With the adoption of the first 5G New Radio (5G NR) standard, the focus of new wireless product development engineers has shifted predictably from research to intensive development of new products that can deliver higher data rates, shorter latency times, and greater energy efficiency than 4G technologies currently in use. Along the way, they will inevitably face a number of technical challenges in designing a physical layer architecture that must cope with a more complex multi-user environment and higher frequency channel utilization.
Understanding how the requirements, as well as new technologies related to the 5G standard, differ from the past, fourth generation mobile communications (LTE) can help design engineers prepare to work with new architectures.
New architecture design and algorithms for 5G NR
Tangible growth of broadband access rate in 5G networks will be mainly due to the use of Massive MIMO technology (this technology involves the use of multi-element digital antenna arrays, with a number of antenna elements 128, 256 and more) in the transmission of data in a higher millimeter wavelength range (known as mmWave, which corresponds to frequencies from 30 GHz to 300 GHz), as well as new radio algorithms that allow more efficient use of the existing frequency resource. The new design of the architectures and algorithms will have a significant impact on all components of the new 5G systems, from the antennas and radio frequency components of the transmitting equipment to the algorithms of source signal processing (baseband). The performance of these subsystems is so closely linked that they need to be developed and evaluated together.
New 5G developments in the new millimeter wavelength range will also require the use of Massive MIMO antenna arrays with hundreds of antennas in network base stations (eNodeB). The presence of multiple antenna elements compactly placed within a relatively small space is an extremely important factor in achieving the high efficiency of the Adaptive Directional Diagram technology for the mmWave frequencies. Since the working wavelength is directly dependent on the allowable size of the emitting elements, the size of the new antenna elements may be up to 100 times smaller than those currently used in modern antenna arrays for ultra-high frequencies. Highly directional beams also minimize transmission losses because the transmission power is focused in a certain direction. Behavioral (functional) modeling of radio frequency and digital elements of Massive MIMO systems can significantly accelerate the development and optimization of adaptive directional diagram.
Behavioral (functional) modeling for Massive MIMO
Achieving the optimal design of modern wireless systems requires the use of combined models of antenna arrays and algorithms to form an adaptive pattern of direction to model their interaction, as well as determine the degree of impact on the performance of the system as a whole. This may be an impossible task for the currently used tools for the design of 3G- and 4G-solutions, in which, as a rule, the functionality for antenna design is separated from the development of system architecture and algorithms for the formation of an adaptive directional diagram. Also, in most cases, MIMO simulation time lasts ten times longer than 3G and 4G simulations.
Simulation of the lattice antenna at the behavioral level will solve these problems and will be more effective the more 5G simulations are performed. Behavioral-level modeling significantly reduces the simulation time. This allows engineers to experiment with different antenna array architectures and algorithms, simulate the performance of these antenna arrays and associated algorithms, and iteratively adjust parameters to reduce the negative effect of antenna signal crossing and achieve better control over the directional diagram.
Form a directional pattern for hybrid antennas
As smaller wavelengths enable Massive MIMO implementations in small form factors, engineers will face new challenges in signal path detection and overall transmission, as the load of the mmWave frequency will increase as more and more wireless systems and solutions with 5G New Radio support emerge. Ideally, in order to achieve better directional diagram control and greater flexibility for future systems, it is necessary to have independent load control for each element of the array antenna implemented by means of a transmit/receive (T/R) module dedicated to each such element. However, this approach is generally not feasible due to cost, size and power consumption constraints.
Hybrid Antenna Direction Diagram is a method that divides your task of providing better control of the directional pattern into digital and radio frequency components, allowing you to reduce the cost of the final product directly related to the number of radio frequency circuits used. Hybrid Antenna Directional Diagrams allow you to combine some elements of the antenna array into sublattices, defining one transmit/receive module for each sublattice.
For engineers who want to use this strategy in their 5G projects, the key challenge they will face is finding the right balance between providing the required performance parameters and meeting the cost constraints of the final product. Software platforms such as Simulink (for more information on the software platform, see
Figure 2 shows a segment of a multi-domain model containing digital values of the directional diagram, which are used to generate signals of the radio-frequency sublattices, where phase shifts are applied. The resulting hybrid values create the necessary patterns for the lattice antenna and help prepare the systems for the simulation.
Modelling and linearizing power amplifiers for 5G New Radio
The stringent linearity requirements of power amplifiers (PAs) were both important to past standards and will continue to be a critical feature of every future 5G transmitter. Lack of attention to the performance of power amplifiers when developing new products for high line regions will result in solutions that are simply uncompetitive, especially when it comes to higher frequencies and broadband links that are directly associated with 5G. For this reason, digital predistortion (DPD) techniques are usually used to improve transmitter performance and simultaneously limit signal distortion and level out interchannel interference.
Developing a high quality DPD algorithm is challenging because it requires a deep understanding of the effects produced by a power amplifier and related subsystems such as the antenna. Since power amplifiers themselves generate non-linear distortions that depend on the memory effect, the performance of the power amplifiers is highly dependent on the type of signal they are working with.
Because of these complexities, DPD algorithms are often developed in laboratories using rapid prototyping platforms that allow algorithms to be tested based on an existing power amplifier. While this approach is useful for testing and fine-tuning algorithms, it is much more difficult to use when a real existing power amplifier is not yet available, or when the main purpose of the simulation is to study the area of digital prediction algorithm design.
Adaptation to new 5G
The 5G wireless standard will provide significantly higher throughput for mobile broadband through the enhanced Mobile Broadband (eMBB) mode. The key elements of the new 5G standard are:
- Slot duration is shorter, corresponding to an increase in the number of carriers, which allows for increased bandwidth and reduced delay time.
- New advanced encoding schemes such as LDPC (Low-density parity-check code) for data transmission and polar codes for controlling the transmitted information, which are used to correct errors and increase data transmission speed.
- Advanced transmission form with improved out-of-band emissions (OOBE), which allows more efficient use of bandwidth resources.
- Space channel models for current (<6 GHz) and higher frequencies.
While these elements have the potential to increase the efficiency of future systems, they can significantly increase the complexity of new solutions design and can cause significant delays in your development schedule. Therefore, when implementing the above technologies for new 5G products, the use of modeling will help engineers to cope with the above difficulties at earlier stages of the project.
Specialized design tools make it much easier to learn different algorithms and find compromises when designing an architecture early in its development. By simulating different components of a 5G system in the same environment with realistic distribution channels, engineers have the opportunity to test the performance of a complete system before it is sent to the test lab. As a result, the implementation and integration of new innovative 5G wireless systems may not be that difficult (for more details on how to use MATLAB and Simulink to create 5G products, see
Cellular operators around the world are actively looking at fifth generation networks. Commercial operation is not expected before 2018 – 2020. Huawei, together with Japan’s NTT docomo, has already tested 5G in the field and achieved a data rate of 3.6 Gbps at frequencies below 6 GHz. According to official 5G documents, the maximum possible speed is up to 20 Gbit/s.
In Russia two telecom operators have already announced the beginning of negotiations on the creation of a pilot zone for fifth generation networks. MTS and Ericsson will start moving in this direction in 2016 in the 5 GHz frequency band, because no permissions are required for its use if the work is carried out inside the premises. At the same time, Ericsson Lean Carrier technology will be tested, which allows to effectively distribute signal traffic between the cells. There will also be a demonstration of new interfaces for connecting different devices to the Internet (M2M): EC-GSM (Extended Coverage GSM), LTE-M (LTE Machine) and NB-IoT (Narrow band LoT LTE based).
Interestingly, the technical requirements for 5G have not yet been formed – the International Telecommunication Union will define them only in 2019-2020. However, it is already expected that in the end 5G will be allocated bands over 6 GHz, which are not yet occupied by existing technologies. In 2017, the International Telecommunication Union will determine the bands over 6 GHz, which are not yet occupied by existing technologies. MTS and Ericsson have planned to create a 5G test zone in the 15 GHz band. The parties expect the regulator to allocate frequencies in this band. And in 2018 it is planned to organize a 5G test zone in the venues of the World Football Championship.
Since the frequency bands are standard and mostly covered by existing solutions on the market, the test equipment manufacturers have already released specialized software to their signal and spectrum analyzers. For example, the company Rohde & Schwarz has developed a solution that allows you to assess the spectrum of 5G signal. The solution consists of a spectrum and signal analyzer FSW, a vector signal generator and the new software R&S TS-5GCS. Users can select the required oscillator and analyzer models from a range of instruments, taking into account the required frequency range.
MIMO technology will also be used in 5G networks, but if MIMO 2×2 is used in 4G, the number of antennas must increase in 5G technology. This technology has two strong arguments for application at once: the data rate increases almost proportionally to the number of antennas, while the signal quality is improved by receiving a signal from multiple antennas.
When testing fifth generation networks, technicians must be able to measure the pulse characteristics of the MIMO channel (channel probing), which is the main task when measuring the MIMO channel. The obtained measurement data are further used for MIMO channel modeling.
The process of channel probing makes it possible to evaluate the characteristics of a transmission channel or a mobile radio channel when establishing data transmission (
To measure the parameters of the MIMO radio channel, it is necessary to ensure synchronization of the transmitter and receiver of the measuring unit to probe the channel. To ensure synchronization, the following methods are used:
- Cable connection synchronization – this method is used for indoor measurements;
- Synchronization on the received signal is a simple way, but is subject to a large influence of interference and distortion in the channel, has low accuracy;
- Synchronization by GPS signals – GPS receivers allow to use standard PPS or 10 MHz signals as reference signals for synchronization;
- use of highly stable reference generators, e.g. based on rubidium. Rubidium oscillators are synchronized initially before measurements and maintain a high frequency stability for several hours.
The characteristics of the radio channel depends on many parameters, such as: loss, reflection, absorption, multiple radio signal propagation, Doppler effect, etc. The pulse characteristics of the channel are measured by the correlation principle. The vector signal generator is used as a signal source for probing the channel and generates a test pseudo-random signal with good correlation properties, and the signal analyzer and spectrum FSW acts as a receiver. Application TS-5GCS works on the basis of MATLAB and directly receives a signal from the input and output ports FSW, resulting in a comprehensive pulse characteristics of the channel. The result of the correlation of signals is displayed graphically.
evaluate the quality of the communication channels.
TS-5GCS software has a built-in library of different types of probing signals for effective analysis and modeling of data transmission channel in 5G networks.