Physical Layer Signal Processing

Physical Layer Signal Processing

Multiple-Input-Multiple-Output (MIMO) technology is considered as a powerful approach to improve the system's performanceof wireless communication networks, in terms of data transmission reliability and throughput. MIMO systems consistof the implementation of multiple antennas at the transmitter and/or at the receiver. By increasing the number of antennas,features such as diversity and spatial multiplexing are observable when applying proper signal processing, e.g.,design of transmit/receive beamforming and receiver decoding among others.

General Rank Transmit Beamforming

Due to the rapid increase of the number of users in the modern wireless communication network, the users' demands for high data-rate also increase.Multicasting networks can meet these increasing demands and provide wide application spectrum such as data broadcasting and video conferencing.However, the presence of multi-users which share the same spectrum causes multiuser and co-channel interferences.

Downlink Beamforming Scenario
Downlink Beamforming Scenario

Assuming a perfect channel state information at the base station, transmit beamforming techniques can be employed at the transmitter side and multiple users can be served simultaneously.A multi-antenna base station broadcasts the information to a single group or groups of users based on their service subscription.In order to steer the transmitted energy in the desired direction, beamforming techniques are employed at the base station.

In the Communication System Group, General rank transmit beamforming techniques are investigated, where space-time block codes (STBCs) are implemented in the transmit beamforming. The main research direction is to implement STBCs based beamforming schemes which improve the performance compared with the stat-of-the-art techniques, can be Applied in multi-group multicasting scenarios, and enjoy low computation complexity.

Associated Researcher:

Massive MIMO

With the ever increasing user demand of higher data volumes, the need to enhance the network's capacity has motivated the application of MIMO technology with growing number of antennas. In that matter, the current 4G standard supports deployments of up to 8x8 MIMO. In 5G, large-scale and Massive MIMO consider the case where the number of antennas increases to very large numbers (even infinity). Several studies in the literature have analyzed from a theoretical point of view the benefits of Massive MIMO, under the scenario of simultaneous utilization of the same frequency (or time) resources for data transmission by many users (MU-MIMO), showing significant gains in network's capacity. A main challenge associated with the massive MIMO technology is to design reliable and fast symbol detectors. Even though extensive research work has been done in the last decade, and numerous techniques have been proposed, many of them are not well-suited for massive MIMO systems for two reasons. Firstly, the most of the optimal or near-optimal detectors become infeasible or extremely slow for large systems. Secondly, suboptimal detectors such as the ZF detector, MMSE detector, are highly sensitive to noise, and offer significantly higher bit error rates (BERs) compared with an optimal detector. Moreover, the performance of these detectors degrade further as the number of transmitted data streams approaches the number of receive antennas.

Massive MIMO and Equivalent SISO Model
Massive MIMO and Equivalent SISO Model

Over the last three decades, there has been tremendous advancement in semiconductor technology, with a major focus on parallel processing. With the development of massively parallel processor array (MPPA) technology, it is not a distant future that we have economically viable embedded systems with hundreds or thousands of processing units on a single millimeter-size die.Such parallel hardware architectures can benefit the massive MIMO systems to process a huge amount of data in real-time.By designing detection algorithms which facilitate the parallel processing, the available hardware can be efficiently utilized to speed-up the detection process.

In the communication group, we aim to develop massive MIMO detectors whose complexity scales well with the number of transmit/receive antennas and the constellation size, along with inherent support for parallelism. We consider, from a more practical implementation perspective, different aspects of large-scale and Massive MIMO technologies. Advanced receivers are designed by utilizing convex optimization techniques and compressive sensing in order to be able to decode the transmitted information with low computational complexity and high speed. Additionally, system level simulations are applied where the principles of Massive MIMO are used such as favorable propagation and channel hardening, in order to reduce simulation complexity and provide a deeper insight into achievable bounds in practical scenarios.

Associated Researchers:

Projects and Collaborations:

Physical Layer Security

Physical layer security has recently drawn considerable attention as a complementary strategy to provide secure data communications. In particular, physical layer security is viewed as a promising solution to provide wireless security in 5G, since it does not depend on computational complexity, and has a high scalability to allow the coexistence of communication terminals with different levels of hierarchical architectures. What's more, physical layer security can either provide direct secure data communication or assist the distribution of cryptographic keys, which makes it particularly favorable in 5G networks.

Eavesdropping Scenario
Eavesdropping Scenario

Figure 3 illustrates a communication system, where Alice aims to communicate with Bob, and Eve is an eavesdropper. At the physical layer, there exists numerous approaches to shield messages from Eve and enhance the security of the wireless communication between Alice and Bob. One popular scheme is to introduce controlled artificial noise (AN) to efficiently jam the signal reception at Eve. By applying an additional processing unit to generate AN, this approach can be easily implemented in a practical system to secure wireless communications. In most existing literatures, AN is generated from Gaussian distribution, which increases the effective received noise at Eve. However, such a generation may not be optimal in terms of energy consumption. This technique has recently been studied from an information-theoretic perspective, such as secrecy rate and secrecy outage probability. In order to take discrete modulation alphabets and finite block lengths into consideration, other security performance metrics have also been proposed, such as bit error rate.The observation that in modern wireless communication systems square quadrature amplitude modulation (QAM) is widely used motivates us to address the problem of how to optimally apply AN to enhance the physical layer security. The symbol error rate (SER) of the demodulated signal at the eavesdropper can then be used as a performance metric.

To secure the signals from the Eve, Bob can design and transmit AN symbols to jam the signal reception at Eve. In the communication group, we designed an optimal AN generation scheme. The objective of this scheme is to degrade the error probability performance at Eve, under an average power constraint. This AN design problem was mathematically formulated and solved using existing optimization tools. Moreover, the optimal AN distributions were explicitly determined based on various channel knowledge that is available at Bob. One important open question for physical layer security is that the eavesdropper Eve is likely to be passive. In this case, the channel information of the eavesdropper is generally not known. The design of physical layer security schemes become very challenging, and worth attention and efforts.

Associated Researcher: