components needed to build a complete signal chain for moderate to high-power RF systems. For consumer-level applications, or when operating in Bluetooth WiFi, there are some highly integrated RF MCU SoCs that include the entire front end built into the component. There are also wireless modules that can be selected for these systems, which will include the entire front end of the chip. Other radio bands, which do not have the same level of market penetration, generally lack these integrated solutions, and designers will have to take the approach outlined here.
An application like software-defined radio, amateur radio, or operation in an ISM band will likely require building your own signal chain entirely from discrete components. Some of the components you’ll need in this application include a digital processor for controlling the entire system, as well as each of the RF elements listed above. Some of the components you need can be found in the following resources:
Designers working on RF systems using integrated or discrete components can access the data, insights, and sourcing information for free using the search features in Octopart. Only Octopart provides advanced search and filtration features to help buyers find components and up-to-date distributor pricing data, parts inventory, and parts specifications. Take a look at our integrated circuits page to find the components you need.
Stay up-to-date with our latest articles by signing up for our newsletter.