This thesis aims to abstract the physical layer (PHY) performance of current and upcoming technologies, so that, their suitability for various use cases and scenarios could be evaluated within an affordable time. For the said purpose, a new effective SINR mapping technique eEESM along with the dynamic optimization of the fitting parameter is proposed. The mapping accuracy of proposed eEESM techniques is analyzed and compared against the other state-of-the-art methods in the doubly selective channel. The results show that the proposed technique is more accurate and map closest to the reference packet error rate (PER) curves. Moreover, the mapping error of eEESM is the lowest for all considered MCSs. The justification for its better performance is the tighter symbol error rate (SER) approximation used to derive effective SINR and the proposed optimization approach. The main purpose of using PLA instead of full PHY simulations is to reduce simulation time. Therefore, a novel concept is presented to abstract PHY performance depending on the time and frequency selectivity of the channel. This further reduces the number of computations required to estimate performance using PLA. To demonstrate the gain in terms of simulation time, the computation complexity of PLA is compared against full PHY simulations. Results show that PLA is roughly 1000 to 1000000 times faster (depending on the abstracted fading conditions) compared to the PHY simulator. The effective SINR mapping approach is then further extended for future candidate multi-carrier techniques (i.e., OFDM, DFT-s-OFDM, GFDM, OTFS), which could be adopted by the upcoming technologies. For this purpose, the received SINR of symbols received through these multi-carrier techniques is derived. The resultant received SINR also considers the impact of ICI due to Doppler. Subsequently, the received SINR of symbols is mapped to effective SINR considering the selectivity of the channel. By comparing the effective SINR, OTFS outperforms other techniques. The reason for the better performance of OTFS is due to the spread of symbol energy over time and frequency, which results in higher effective SINR due to higher diversity. Furthermore, evaluation results show that the proposed PLA can accurately model the performance of these multi-carrier techniques under various fading conditions. Multi-connectivity is another enhancement being considered for future technologies, as an enabler for ultra-reliable communications under harsh channel conditions. Therefore, multi-connectivity communications are also studied in this thesis. Specifically, the frequency domain multi-connectivity networks are presented. To fully exploit frequency diversity under frequency selective channels, the subcarrier-based link combing scheme is proposed. The earlier derived received SINR is then extended for the state-of-the-art link combining schemes, i.e., SC, EGC, and MRC. The multi-connectivity gain in terms of the average received SINR is derived and compared for the above-mentioned combining schemes. To abstract the performance of multi-connectivity communications, the post-combined effective SINR mapping is proposed, where effective SINR represents the combined performance of connected links. The developed PLA performance is validated against the PHY simulations for the case of MRC. Results reveal that with the increase in multi-connectivity order, the RMSE error decreases due to the decrease in the variance of mapping SINRs. In the end, various applications of PLA are demonstrated. The developed multi-carrier PLAs are used to compare the performance of multi-carrier techniques under various fading conditions. Results depict that PER of multi-carrier techniques generally decreases with the increase in time or frequency selectivity, given that, the ideal channel estimation, ICI, and inter-symbol interference (ISI) cancellation is used. The multi-connectivity evaluation results depict that with the increase in channel selectivity higher diversity gain could be achieved. Besides, the proposed subcarrier-wise combining scheme achieves better performance compared to the traditional link combining approach. The next PLA application demonstrated is the performance comparison of V2X technologies, i.e., IEEE~802.11p, LTE-V2V, IEEE~802.11bd, and NR-V2X, in an Urban NLOS communications scenario. It is observed that 802.11bd outperforms other technologies in terms of PER and packet reception ratio (PRR). Its better performance is due to lower ICI compared to LTE-V2X and NR-V2X, and due to the use of LDPC codes compared to 802.11p. In contrast, NR-V2X outperforms other technologies in terms of data rates and packet inter-arrival time. The last PLA application shown is the link adaptation for single-link and multi-connectivity communications. In single-link communication, the performance of various PLA techniques is compared in terms of achieved data rates and outage probability against the case of perfect CQI. The CQI based on the proposed eEESM technique improves the data rates and reliability of the link, compared to other schemes. Further, in the case of multi-connectivity, the post-combined effective SINR mapping proposed in this thesis is used for link adaptation in terms of both MCS selection and adapting the number of links. The proposed scheme optimizes multi-connectivity data rates while using the lowest possible number of links required for the desired quality of service.

In 5G systems, two radio air interfaces, evolved LTE and New Radio (NR), will coexist. By using millimeter waves, NR will provide high throughputs, but the higher frequencies will also lead to increased losses and a worse coverage. Multi-connectivity is therefore envisioned as a way to tackle these effects by connecting to multiple base stations simultaneously, allowing users to benefit from both air interfaces’ advantages. In this thesis, we investigate how multi-connectivity can be used efficiently in ultra-dense networks, a new paradigm in which the number of access nodes exceeds the number of users within the network. A framework for secondary cell association is presented and an energy efficiency’s condition is proposed. Upper and lower bounds of the network’s energy efficiency are analytically expressed. Algorithms for secondary cell selection are designed and evaluated through simulations. Multi-connectivity showed an improvement of up to 50% in reliability and and an increase of up to 20% in energy efficiency.

The realization of wireless ultra-reliable low-latency communications (URLLC) is one of the key challenges of the fifth generation (5G) of mobile communications systems and beyond. Ensuring ultra-high reliability together with a latency in the (sub-)millisecond range is expected to enable self-driving cars, wireless factory automation, and the Tactile Internet. In wireless communications, reliability is usually only considered as percentage of successful packet delivery, aiming for 1 − 10⁻⁵ up to 1 − 10⁻⁹ in URLLC.

The Internet of today is very different from how it used to be. Modern networked applications are becoming increasingly diverse. Consequently, a variety of requirements must be met by the network. Efforts to make the underlying mechanisms of the Internet more flexible have therefore been made to adapt to this diversification. In this thesis, we explore how information about application requirements can be leveraged to optimize the network protocol stack of end-hosts during run-time. In addition, we improve the visibility of the network to the end-host in order to enable additional flexibility in the usage of the network's resources. We conduct tests in real-world testbeds and examine how services might be developed to optimize latency, throughput, and availability for various network traffic scenarios, including 360-degree video streaming, drone autopilots, and connected vehicles. We show how multi-connectivity, where the end-host is connected via multiple network paths simultaneously, may be used to significantly reduce latency and increase availability, while minimizing the overhead imposed on the network by carefully considering the network selection process. Furthermore, we describe an architecture that allows the user equipment and network functionality inside the 5G core network to cooperatively optimize the resource usage of the network. / The Internet of today is very different from how it used to be. Modern networked applications are becoming increasingly diverse. Consequently, a variety of requirements must be met by the network. This presents a massive challenge, since the Internet was originally designed on best-effort principle.  To address this challenge, we explore how Internet end-hosts can flexibly adapt to the needs of individual applications, by dynamically configuring the network protocol stack during run-time. In addition, we improve the visibility of the network, allowing end-hosts to better utilize the resources of the network.  We conduct tests in real-world testbeds and examine how services might be developed to optimize latency, throughput, and availability for various network traffic scenarios. We also show how multiple network paths can be used simultaneously to significantly reduce latency and increase availability, while minimizing the overhead imposed on the network. Furthermore, we describe an architecture that allows the user equipment and network functionality inside the 5G core network to cooperatively optimize the resource usage of the network. / <p>Paper II was published as a manuscript in the thesis. It is an extended version of the paper, which adds additional material that had to be cut from the original paper due to page limit restrictions.</p>