Wireless Communications

Overview

Wireless communications are now part of everyday life, enabling systems and networks such as 5G, WiFi and the internet of things to name but a few. This module develops the necessary concepts and techniques required to understand the design and operation of present-day wireless communications. It also explores some of the technologies which are likely to underpin future wireless systems such as AI and machine learning. Among the topics that will be covered are:

• Historical development of wireless communications
• Information theory (entropy, probability, and channel capacity)
• Digital communications, channel capacity (noiseless and with noise) and bandwidth
• Signal-to-noise ratio (SNR), bit-error-rate (BER) and Friis Law
• Digital modulation and demodulation
• Radio wave propagation
• Free space propagation
• Reflection, diffraction and scattering
• Large and small-scale fading
• Frequency selective and flat fading channels
• Rayleigh and Rice fading
• Outage probability
• Time series analysis and machine learning concepts for communications
• Time series models (autoregressive, moving average, autoregressive moving average)
• Autocorrelation and partial autocorrelation functions
• Stationarity and unit root
• Data scaling and normalisation techniques
• Multiuser systems (multiple access, multiuser diversity techniques)
• Multiple antennas (e.g., maximal ratio combining, MIMO techniques)
• Multicarrier communications

Coursework:
1. Focuses on time series analysis for wireless communications and introduces machine learning concepts
2. Time series analysis is performed on real wireless communications data set (e.g., device-to-device, mmwave communications data).
3. Autoregressive (AR) and autoregressive moving average (ARMA) models will be used to predict future received signal strength information.
4. Solutions will be obtained using a Python notebook that can be accessed via Google Colaboratory.

Additional Resources and Recommended literature:
Google Colab based Python notebooks containing examples of various time-series concepts and models are discussed in class. Students are also pointed towards time-series forecasting materials readily available on GitHub for more information and additional learning. Other references include:

• W. McKinney, Josef Perktold and Skipper Seabold, Time Series Analysis in Python with statsmodels. SCIPY 2011.
• P.J. Brockwell and R.A. Davis (2002). Introduction to Time Series and Forecasting. Springer.
• P.J. Brockwell and R.A. Davis (1991). Time Series: Theory and methods. Springer. P. Diggle (1990). Time Series. Clarendon Press.

Learning Objectives

• Develop an understanding of information theory and related problems, covering key concepts in wireless communications, such as entropy, probability, and channel capacity.
• Demonstrate a thorough understanding of the principles of noise sources in wireless systems. Learn the application of Friis Law to wireless channels and gain the ability to analyse noise temperature, noise figure, noise factor in cascaded wireless communication systems.
• Understand the principles digital modulation and demodulation, and their application to baseband data for wireless communications. Gain the ability to analyse error probability in signal demodulation.
• Demonstrate a good understanding of the radio wave propagation, relating received power to electric field strength, and the main propagation mechanisms encountered in wireless systems (including free space propagation, reflection, diffraction, and scattering)
• Demonstrate a good understanding of the concepts of effective aperture, gain and directivity in relation to antennas
• Understand the need for statistical approaches to modelling signal propagation and reception in wireless systems (including large and small-scale fading, frequency selective and frequency non-selective fading, etc.)
• Understand the Rayleigh and Rice fading models (including their underlying physical models, signal envelope and received signal power, in the case of Rayleigh, the level crossing rate, average fade duration and outage probability)
• Understand the importance of time series analysis in wireless communications, and the concepts of time-series analysis (autocorrelation function, partial autocorrelation function, stationarity, and unit root). Apply time series models (autoregressive, moving average and autoregressive moving average) to forecast future values based on previously observed values.
• Understand the concepts of multiple antenna diversity techniques including antenna diversity and diversity order. Understand the difference between MISO and SIMO systems. Examine and compare the performances of multi-antenna combining techniques (selection combining, maximal ratio combining, equal gain combining, maximal radio transmission) over Rayleigh fading channel.
• Understand the main concepts and principles behind MIMO systems, and the differences between SIMO, MISO and MIMO systems. Understand spatial multiplexing. Demonstrate the ability to design linear-complexity MIMO receivers (e.g., minimum mean squared error), to model and represent MIMO channel matrices and to evaluate the performance of MIMO systems.
• Understand the OFDM modulation technology, differences between FDM and OFDM, the main concepts behind analogue and digital OFDM. Demonstrate an understanding of the basic properties of OFDM and their benefits (robustness to multipath fading and reduction of inter symbol interference).
• To understand the need for resource management (scheduling), the concepts of multiuser diversity (another way to deal with fading channels) including random access scheduling and greedy access scheduling.
• Apply noise theory and Friis Law to design, analyse and optimize receiver systems in wireless communications.
• Use learning from the Gaussian statistics as well as the Rayleigh and Rice fading models, to work with more advanced (unseen) fading models. Use this knowledge to determine different performance measures related to wireless communications.
• Learn to implement time-series models such as autoregressive, moving average and autoregressive moving average using appropriate software. Learn to check for stationarity and apply scaling as well as normalisation techniques to a given data set using software. Learn to predict future values based on previously observed values using software.

Skills

Assimilation of lecture material, python skills, system model and problem-solving skills as well the application of probability, statistics, electromagnetic theory, and time-series forecasting to wireless data sets.

Assessment

None